Chapter 29 - Flight Testing part 6

A Strong Smell of Gas in Palm Beach

Palm Beach from 7000 Approaching North County F45 This is one of those dufus paragraphs, so keep alert for the clues....
It was raining, but I did manage a couple of trips to the gas station and the hangar to tote a total of 50 gallons of 93 octane gas to the plane. Before adding any gas I disconnected the solenoid and ran the left pump until the last 6 gallons or so of 87 ocatane were pumped over to the right tank. As soon as the pump sucked air with the longerons level I stopped the pump. This is the first time I've filled the tank to the brim from empty. 5 gallons at a time and checking my manual sight gauge calibration lines as I went, I added exactly 30 gallons to the left tank. I'd fitted the sight gauges exactly per the drawings, but it shows full at 20 gallons, and Vance's orange ball is still sitting at the bottom. The last 10 gallons is above the top of the sight gauge. I'm getting quite good at mixing the oil in and pouring the gas into the tank with a wet rag wipe first to cut down static. I only spilled a few drops which I wiped up with my rag. The 30 gallons plus 30 oz of 2 stroke oil brings the tank to the very brim. Excellent. When building the strakes I moved the inside bulkhead in about 2 inches and the outside bulkhead out about the same. I knew I had more than the stock 26, and had hoped for the 30 I actually got. That's a total of 8 gallons more than stock for an extra hour or more of range.

Next I topped off the right tank for a total of 43.6 gallons added. The EM2 said I'd used 45.1, so it's reading a bit high. I tweaked the calibration down a bit. It was still raining, so I put the other 8 gallons in the trunk of my car (Paul doesn't allow gas cans in the hangar) and left the plane ready for flight the next day.

After three days of rain and two of work I finally got the chance to head down to the airport on Friday afternoon. My car stank of gas. I checked in the trunk, and found one of my gas cans on it's side. It had leaked a couple of gallons into the spare tire area and onto the ground. Yuk! I opened the windows wide and drove down to the airport with my head half out of the window. Once there I left the trunk open to help the smell dissipate.

When I opened the hangar door there was a small pool of gasoline under the right vent of the plane. I'd filled the tanks to the brim for calibration, and it must have expanded with the heat of the day. I removed the caps and a cup or so of gas flowed out of each tank and onto the top of the strakes. Anxious to get airborne I wiped up all the gasoline, moved some tools off the pilot seat, pushed the plane back, and went through the pre-flight.

Everything was in order as I taxied to the runup area. I tested both pumps, then left it on left. I like to use up the fuel in the left tank first to help with balance. Also, I don't trust the fuel gauges yet because the calibration needs a lot more work, so I use the left tank first because I can't see the sight gauge on that side without removing my seat belt. After a good runup I took the active and set out for a few hours of fun flying.

Everything looked good in the pattern, so I called for clearance to climb. Palm Beach Approach were busy and wouldn't let me in, so I headed South 5 miles to Boca Raton, then climbed with my wingtip touching the class C line as shown on the GPS. I climbed around the class C, then headed North as soon as I got over 4000. My plan for the day was to treat the "Slick Kitten" to her very first taste of foreign soil. Well, not really foreign, but at least different. You see she's never touched ground anywhere except on my patio and at Lantana airport. I thought it was about time she stretched her legs, or should I say spread her gear, or... Lets just say "landed" ....somewhere new.

Rivera Beach from 7000 On the ramp at North County - F45 At 7000 feet I took a few pictures, then the GPS said I was 5 miles from F45. I took time in the descent to figure out the pattern. They have a 26L and a 26R, and the 26R was busy with ILS approaches. I noticed that Cessna's taking off were being blown north of the centerline, so there's obviously a fairly strong crosswind. My GUMPS check on downwind revealed 1/2 full on the left side. Yea - right! It was full when I left LNA 20 minutes ago. That would be 30 gals/hour. I don't think so. Gotta get those capatitance senders calibrated. I'd tried as I filled the tank, but they're really sensitive to the slightest adjustment.

Keeping my prop strike experience in mind, and knowing I was relatively heavy with gas I kept the speed to about 95kts on final. The crosswind didn't seem to be much of an issue, but I did feel her sag a little at 10 feet, and nudged a little more power on to squeak down "per plans". I taxied over to the ramp and parked in line with the other planes. A fuel truck arrived across my nose before I even got the canopy open. I signalled "No thanks, I only pay $2.12 for my gas" and climbed out.

There was gas coming out of the right vent and forming a pool on the ramp. Knowing how they're a bit touchy about this sort of thing at a lot of fields I went to release the cap, but a half gallon wanted to come out, so I put the cap back on and pushed the plane forward away from the pint or so already on the tarmac. Who me? Nah. Must have been the last guy in that Mooney.

I seem to be leaving a trail of gas everywhere I go today, I thought. Little did I know. Anyway, back to the story...I'd only used the left tank on the flight up, so the right tank was still brim full. Must be the sun on the strakes. Once the tide slowed to a stop I walked over to the FBO for a coffee. "Pretty little plane" said a lady as I passed by. I thanked her and went in for my coffee. When I came back out she, her husband, her mother, the gas truck guy and four other people were crowded around the plane peeking inside. I gave them the full tour, answered questions for 15 minutes, and finally had to almost shoo them all away, explaining that I needed to head on back before dark.

Flying out of a field with 4500' runways, no class C airspace overhead and lots of places to put down in an emergency is much nicer. I stayed over the field until I got to about 8000', then headed back to Palm Beach with PBI and LNA in clear sight. This time I used the right tank to get the level down and avoid further expansion. Over Palm Beach I tried to call approach for clearance into Lantana, but they were busy with an emergency of some sort. It sounded like a plane had gone down somewhere to the West and they were coordinating the helicopters. Eventually there was a break on the frequency and I got a quick N96 Papa Mike, Palm Beach" in sideways. The controller came back with a squark and I began a descent toward Lantana. A few seconds later he came back with the dreaded words "6PM, please call the tower on a land line after landing. Say when ready to copy the number". Oh sh.t. What have I done now? I swear I was exactly on the edge of the class C all the way. I was close, but I never breached it. I was making up my affidavit as I slowly circled down to land listening to all the helicopters trying to avoid each other. Something was definitely going on West of the field, and I heard something else said about looking for a downed aircraft. Poor bas.ard, I thought. I hope he's ok.

As I rolled out someone on the frequency said they thought my ELT was broadcasting. Geese. What next? I said I'd check it as soon as I could, and taxied back to the hangar.

Waiting at the hangar were Paul, another local pilot & friend, Bob, and the local Sherrif. Uh oh. I must have I've p.ssed someone off good, I thought. I shut down and climbed out. Paul looked very relieved. The Sherrif seemed very happy that I was ok, and departed. Bit by bit the story came out. Bob's house is under the pattern. He'd heard my engine and gone out on his lawn to watch. As I went over his house he saw what looked like a cloud of white smoke or vapor coming from my right side. Very concerned, he'd called Paul and they both rushed over to the field. For some reason the FBO wouldn't let them use the radio, so, thinking I'd be talking to Palm Beach, they'd called Approach on the land line and told them the story. At around this time Paul tuned his handheld into the emergency freq, got a strong ELT signal and immediately figured I'd gone down. I'm not sure who called out search & rescue guys. I called the tower supervisor at Palm Beach. He asked if my ELT was broadcasting. While he was on the phone I removed the front cover and checked. Nope. I disconnected the antenna and switched it from armed to off. Pauls radio carried on warbling. Nope. Its not mine, I told the nice FAA man. He thanked me, wished me a good day and hung up. Paul and Bob were very apologetic for calling out the troops on me, but I thanked them. Better that than the other way around. As I pushed the plane in the hanger I noticed that the left tank was down to 10 gallons. Strange. I did 1.6 hours, half of which was on the right tank. Twenty gallons fuel burn in 0.8 hours flying seems like a lot.

On the way home it dawned on me. The crimpers on the pilot seat. I'd left them there to remind me to hook up the disconnected solenoid wire. The reminder didn't work, and the crossfeed warning light only works if the right pump is energized. I'd flown about 60 miles with the left tank return going to the full right tank. The "cloud" that Bob saw was vented fuel, and that's why I was dribbling gas from the vent on landing at F45. I'm guessing that I vented about 10 gallons of 93 octane over Palm Beach, hence the title of this paragraph. I suppose it could have been worse for the megabuck people of Palm Beach. At least it wasn't the cheap stuff. Ah well. Another 1.6 hours on the hobbs, another adventure in the log and another lesson learned. ALWAYS leave a note on the panel if leaving the aircraft unservicable. I'm so used to leaving it unserviceable I haven't got into that habit yet.

Did I mention that the engine performed flawlessly for the entire trip and that I had a really good flight, met some nice people and actually visited another airport? 21.4 on the hobbs, and I hope the guy with the ELT is ok.

I'll be flying again tomorrow, if they'll let me in the gate.

The Sweet Spot

Startup outside the hangar Fly-by Click on the pictures for full size images

Today's flight inspired a new line for the limerick.

Seventy-two, skies of blue
Much better luck with the EC2
Slick Kitten is faster than I ever knew!

I reconnected the solenoid, tested the flow, added what was left of the gas from my truck - no I didnt squeeze out the carpet - and went flying. Paul parked his golf cart next to the runway and snapped the picture at left as I did a low pass. I cleared out below the class C and decided to drop in on Buly, another turbo rotary Cozy builder, at FXE (Ft. Lauderdale Executive). I need to get back into the swing of entering and leaving busy controlled fields in Class C, talking to clearance delivery etc. etc. Teasing Buly would just be a bonus.

Outside Buly's hangar at FXE - No Buly 10 minutes later. Arrival at FXE, 15 minutes after leaving LNA, was a breeze. Temps were 135/140 as I taxied up to Bulys hangar. He wasn't there. I checked my cell phone and there was a missed call - from Buly. Hmmm. A chance for some fun. I called him back, chatted for a while and asked if he was working on the plane today. He said he was on his way to the hangar. Perfect. Then he spoiled it and asked where I was. Outside you're hangar. Buly showed up 5 minutes later and a bunch of canardians came out of the woodwork for a peek. There were VariEZ drivers, LongEZ drivers, a Dart (whatever that is) driver, a Velocity guy, some builders and Buly, leaning against the cowl. He probably got oil on his shirt from that. Buly gave me the names later: From left to right: Buly - Cozy IV, Glen - VariEZ, Oscar - Velocity, Me - Cozy IV, Richard - LongEZ, Jim - Dart, No name - VariEZ.

After a half hour visit and a quick snack on Buly's French Fries I taxied out for departure. It's a LONG way to the hold point for 26 and my temps, which started at a heat soaked 170/160 were 205/185 by the time I was lined up for take-off. The canardian crowd was lined up by the runway as I took off. Later Buly said it sounded "Sweet. Like a real airplane." The tower approved a pattern, but then talked non-stop until I was turning final, by which time the temps were back down to 170 or so. The controller wasn't happy that I'd done the pattern in half the time he was expecting, thus screwing up all his arrivals, so my planned fly-by / low pass got dumped and I was curtly instructed to "fly South and stay clear of everyone's airspace" I guess the fact that I wanted to go North wasn't relevant. Ah well. I circled the field to the South, then headed West and eventually North under the 1200' class C for Ft. Lauderdale International, then climbed back East to get over Boca. It's nice toodling along at 1000' without an ever present fear that the engine is going to die any second. That hasn't actually happened, but the concern has always been there. Now that concern is moving into the background.

Back over Boca, and then Lantana at 6000' I decided it was time to visit Pahokee. At 200 mph for 40 hours you have to go somewhere! I climbed up to 11,500 to give me glide distance coverage and looked for my pen to write down some readings. I couldnt find it, so sorry - not many numbers for you today - but the ones I DO have are doozies. I'm going to buy a handheld voice recorder for this. Writing is a pain anyway.

Once at the top of the climb I let her settle down a bit, then gradually moved the throttle forward. So far my max rpm has been 5700. As I passed through 5800 with 36 MAP I noticed something interesting. The relatively minor vibrations I'd gotten used to just smoothed out. At 5900 rpm and 38 MAP it was even smoother. Cool. I could get used to this. By this time I had about 165kts indicated and 6000 rpm the engine was simply purrrring like the kitten she's supposed to be. I checked intake temp - 134. I richened the mixture a little and nudged the throttle a bit more. She kept on coming. I backed off - ok chickened out - at 6100 rpm and 185 kts indicated with a 44 MAP. GS showed 234kts (269.2 mph to save you working it out), and no, I didn't measure it going the other way.

Throttling back to 5000 rpm and 140 IAS brought back the minor vibration I'd gotten used to, but I wanted to do some hours today, not just burn off the gas. As I headed over to Pahokee I noted 11.4 GPH on the now calibrated EM2 with 8 bars of mixture. The guys on fly-rotary were talking only yesterday about running on 2 or 3 bars, or even less, but they're not turbo. I tweaked the mixture a tiny bit toward lean, and was rewarded with 10.5. Another tiny tweak - man this is sensitive - and I had 9.4 gph with no noticeable change in engine rpm. EGT at this point was 1640. I settled for that and cruised the 38 miles to overhead PHK in about 10 minutes.

I did a wide circle over Lake Okeechobee, considered taking a trip North to F45 again, but decided on heading back toward LNA. I gave her another bit of a high speed run on the way back with similar results. There's definitely a noticeable reduction in vibration as I pass through 5900 rpm. I think some dynamic balancing is in order, but first I'll send the prop back to Clark for "final finishing" and see how it is when he's done with it. Since I'm getting 6100 I may not want to cut it back much, if at all. I'd rather live with reduced take-off and climb (where higher power may get me into cooling problems too) in exchange for high speed cruise. But, then again, if I cut it back a bit I might be able to do the same with less boost. This would be better - right? By the way, perhaps I should mention that the manual wastegate was wide open for the entire flight.

The flight had been picture perfect so far, and I was planning another couple of hours of experimentation, possibly a trip up to Stuart, when Char blurted out - "John. The oil's getting hot". She doesn't know this, but that really means that the low oil level sensor has activated. I checked the oil pressure - it was sitting as usual at 72 - and throttled back. The problem with a low oil warning is that you don't know HOW low, and you don't know how long it's going to be from the first warning (at 1 QT low) and the final warning, which is an automatic feature built into the engine - it disintegrates and quits. Char kept on telling me about the oil, so I hit the STFU button, figuring she'd wake up in time to remind me about the gear. A call to Palm Beach advising "engine warning light" got me a quick clearance and I descended into the aluminum infested pattern picking my spot carefully. As I came down I'd asked for aircraft in the pattern. I had one guy on final as I hit the crosswind at 1200'. Then I spotted a guy turning a 4 mile base. Why do they DO that? I backed off on the speed, realized he was going to be in the way and asked if Cessna on base would allow the Cozy to do a precautionary landing ahead of him. No reply. I blipped the throttle. Still smooth and good oil pressure. Torn between keeping the engine at idle and maintaining height I snuck in behind him as best I could. He JUST cleared the runway as I touched down.

As it happens, I was a Qt low, so in the past 6 or so hours I've blown a qt of oil out from somewhere. Next time the cowl comes off I'll do another search, but for now I'll top her off and keep wiping the cowl. Today's flight was an even 2 hours. It would have been longer, but for the oil warning.

Sunset Pattern Work

Sunday I topped off the tanks, topped off the oil, and adjusted the altimeter. I'd been rereading the EM2 manual and realized that I'd adjusted the fuel flow calibration the wrong way. I reset it to 161. By the time I was done with all this it was hot and I was hungry. I went home for some lunch and came back in the early evening. The airport was quiet as I pushed her out. I started up and taxied to the active. After the runup the temps were 170/160. There was a 172 on base, so I waited. A Cherokee appeared behind him, so a waited. By the time the Cherokee was on final there were two Cessna's on the downwind and one on base. I waited. I began to realize what was happening. All the Sunday fliers were returning to the nest after a day out. A constant stream of aluminum came in for the next 15 minutes without a gap big enough for me to sneak in. It was entertaining watching them all jiggle for position with some coming in on a right base, some on left. Some joining on the crosswind, some over the field, some straight in. Sunday flyers! I think I'll avoid Sunday evenings from now on.

The wait turned out to be a useful test. With an OAT of 65 my engine temps very gradually climbed to 205/185 over about 10 minutes, then stabalized. I just sat there idling. The engine seemed happy and temps never went above 205/185. Finally I saw a big enough gap and took it. It was getting close to sunset so I put my strobes on, stayed in the pattern, and became part of the juggling. Temps dropped to 170/160 in the downwind. Mostly I stayed at pattern altitude and just watched. At one point, as I turned final but stayed at 1000, I had two below me also on final and one on a tight base that would have been a conflict if he hadn't done a dog leg. By the time the sun started to disappear below the horizon I had the pattern to myself. I put the gear down and descended for a squeaky landing and held the nose off, but not too high, as I rolled to the end of the runway. Since the prop strike my landings have tended to be faster and flatter. Now I'm getting them back to the middle of the envelope. A pleasant evening flight, and another 0.6 on the hobbs.

I've now done a full 24 hours flying in the plane. 16 more and the 40 hours are done. I have a lot of tests still to perform, but the biggest job seems to be largely behind me. One of the most important parts of flight testing, at least for me, has been getting to know the plane, in particular the engine. I'm learning how it "normally" behaves in terms of engine data and I'm learning my way around the fairly complex panel. The more familiar I become with all this, the more I can relax and start thinking about the standard flight test procedures. I have the flight test checklist on the passenger seat. Next week I'll start working my way through it, beginning with ASI calibration.

Life is but a Circle

On the way to the airport I picked up a handheld voice recorder. Once airborne I got to 5000' over the field and did airspeed calibration circles recording the readings as I went. I did three different speeds at 5000, noted the engine parameters, OAT and altitude readings, then climbed to 10,000 and did it again. Total time on the flight was 1.2. When I got back I couldn't hear my own words on the tape. I'd been holding the microphone too far from my mouth. Thankfully, Char is a medical transcriber. She could understand the tape as I played it at my desk, even while she was across the room working on a different tape with headsets on. I'd hate to have hearing that good.

There was one strange experience during today's flight that bears reporting. Everything seemed normal. I was boosting about 38 MAP at 4500 rpm and 5000'. I was gently adding a little throttle to increase speed for another ASI calibration when I felt a surge of acceleration way beyond what I'd expect for the minor throttle adjustment I'd made. This was literally a shove in the back. It lasted maybe 2 seconds, then the engine carried on at the new power level. RPM jumped up a couple of hundred. I glanced down at the MAP. It had dropped to 34. My best guess is that I'd been running with a gummed up plug which suddenly cleared and gave me added power.

A Slight Delay

The next day I went to the hangar armed with 20 gallons of hi-test. All was well until I got to the runup. 3800 rpm was all I could get. Since I was already at the hold point I took the runway and opened her up. She would have flown, but there's definately not as much power as there should be. I took her around for another try. Same reaction. Recalling the slight surge yesterday I'm think the plugs may need cleaning.

Back in the hangar I noticed a bit more oil on the cowl, and this was just from a taxi test. I took the cowl off and looked around. There was a slight oil film on almost everything on the right side of the engine and, strangely, a small pool of oil on top of the turbo connector where the oil feed goes in. I can't image how oil could have accumulated there except by coming out of the turbo oil fitting. The engine was too hot to work on comfortably, so tomorrow I'll be doing some maintenance. Clean plugs, replacement turbo oil fittings, move the exposed oil temp sensor....I'm sure there are more things I haven't thought of. I also noticed that the cover for my plenum AC outlet had blown off. It was clamped in place and safety wired. Now it's just hanging on the safety wire, so that needs attention. Finally I'd like to install some additional duct work inside the cowl to stop the air flowing into the wing roots. At least the oil leaks have been useful in this area. After a flight I can see streaks of discoloration where the oily air has been escaping the cowl in places I hadn't planned on. I don't want to loose the momentum on flight testing, but these basic items need attention before she flies again.

Wednesday 3/16 I found a replacement fitting for the turbo oil feed and took it to Charlie to have a steel AN connector welded onto it. He'll have it done in the morning. Back at the hangar I removed the aluminum oil feed junction box I'd made and plumbed the redrive direct. Now I needed a new place for the oil temp sender and the hobbs sensor. I don't know why I didn't do this before. I removed the oil filter by-pass block, drilled & tapped it for 1/8 npt and installed the temp sensor there. Now what to do with the hobbs? I like the hobbs meter. Its like the mileometer in a car. Simple and trustworthy. There's a hobbs built into my EM2, but it gets reset every time I send the unit back for repair (twice already). There's just something nice about a physical clock.

That leads me to a side comment I've been meaning to mention. The EM2 is a handy device. I wouldn't be without it, and I'm sure it'll become more and more useful as I settle down and get more time to study the data it's giving me. Having said that, the most important inputs I get right now are from the cheap analog engine gauges. Perhaps it's a personal or a right brain / left brain thing, but I've found the quick glance analog input, and the trends that it gives you, to be absolutely essential for the early stages of flight testing. My recommendation, for anyone with an alternative engine installation and digital engine monitor, is to also install a basic set of analog gauges in a temporary panel where your nice Garmin 540 or your Blue Mountain EFIS will eventually go. After the flight testing is done and the engine is settled, the analog guages can come out. At least that's my experience, for what it's worth.

I decided that the best way to fit the hobbs was to swap a 3 way 1/8 NPT connector currently feeding the oil pressure sensors with a 4 way, and add the hobbs to the extra outlet. Sounds simple, but this connector is buried deep under the intake. I didn't want to take the intake off. It might have been quicker to remove and replace the intake, but I eventually got it all done and tightened up. Next I removed the plugs. They were mostly brown. That's a first! They've always been black and sooty. I must be getting closer to optimal mixture settings. I cleaned them and put them back. Finally I did a quick recalibration of the oil temp senders to make sure the readings I'd been getting on one were false. They checked out within 5 degrees of each other and the mechanical gauge. I looked at the cowl duct issue, and decided to put it off for now.

Thursday was heavy rain and 600' ceilings, so at least I didnt loose a good flying day. I picked up the turbo feed. Much better and sturdier than the old one. I reinstalled the turbo oil pipes, heat shrouded the steel feed tube and cleaned up the oil film. I'll run the engine a little before putting the cowl back on then, when the weather clears, its time to get on with the testing.

Back to flying

Friday 3/18 was nice so I reinstalled the cowl and went flying. On the runup I got 3700 max. Hmmm. I wonder if using the B computer has anything to do with it. The B computer doesnt modify mixture depending on intake temp. While holding 3700 rpm I tweaked the mixture a little to the right and instantly got 4100. OK. This is better. Tracy Crook has just annouced a new auto-tune feature for the EC2/EM2. He says it reprogrammed his computer in 30 minutes much better than he's ever been able to do in 2 years of tweaking. I'm looking forward to getting one of those new chips.

After checking both computers, both pumps, both batteries, both sets of injectors both sets of coils AND the landing brake, I took the runway and rolled. Take-off and climb were good. I settled down to do some patterns. After a few minutes I noticed that the EM2 coolant temp was showing 275 and flashing at me. The analog coolant temp read 170. The EM2 has to be wrong, but I put her back down to check.

In the hangar I removed the cowl and found that the rerouted turbo oil feed braided steel hose was touching the coolant temp sensor connector and shorting it out. Problem solved. I tied the cables and hoses a little better. At this point Mark, an A&P from the hangar next door wandered over for a chat. He mentioned that he'd heard what he described as a pop pop sound over the normal engine noise during my last take off. In fact he's heard it a few times when I've been flying. I can't hear anything particularly unusual from the cockpit. On the next take-off I listened intently and could just detect a slightly different sound at the edge of my boost limit. I wouldn't call it a popping. More of a rumble. I back off whenever I hear it. Perhaps it's the onset of detonation, but whatever it is the engine seems to have tolerated it. Based on what Mark described I think I'll cut my boost back a bit. I've been working with a limit of 46 MAP, but I may have exceeded that by 2 or 3 PSI on the last take-off. Oops.

The second flight was almost perfect. Temps stayed below 160 as I buzzed around the pattern at 145kts. By this time the sun was getting low, so I just circled the field for a while to test everything out. All the readings were in the green after a half hour or so, so I decided to put her down and do a longer flight tomorrow. I had to do a 360 on base for spacing, then on short final the sun was right in my face and I misjudged the landing, bouncing to about 3 feet before settling down again. This was probably the worst (as in least comfortable) landing I've managed so far, including the one where I got a prop strike. It counts for excellent, but I don't think you could call it pretty. 26.1 on the hobbs and nothing negative to report, other than a slight smear of oil on the cowl. It could be left over from earlier, but somehow I suspect not. For now I'll wipe it up and keep a real close eye on the oil level.

More flying

KNLA from 6500 feet I'm having trouble picking titles for these paragraphs. Nothing special happened. Saturday 3/19 I did two flights. The first, in the morning, was 0.6. I landed to check the oil and reset computer A which I'd omitted to do on pre-flight. After a drink of iced-tea I taxid back to the active for the planned long flight. After a good runup I took the active and opened her up to 46 MAP. As we rolled I checked the gauges, and noticed the the mixture bars were down to about 3 instead of the normal 6 or 7. The engine was still running well, but as I watched, the reading dropped to 2, then 1, then zero. Fuel pressure was also a little low. I cut the throttle and taxied back. As I taxied I experimented with the mixture setting and was able to bring it up to half way with the manual mixture. When I turned on the other pump the mixture went off the scale. Hmmm. Back in the hangar I checked the fuel pressure on both pumps. 35 on right. 29 on left. 38 on both. I climbed in the back to make sure the manual fuel shut-off hadnt worked it's way partially closed on the left side. It hadnt. My theory for now is that the left pump runs off the left battery, and I havent always been turning on the contactor, so it hasnt been getting a charge. I need to put a diode in the line so it can get charged without being in the circuit proper. The voltage on the left battery reads 12.6, the same as the right, and I recently did a load test on both batteries with good results. Maybe the pump is a bit weak. I doubt it's the filter - I cleaned it recently. I think I'll order a new pump. By this time the engine had heat soaked nicely, so I decided to let it cool off while I got some lunch.

That evening I got back in time for another 0.8 hours of local flying. I used the right pump on take-off, then switched to left once I had some altitude. Everything seemed fine. It was a cool (62F) evening and the air was calm. At one point cruising at 6000' indicating 140kts I checked the fuel burn. 4.5 GPH. As I increased power it came up to 10.5 gph, but I didn't let it settle long enough to get a good IAS reading. The sun was beginning to dip below the horizon, so it was time to come back to roost. I've figured out how to get into the Class C easily. It's easy - all you have to do is behave normally. If you call from 10,000 over the field and tell them you want to land, this isnt the normal routine and they try to ignore you. So, at 5000' I fly away from the airport about 6 or 7 miles, then call Palm Beach approach, N96PM 12 miles south, 5000, landing Lantana" This time he didn't even bother giving me a squawk because I told him I had the field in sight. I was with him for maybe 25 seconds. To get out I call them with something like Experimental Cozy N96PM out of Lantana at 1000 destination North County. Again. He's used to this, and gives out squawks like candy. Bingo. No more sneaking under the airspace for me. Hobbs time now 27.5.

Time for a clean-up

Sunday 3/20 didn't work out quiet how I'd planned. Char volunteered to come down to the hangar and help me clean up. I flew for 0.9, then came back and helped. After we'd done a good job in the hangar and had some lunch I snuck back and did another 0.8 in the sunset. No high power runs today. Just some ASI calibration and continued familiarization with the airplane. I collected a few numbers...
On the first flight:
5500' OAT 57, IAS 130, 29.92 alt = 5300, 4320 rpm, 
Average ground speed 145, coolant 175, oil 186, oil pressure 72, 
coolant pressure 5, gph 9.2, egt 1419. Leaned mixture to gph 8.6, egt 1505

Second flight:
1000', OAT 69, GPH 10.7, IAS 138, calm
5000', OAT 61, GPH 10.3, IAS 135, Average Ground Speed 149  
6500', OAT ?,  GPH 10.3, IAS 138, Average Ground Speed 154, MAP 32
Obviously these numbers are very provisional. The gallons per hour (gph) calibration needs to go through a few more cycles before I can be sure of it, and I'm a long way from perfectly tuned.

The Palm Beach controller was as busy as a one-armed paper-hangar today. At one point I heard him say VFR Traffic - don't bother calling for advisories. VFR traffic already in the system - I'll be dumping you shortly. He then proceeded to hand out "1200's" to four or five aircraft. I took his advice and didn't bother to call. Tripping out under the Class C isn't as nerve racking as it used to be.

The one problem I had was with the left fuel pump. It seems to work ok for 10 minutes or so, then the pressure drops off. If I leave it the mixture drops off the scale and the engine starts to faulter. As soon as I switch to the right pump the pressure comes back up and the mixture comes back to the middle. I had fuel in the left tank I wanted to use, so I used the left pump intermittently and during descent, then flipped to the right pump for landing. There seems to be little doubt that I need a new pump. Total time now 29.2.

As I come up to the 30 hour point, one thing is becoming obvious. There's no end to the testing. Certainly not at 40 hrs. The sign-off is like a new pilot ticket - it's a license to go out and learn. I'm sure it will be many months, if not years, before I'm "finished". A little tweak here, a little (or a lot of) paint there. I have a lot to do before I can settle down and say I'm done developing this airplane. That's the nature of the beast. Tracy Crook, Ed Anderson and others have been flying they're rotaries for years and they're still making them better, and discovering new issues on an almost daily basis. Passing the 40 hours is just the next stage. That doesn't mean I'm not looking forward to it.

Pulling My Weight

Today 3/21 was perfect for my first C of G and gross weight test flight. The wind was significant and straight down the long runway. On the way to the hangar I stopped at the local Firestone store and, with permission, emptied their bucket of used balance weights into my trunk. I found a suitable box, added all the lead weights, a spare turbo manifold, and a couple of old batteries to get to 135lb, then strapped the box into the passenger seat. I took the 65lb of lead out of the nose and put it in the box. Now I had a simulated 200lb passenger. With 18 gallons of fuel on board that put me at just aft of center in the c of g box and a gross weight of 1776 lb. By the time I had everything prepared it was getting close to dusk, but there was time for a little pattern work. I checked Palm beach ATIS. Wind 160 @ 13 kts, Density altitude 1100'. We were using runway 15.

Take-off roll seemed normal. I was at 75 kts at about the same spot on the runway. I held her down until I saw 85 kts, then let her unstick with maybe 1000' to go on the 3200 foot runway. No problem, she was flying well. That's when everything got different. I was looking down at the shopping center as I started the crosswind turn. It seemed a bit too close. My RAF instructor taught me to judge 100' for low level work by counting the legs on cows. There weren't any cows, but I could count the legs on the people. A little closer and they'd recognise me. A glance at the altitude showed that I was at 150' where I'd normally be at 400'. Uh oh. Then I saw the airspeed. 125 kts when I would normally be at 110. I gradually figured it out. Not wanting to get into a canard stall on take-off I'd overcompensated. I need to pull harder. I pulled back to 85kts and she climbed nicely to pattern altitude. On the downwind I thought about it some more. They don't call it the "Strong" pitch trim for nothing. You have to be strong to pull against it. Well, not really, but I'd got used to the amount of pull needed and used that pull automatically. In reality it only took a few seconds for me to realize that I needed to pull more, but it was definitely a bit of a suprise. In the pattern the plane flew just as it always has. The approach and landing were quite natural. No big changes there. That take-off needs a bit of practice, and I should get into the habit of adding more take-off trim before I swap the lead for a live person that can smack me. (naming no names, of course). I could definitely use a bit more power on take-off - that's no suprise since I'm only turning the big prop at 4050, but I have to remember what it was like to take off heavy in a Cessna or Cherokee before I over-react and chop it down by 6 inches. Maybe it's just a matter of getting used to it. Just 0.3 hrs today, but it's another step in the right direction.

Back at the hangar I added 5 gallons of fuel to each side. I'd brought 20 gallons, but I think I'll hold off on the other 10 and build up the weight slowly to full gross. Besides - I've got used to the smell in my car.

Graduation from Builder to Flyer

I've noticed that my old epoxy stained work shorts and tee-shirts have been disappearing. I put them in the wash (i.e. leave them on the floor in the bedroom) and they just go away, never to return. Yesterday Char showed up with a bunch of new casual clothes for me. Flying Clothes, She announced. I must have mentioned that I felt a bit scruffy when visiting other airports as a "traveling pilot" dressed as an "airplane builder".

Something else has changed. My goodbye words used to be Have a good day or something similar. Once I started flying, the parting words changed to Play Nice!. Since I dinged the prop I get Play Nice, and don't be rough with your toys!"

The next few days were a bust. On Wednesday I was standing on the ramp considering the flight prospects. Wind was still 18 gusting 23 at a slight crosswind to the main runway and the forecast is for worse over the next few days. The aluminum was using the short runway. Maybe they're practicing crosswind landings. Ceiling was 2500 broken. Hmmmm. About then I felt some rain on my face and heard a crack of distant thunder. OK. That's at least "three things". I climbed in the plane, extracted the EC2 and headed for FedEX. I spoke to Tracy earlier. He promised to replace the bad chip and return it as quickly as possible, together with a beta version of the autotune chip for the EM2. I guess he figured that if I can't break it, nobody can. In the meantime I have a replacement fuel pressure sender on order - the one for my EM2 died, and I'm looking for a replacement fuel pump. Hopefully by early next week I'll have all the remaining squarks (and squawks) fixed, and will be able to finish off the test flights. One reason I wanted to get the EC2 A computer fixed and get autotune now is that I need all the power I can get to launch full gross off this 3200' runway, and I don't need to be fiddling with the mixture during the take-off. The B computer doesn't adjust for intake temp, and that's probably costing me some power. If the autotune system can tune Tracy's computer better than he's ever been able to do, then it's bound to improve my amatuer tuning by a large margin. So. Don't expect any more updates till Saturday, or Monday of next week.

The bad news is that it was PERFECT flying weather for the next 2 days while I sat at my computer working. The good news is that my EC2, now repaired and tested, an auto tune chip for the EM2, a new fuel pump and a new fuel pressure sensor are all on their way. The EC2 and autotune chip came in late Saturday.

Canardian Chance. Go back 18 months. Do not pass Go. Do not collect $200

Easter Sunday started off as a great day with wonderful news from my daughter & her husband in CA. They're expecting their first child in November. Julia, who's working through Med school, has the birth perfectly planned to occur in-between her Gynecology and Pediatrics rotations. She got an A in her Gynecology studies. I think she's trying for an A+.

After consuming half my chocolate rabbit I searched around and finally found some work clothes Char had missed. Down at the hangar I reinstalled the EC2, then wrote down all the calibration data in the EM2.With this done I removed the EM2 board, swapped the chip for the new autotune version, reinstalled the EM2, then reentered all the data for my sensors. All this took a couple of hours. I was anxious to test the new autotune feature, so I wheeled her out and climbed aboard. The EC2 MAP table data was a straight line with all the bars in the middle. I recovered the default settings and copied them to the B computer.

She started, but ran very roughly for a few seconds, then cut. I tried again. This time I was able to get maybe 30 seconds out of her. The engine was running so roughly that the canard was vibrating maybe 3 inches or more at the tips.

To cut a long story short I spent the next seven hours changing plugs, trying different ideas and calling Tracy (Yes - I got him on Easter Sunday). All engine start attempts were essentially the same with no improvement from any of changes I made. The engine fires right away, runs very very rough, shakes the plane and is barely able to keep itself going. After 30 seconds or so it usually stalls, but will start again if I hold the starter. There is a rich smell from the exhaust. When removed both trailing plugs are wet and black. I cleaned and/or replaced the plugs between each engine start attempt. Here's a list of the things I tried:

- Checked all coils for good spark. 
- Recovered default injector flow and timing settings.
- Adjusted mode 3, (injector flow) 20 + clicks toward lean. No improvement. 
- Adjusted mode 8, (timing) 10+ clicks in either direction. No improvement.
- Reset both mode 3 and mode 8.
- Checked vacuum hoses to the EC2 (right to the board) 
- Temporarily clamped off fuel pressure and manifold pressure T'd lines. No improvement.
- Checked the EC2 connector for loose wires. No problems seen.
Not wanting to call Tracy on Sunday evening I sent him an email listing what I'd tried. He replied within the hour, but wasn't able to suggest anything new. Unless the gremlins are down here for Spring Break, it seems unlikely that something major has failed all by itself while the plane sat unattended in the hangar. The EC2 is the only thing that changed, so if I can't get anywhere with it in the morning, it has to go back once again and another week will be lost. Color me frustrated.

Just to finish off the day, the manual shift in my Saab seems to have come loose and I can no longer get 1st or second gear. Pulling away from lights in 3rd is getting to be a pain, but down here, no-one seems to notice. A visit to my Saab expert is planned for the morning.

Turning a Setback into an Opportunity

I sent Tracy a note telling him the EC2 and EM2 were on the way back. I asked him to let me know what he finds, then added "Please find SOMETHING!"

The plane is grounded for the rest of the week. Rather than bitch about the fact that the entire week was perfect flying weather, I decided to treat this as an opportunity to do some maintenance work.

First, I've been planning on doing something with internal cowl ducting. The oil leaks have shown that air is exiting the cowl in all sorts of places I hadn't planned on. It's coming out at the back corners of the wing roots and even from the screw holes. There's obviously a lot more pressure in the cowl that I'd expected and, even though it seems to be cooling just fine, the airflow is all wrong. You watch. I'll fix the airflow and screw up the cooling. Ah well.

I wanted to seal off the wing roots, but it's hard to measure, let alone do a layup, inside the cowl with it in place. I made ribs with urathane foam and adjusted them bit by bit removing and reinstalling the upper cowl until it sat nicely on top all the way along. I duct taped the lower cowl and the ribs along the inside and the top, did a 3 BID layup, added some flox on top ala strakes construction, then lowered the upper cowl in place and screwed it down. After cure the new fiberglass ribs were attached to the upper cowl and had a nice ledge that was a perfect fit where they intersect with the lower cowl. I added a couple of nutplates to secure them to the lower cowl. Thinking that this is about pressure as well as flow I made baffles to seal off the aileron torque tubes and rudder cables. I riveted a strip of silicone baffle material to an aluminum angle. I'll make holes in the silicone to let the aileron tubes and rudder cables pass through. These silicone flaps "should" seal off this area fairly well.

Why am I using fiberglass close to the turbo? Because it's much easier to form than steel, and I've found that it takes the heat very well. The top of the cowl is just as close to the turbo, and I've had no problems with it melting. Even behind the hot exhaust the fiberglass stands up very well. It's a bit discolored, but its structurally fine. I'll put aluminum duct tape on the inside and I think it'll be fine. If not, then I can do something with fiberfrax and/or SS sheet later using (what's left of) the glass as a base.

Next came the outlet holes. When I made the cowl I formed nice curved outlets, but left the inside as a barrier to airflow. The air must be hitting the inside of the curve and bouncing around all over the place. Using "great-stuff" (Urathane pour foam would have been much better, but I didn't have any) and micro I formed a bump around the exit to smooth the airflow. The bump will also help accelerate the air as it heads for the exit. I'm going to have to modify the rib on the left side to get around the intercooler which sticks into the wing root a little.

News from Shady Bend

I'm not sure if it's good news or not. Tracy tested the EM2 and EC2 and found nothing that would explain the symptoms I had. He WAS able to duplicate the scrambled data syndrome on the old A chip by introducing external noise, but not on the new "flash" chip. There was a "turbo" bit set wrongly on the EM2, but this wouldn't affect the engine, just the MAP read out. In an attempt to cover all bases, Tracy is sending the EC2 back with the new chip in A, and the old chip (the one that worked) in B. He's enclosing a new chip for B in case this isn't the problem and I want to upgrade. The bottom line is that if it doesn't work, something else is broken.

In the meantime I removed the intake system - no mean task - to double check for stuck injectors, chase after the oil leak and, as it happens, replace the injector O rings which seem a little worse for wear. One of them got crimped. While the intake is off I'll clean up the top of the engine, tidy up some wiring and check all the fuel lines.

I'm also getting on with the annual condition inspection using a modified version of Marc Z's checklist. It's exactly a year since the plane was inspected and approved for flight. I'm not sure if the annual runs out a year after the first flight, or a year after the inspection so I'm getting it done now to be sure. The EC2 will probably be back on Sarurday or Monday. I'm putting in a lot of hours down at the hangar, but I probably won't be ready for it when it arrives.

More Maintenance

While I had the intake off I checked all the fuel hoses and put 12v on the injectors to make sure they clicked properly. I went down to the Mazda dealer to pick up new injector O rings. Eight o rings, and he wanted $118 plus tax. I took four, but will be returning them for a refund since the old ones are in perfect condition. What really needed replacing were the rubber sleeves (I think they call them insulators) that hold the primary injectors central in their slots in the engine. Mazda didn't have these in stock, so I made them by grinding a piece of silicone hose to the right size ring. Total income from me at the Mazda dealership = $0. While working on the intake I noticed that my boost control cable had lost it's C clip and was hanging loose. I replaced the C clip and safety wired it. I did an overall check and cleanup of the entire engine, checked the surpentine belt, neatened up all the cables, checked all the hoses and connectors, and replaced a frayed thermocouple wire. As part of the annual I removed the spinner and retorqued the prop. Interestingly, two adjacent bolts took a half turn, and the rest were good.

It could be wishful thinking, but it looks like one of the sensor hoses under the intake may have been seeping oil. I used teflon paste on all the connections and tightened down the sensors. I finally got the intake reinstalled on Friday. Did I mention that removing the intake is a tough job. Reinstalling it is much worse. The clearance is so tight that the intake gasket is enough to stop the intake going over the engine. I have to slide it in place after the intake is on. The two rails and four injectors are on the underside of the intake, but can't be bolted up until the intake itself is in place. It's an engineers nightmare, but it works and it fits - just. Several scraped knuckles and 5 hours of hard work, and the job is done. The annual is complete from spinner to firewall.

I spent nine hours on Saturday working my way through the annual check list. Anything that hasn't been off recently came off and got checked. I also completed and fitted the silicone baffles to go around the aileron torque tubes. These will sit on the inboard side of the new ribs I made for the upper cowl and seal off the wing roots. I'm a bit wary of installing something over the aileron torque tubes, but they dont interfer at all, and they passed Paul's inspection so I feel better. The EC2 and EM2 came back mid-day so I went back down to the hanger and fitted them. The engine still needs some fluids replaced and it was getting dark, so I held off on trying a start until the morning. Instead I did a small layup on the inside of the cowl to stiffen up the baffles where they join the curve of the upper cowl.

Reptiles

Sunday started off well. Once the traditional Sunday morning "duties" were completed I headed down to the hangar. A few minutes later, as I was just starting to apply some RTV to the new aileron baffles, Char called. There's a snake in the living room!" Apparantly one of the cats had caught a small snake and brought it home to play with. Both cats were going berserk chasing it around under the couch, even under the carpet. Grabbing my cape, I headed home to deal with the snake. On the way I found myself following a truck with the bumper sticker You. Me. Whipped cream. Handcuffs. Any questions?" Hmmm. I didn't know the Cozy girrrls were in town. Then, of course, the sticker didnt mention oil, so it couldn't be them. As I pulled alongside curious to see the driver, I saw a very large igwana on the back seat. He had his paws(?) on the window ledge and was watching the traffic like a pet dog. OK. I pulled level. The driver was a young guy, so I lost interest and headed home to be a hero. By the time I got home the cats had cornered the snake and Char had siezed the opportunity, covered him with an upturned bowl and weighted it down with our wooden Pedro statue. I slid a board under the bowl and carried the snake to the very back of the yard before releasing him. He was a good 18 inches long, and not at all happy.

Back at the hangar I removed covers and carried on through my annual checklist. Nothing was bent, rusted or broken. All the welds looked good and the engine mount is fine. I don't like the way it hangs on the upper mounts and compresses the lower mounts. This must allow a slight upward movement of the engine on thrust, but it's secure and there are no signs of cracks. The main tires are OK, but they'll need replacing soon. Brake pads are fine. I installed the new fuel pressure sensor and calibrated it. OK. I finished the checklist. Using my new Repairman Certificate, I signed off the conditional Inspection in the Aircraft logbook, and signed off the bolt torque in the prop logbook.

Note: I took a high resolution image of one side of the engine which some might find interesting. Click on the image for the full size one.

It was time to test start the engine. But first I needed to clean off all the dust from the wings and canard and clean up the mess I'd made on the strakes. Finally I pushed her out (with the cowl still off), did my walk around and climbed in. Please start normally, I thought. That shaking and rough running I'd had last time felt an awful lot like serious internal damage. (Yes. I DO know what that feels like). I'd been second guessing this ever since Tracy said he couldn't find anything wrong with the EC2. Maybe I overboosted on that last heavy take-off. But she'd run just fine around the pattern. I'd been able to get each rotor to run with the plugs off the other one, so it wasn't damage to one rotor only. Maybe the side seals had given way because of detonation. The compression seemed good, but without the proper tester it's hard to be sure that all three pulses are equal...

I set computer A, cold start on, full rich, fuel pump on, pressure good. I shouted clear and cranked the starter. She fired up and ran exactly as before. Very very rough, shaking the canard and no throttle response. After 15 seconds she gave up and died. This was what I'd been dreading. I tried again. She shuddered some more and staggered up to about 800 rpm before dieing back down to nothing and stalling. Damn. One more time. Same result. I flipped over to the B computer and cranked.

WvvvvvvvvvvRRRRRROOOWWWWW! I had to stand on the brakes because she wanted to FLY. Hmmm. At least there's nothing wrong with the engine. I taxied out to the ramp on B. As I taxied I flipped over to A. No difference. The engine was running fine. When I throttled back and the rpm dropped below about 1500 the engine immediately went into the shake and shudder routine and tried to die. I flipped back to B and it picked up again. After a few minutes of experimentation and a short runup I taxied back to the hangar. There's something seriously wrong with the A computer. It could be mixture settings - I haven't checked the MAP table yet - but the mixture control didnt solve the problem. I think theres more to it than that. I decided to report back to Tracy and see what he has to say.

After shutdown I found another little problem. There was quite a bit of smoke coming from the turbo heat shield. At first I thought it was oily fingerprints burning off as I've seen before, but no - this was different. With a flashlight I could see that the side of the engine in front of the turbo was wet. I wiped it off, and it got wet again. I smelled the rag. Gasoline. I shone my flashlight in between the intake and the engine. There was a small pool of gasoline on top of the engine which was spilling slowly down the side of the engine and onto the heat shield. Yikes! I soaked the gasoline up with the rag and stood ready to run for the fire extinguisher. In a few minutes the fuel dissipated, the heat died off and I was able to relax. I turned on the fuel pumps, then checked the intake. No leaks. The leak must only occur when the injectors are firing. I must have gotten something misaligned or crimped an o-ring when reassembling the rails. Removing the intake again wasnt on the top of my list of things to do Monday. Ah well. At least the engine is in good shape. Everything else is a minor issue compared to that. Content with the day's events I went home to draft an email for Tracy.

When I got home Char mentioned that there was a large lizard under the table. It wasn't moving. Just sitting there and staring at the cat. Perhaps there's something wrong with it, I thought. I picked it up by the tail. I'll tell you what's wrong with it, my lad! It's dead! I sent it for it's last long swim to wherever dead lizards go. Another day in the life of a Florida experimenter. As a side note, I hear that Rich Hughes picked up his new Cozy III today. Congrats, Rich. Welcome back to the world!

Nothin's Up

OK. I know you're wondering what's up. Answer: Nothin's up. Much as I'd like to get down to the hangar and fix the fuel leak, something got in the way every day this week. Frustrating, but to quote a certain dogmatic builder I know, that's just the way it is. Maybe tomorrow...

Friday 4/8, I finally made it to the hangar in the afternoon. I wanted to experiment with the EC2 first, so I pushed the plane back and fired her up. No start on A, but she ran fine on B as before. I checked the MAP table for A. It was a straight line of bars showing 50%, So I reset the defaults and tried again. No better. It ran well on B so I ran the engine for a couple of minutes switching back and forth between the computers. Once started on B it runs fine on A down to 17 MAP, then, if I go lower it backfires and cuts. If I'm quick switching back to B it'll pick up again.

After a couple of minutes I shut down to check for fuel leaks. Sure enough, there was a small pool of fuel under the intake again. OK, the intake has to come off. I'm getting pretty good at removing the intake. In half an hour I found what I think was the problem. My home made sleeves for the primary injectors were a bad idea. When I first assembled the intake I'd used O rings, and they'd worked fine until I crimped one during one of the rebuilds. The proper sleeves weren't in stock at Mazda, so I replaced the sleeves with new O rings being very careful not to crimp them as the injectors went into the holes in the top of the engine. I realigned the secondary rail and in another 3 hours had everything back together. This time, no fuel leaks. The engine runs very well on either computer. It just won't idle, or even start, on A. I called Tracy to discuss the next move. He feels it must be something to do with the ignition logic on the A computer but he doesn't know what. He's been looking over the code trying to find a problem without success. I suggested that it might be an incompatibility between the new chip and my older board that only shows up in some weird circumstance. I'm very good at creating weird circumstances.

Tracy agreed that the next step was to swap out the entire board and replace it with a new one. The only problem with this plan is that he's out of stock. The next batch of boards is due any day, but Shady Bend is flooded and there might be a problem with the delivery, and then he's away at Sun & Fun all next week. Ah well. Tomorrow I'll run the plane on the ground with the cowl off. If everything looks good I may decide to continue the flight testing. If the A computer was totally dead, then I'd be doubtful, but it runs fine at anything above 17 MAP. I didnt use it on the last few flights anyway. Hmmm. I'll try a few run-ups and taxi tests, then decide. Of course, I could always get on with the painting instead......

Tomorrow is the anniversary of the first flight. I really wanted to have the flight testing done by now, but it looks as though the kitten has it's own schedule. Serves me right for trying to tame a cat.

Happy Anniversary

Saturday 4/9, was kitten's anniversary so I gave a her a new fuel pump to celebrate. I think I found the problem that had been causing intermittent low fuel pressure on one side. There was a small amount of "crud" in the inlet side of the pump. I'd estimate about 1/2 cc of what looked like urethane foam dust that had got through the strainer and, presumably, was shifting around and occasionally causing a minor blockage. My EFI filters are after the pumps. Put them before the pumps and they can block and cause the pump to cavitate. The failure mode this way was fairly begign. A few short unexplained drops in fuel pressure, one of which was enough to slow the engine a little. Hopefully this is the last of the construction crud, but at least now I know the symptoms if it happens again.

Since I'd already obtained a replacement pump I put it in anyway. Interestingly the original pump, that I got from Tracy, was a Walbro Inline fuel pump-GSL393. This is identical in outward appearance to the Walbro GSL394 I just replaced it with. The 394 is a slightly higher price model with slightly better specs. Both models have all metal construction and 1/8 NPT metal fittings either end. I got mine from Lightning Motorsports, an aptly named company - they're unpredictable, unresponsive to suggestions and on time delivery of their products seems to be about a million-to-one chance. Given the crud on the left side I removed and checked the right pump. No crud. By the time I was done testing the pumps (34 PSI each, 38 PSI together) and reassembling the back seat area it was time for food so I didn't even get to run the engine today.

On Saturday evening I went back to experiment some more. Tracy had suggested I moved the staging point (the MAP point where the secondary injectors cut in) to see if this changed the MAP at which the A computer cut the engine. I set the staging at 20 MAP and this did seem to advance the rpm / MAP at which the engine cut on A. I always get a distinctive backfire when it cuts. The problem came when I tried to set the staging point lower again. I thought that maybe I could set the staging point at something really low, like 10, maybe this would eliminate the problem on A. I don't seem to get below 15 - 17 MAP at any throttle setting. Now I'm wondering if this is normal. On the B computer closing the thottle gets me a slightly uneven idle at about 850 rpm, but the MAP still reads around 15 or 16. I'd expect lower. Maybe the turbo is still pressurizing the manifold a bit. My next experiment was to give the autotune a try. With the EM2 in track mode I brought the MAP up to about 22, cut over to the A computer and changed the EC2 to mode 9. The engine cut. I tried this about 3 times with the same result each time. I returned the EC2 to mode 0, went back to computer B and did some taxi testing. The EM2 was still in track mode. A couple of times I noticed the bar graph change completely for a second or two, then change back without any throttle adjustment or rpm change. The briefly displayed screen showed a completely different bar graph with a few higher bars and gaps between them, then I was back to a more reasonable looking graph. Very strange. All running on B was fine. Everything seemed normal on a runup and a couple of high speed taxi tests. By this time the temps were getting high, so I taxied back to the hangar and shut down. There doesn't seem to be any leakage of oil or fuel, so at least this is fixed.

Grounded

On Sunday I ran the engine again using computer B. Everything was normal. I tried a run-up and a high speed taxi test on the A computer and again, no problems. Basically it comes down to the fact that I can't idle on computer A. I'm beginning to think that I could live with this until I get it fixed. I installed the cowl and got the plane ready for flight, but there wasn't time on Sunday. I had a social event to attend.

Monday morning I donned my best white flight testing underware (Hey - how else you gonna know?) and went down to the hangar prepared to take her up provided nothing else came up. Something else came up. The A computer data was a mess, so I went to recover the defaults. When I turned the master back on the EM2 flashed NOP in the rpm readout. I don't know what NOP stands for, but it indicates that the EC2 and EM2 are not communicating. This will happen, for example, if the power to the EC2 is off. Both EC2 power switches were on. I checked the power at the EC2. 12v. OK. I can't see the B computer data, but I can test it. I started the engine on B and it ran fine. Ergo - the A computer is dead. That decided the issue. I'm not flying with only one computer which just happens to be in the same box as one that just upped and died. This plane is grounded until I get the replacement computer. It'll be a week to 10 days before I get it, so I reached for the masking tape and started remove panels for painting.....

So what's the good news for today? Tracy will be at Sun & Fun when I drive up on Friday. I'll save the FedEx costs and will be able to deliver the EC2 by hand. I hope I can restrain myself from delivering it from 50 feet, "frissbee style".

I'm looking forward to meeting old friends and new builders at the Cozy BBQ on Friday evening, then it's back to work preparing the plane for repainting. Why does it need repainting? In case you haven't read that section, the PolyFiber Top Gloss paint I used is absolute crap. So much so that they withdrew it from the market after numerous builders experienced peeling, rash-like bumps, gasoline staining and various other problems. The president of Poly Fiber sent me, at no charge, a complete set of materials to repaint the airplane with their new paint system. What he didn't send, unfortunately, is the man hours needed to remove the old paint and prepare the surfaces for repainting. Ah well. It's not like I have anything else to do for a week.

Sun & Fun 2005

We got up early and drove up to Lakeland. As we pulled into the parking lot I managed to reach my daughter, Vicky, on the cell phone and wish her a happy 24th birthday. Then came the dangerous question - What would you like for your birthday? Vicky has unusual tastes, and fullfilling her birthday gift requests has led me on a few strange searches. This year was no exception. A Didgeridoo. OK, I said, unshaken. Consider it done. Hmmm. Who do I know in oz? Maybe Max Heywood at Turbonetics could help. I didgeridon't think they carry them at Sears.

Char & I headed across the field to deal with task number 1 for the day - delivery of the EC2 to Tracy. As usual his RV was surrounded by prospective wanklers, so I waited until there was a gap in the traffic to hand him the bubble wrapped bereaved board. .I'll find out what's wrong with it, said Tracy. Thinking of the lizard and John Clease's parrot I couldn't resist such a perfect entry line....

I'll tell you what's wrong with it, my lad. It's dead. This board wouldn't "voom" if you put four million volts through it! It's bleedin' demised! It's passed on! This board is no more! It has has ceased to be. It's expired and come to meet it's maker! It's a stiff! Bereft of life. It's kicked the bucket, shuffled off it's mortal coil, run down the curtain and joined the bleedin' choir invisibile! THIS IS AN EX-BOARD!!

OK. I didn't say all that with prospective customers in earshot, but I thought it. What I did say was that this board has gone back and forth too many times, and I'd rather not see it again. Ever. Unless maybe we could pot it in clear epoxy and frame it. Tracy agreed to send me a replacement EC2 as soon as the new batch come in. The problem is that the road to Shady Bend is 2 feet or more deep in water. He got his plane out before the runway flooded, but he and Laura literally used a canoe to get to their car. After a pleasant visit with Tracy, Char & headed over to our next port of call - the Polyfiber booth. I wanted to talk with their President, Jon Goldenbaum about my Top Gloss paint and discuss the best way to deal with it. Jon was over in their other tent and I never did manage to hook up with him. The next 30 minutes in hangar 1 were expensive. Char needs a headset. I'd been looking at the price range and considering one of the less expensive ones. T'was not to be. Char made a bee-line for the Lightspeed booth and fell in love with their top-of-the-line ANR model. We were discussing it over the intercom in their sound booth, and I was about to resist, when I realized that we were having this conversation in very low tones. I was wondering where the volume control for the intercom was, then realized that I didnt need to turn up the volume. I have Tinitus which makes seperating the wheat from the chaff very difficult. With the ANR and very high passive noise reduction there was zero chaff. I could hear every word with perfect clarity. I have a Pilot ANR and it's very good, but this was dramatically better. OK. We bought the lightspeed for Char, and they gave me a nice (free) Lightspeed flight bag to go with it. Char gratiously agreed to let me use the Lightspeed when solo. If it proves to be as good as I expect I see a matching set in our future. Char suggested that we use them around the house.

Next we came across the MicroMesh booth. My canopy has a lot of light scratches which get in the way a bit when flying into the sun. I used the old micromesh kit on it years ago but the hand sanding is very tedious and there are new scratches since then. Todd says they're probably caused by my wiping the canopy with dust on it. Air blowing is the safe way to clean a canopy, but it doesn't seem to work well. The new Micromesh kit comes with velcro pads for use with an drill or DA sander. I bought one.

As we walked by the Wicks booth they had a guy demoing the Fein sander. I've heard wonderful things about these, but had decided it was too late now and not worth the investment. I bought one.

As Char admired the jewelry I spotted a booth with LED flashlights. They have small red, green and blue map reading lights that last 200 hours. I bought red and green. If I have a nav light failure I can stick these to the winglets. They even have flash mode. One weird thing about these map lights is that you can't turn them off. The "off" setting is a low glow that lasts 20,000 hours. This way you can always find them in the dark. They also had a small white LED flashlight. I bought one.

As I escaped hangar one I passed the AirCharts booth. They looked me up and offered pro-rata credit for the 4 months I'd missed last year. I bought the upgrades. On the way to the food section was the Snap-on-Tools booth. I got some indestructible torx drivers for my cowl screws. Finally (almost) I passed the duats booth and picked up their new (free) flight planning CD.

I never did make it to the flight line. When the warbirds and noisy aerobatic guys started their annual battle song we toured the other hangars. It was nice to walk quickly down the aisles looking side to side and thinking, Got one of those, Got one of those, Don't need that, Got one of those. Apart from some new screws for the spinner at a hardware booth I was done shopping, or so I thought.

I spent an hour at the Cozy booth chatting with builders. Dan Crugar (Flying Cozy IV) has agreed to sell his Lycoming 360 to Jerry Schnieder. Why? Because his 13B Renesis rotary comes in from Bruce next week. Jerry is also planning a rotary, but wants to get in the air quickly, then he'll pass the Lyc on to some other builder once he has his rotary running on a false firewall. Makes sense. I think I see a trend developing. Nat seemed particularly disappointed that my plane was not there for him to see, and unusually interested in why I'd taken a year to fly 30 hours. I wonder, could he be looking for a way to put out the rotary forest fire that's sweeping across his domain?

Two other booth visits were interesting. Greg Richter showed me his new Sport unit and got me drooling again. At the end of the evening I spent a good hour chatting with François Badoux, CEO of Mistral Engine. We discussed the stock turbo which I'd noticed in their engine shots. They don't work, said François. They blow up at 10,000'. I smiled. Yes, I know. I explained my experience with two stock turbos while François nodded, then, when I described the solution I'd found at ATS in Au, he started making notes. We discussed his planned approach to running the rotary on Jet-A, and then I started making notes. I got the impression that the delays and all the custom tooling have left Mistral a bit cash-weak. In some cases, like the turbo issue, they seem to be a little too quick to head off and develop their own custom solution when there are existing options already available. Like many European companies, they're openly looking for investors, which is never a big plus when you're trying to create credibility. If it were me I'd be looking for customers. According to François they now want $32k for their base engine which seems like a lot until you consider their market, and the man hours and risk associated with developing your own unique installation. They're not targeting home-builders, they're after the bigger and currently "available for the taking" market of certified engines for 172's and Pipers. I found François's pitch on redrives suprisingly tacky, inappropriate, and out of character with the rest of what he had to say. "It'll only last two or three hundred hours, he said, referring to Tracy's redrive. Countering that, he was pleasantly open about their plans and their failures, which I found refreshing. If Mistral can get their Jet-A conversion working they could well be poised to experience corporate VTOL. As we chatted, Mrs. Lamar stopped by to invite us to the ACRE evening dinner. François politely declined, as did I.

At the Cozy dinner I enjoyed chatting with old friends and new builders. It was a strange feeling to move from group to group. I knew 50% of the people there, and the other 50% knew me. Nick Ugolini described the baking soda power wash method of removing paint. Sounds a bit drastic because it takes off some of the micro too. Blake showed me the Icom handheld Nav/Com he'd picked up for $239 at Gulf. I wish I'd seen them, I said. I'd have bought one. Unfortunately I'm driving back home tonight. No problem, said Blake. I'll pick one up for you in the morning and drop it off at the hangar on Sunday. You can pay me back then. Char thought we were done shopping for the day. The booths were closed - right. Apparantly not.

In all it was a great day. 5 hours of driving. 1 hour of shopping and 13 hours of talking Canard. We got home exhausted and broke at midnight, but we'd both had a fun day. I'd packed a lot into the trip, and had a bunch of new toys to play with. Sorry there arn't any pictures. I left the camera at the hangar. Simon Ramirez posted posted some pictures on his web site., so I'll just link to these.

Payback's a Bitch, ain't it.

We've ordered a wood floor for our living room, and the materials come in on Tuesday. Instead of spending Saturday prepping the plane for paint, I spent it prepping the living room for wood. I got about 1/4 done.

How's it Going?

Thanks for asking. Pretty good. After two days of hammering, I managed to get the three layers of 20 year-old tile off the slab and scrape up the nasty tar adhesive. Once my right arm recovered, I removed the baseboards, spackled the walls where needed and repainted all four walls. As of today I'm almost done laying the new wood floor.....

Oh - you meant the airplane, didn't you. Airplane. Hmmm. I think have one of those parked somewhere.

I got an email from Tracy telling me he still could find nothing wrong with the EC2, but will send me a new replacement unit anyway. He still thinks it's a "noise" problem. I don't see how electrical noise can disable a computer in a totally consistant manner, while leaving the one next to it unaffected. Also, I can't reconcile the noise theory with the fact that both computers were behaving in exactly the same manner until one was swapped out for an older version, which then worked normally. I followed the wiring instructions carefully with respect to shielded wires and keeping the injector wires away from other wiring. I've checked all the grounds and the power with a good multimeter. Without a scope I'm not sure what else I can do. Another consideration is that if electrical noise can cause such symptoms in the ECU while all the other electronic gizmos - radio, intercom, transponder, GPS, etc. are totally unaffected, then I think I'd rather have an ECU that isn't so susceptible. The most recent failure has me particularly worried - the engine wouldn't run at all on A, yet Tracy says the A computer is fine. This does not leave me with a good feeling. My life depends on this box.

Personally, based on limited understanding of the EC2, but lots of experience in logic, I just don't believe noise is the problem. Either the code is different for the new computers in some way which impacts my installation, or the settings in my B computer were sufficiently different to make the difference. The user has no way to see what the B computer settings are, so at my request Tracy extracted the staging, timing and mixture data from the B computer and sent this to me via email. When the new board eventually arrives I'll be able to program it with these settings if necessary.

If I'm still unable to get the EC2 working Tracy has agreed to send me a full refund which will go part way towards the purchase of a Microtech ECU. This unit seems to be working well for Paul Conner, and it also has a timing split capability for turbo applications. I'm not looking forward to having to install and learn a new ECU, and I don't want to give up on the EC2. The product is working well for 40 or more other rotary flyers (some of whom have the same upgraded old boards as me) but, after fighting this thing for over a year and sending it back a total of four (or is it five) times, the frustration level is getting pretty high. I'd like to get to the root of the problem. Whatever the problem is, finding it will help Tracy and other EC2 users, but if it doesn't work this time, enough is enough.

The rest of the bad news is that Shady Bend is still under water. Delivery of the new boards is now expected around the end of this month (April '05). Looks like I'll have time to finish the living room, do a few other honeydos, then spend a week or more getting the plane ready for paint.

Turning the calender

Geesh. Here we are in May, and the airplane's still grounded. The living room's all done, I'm pretty much caught up on work, and the flood at Shady Bend has subsided. There's still no word on my new EC2, so I've been doing a bit of work stripping the #@$%ing Top Gloss off the plane. I've found that if get a sharp razor scraper under the paint it comes off in sheets. I've also noticed that the two coats are not even stuck to each other. I can take a stripped off strip of paint and seperate the two coats. I'm sending a sample to Jon Goldenbaum at Polyfiber for his comments, with a few comments of my own. The repaint is going to be a massive job because ALL the paint has to come off, and the plane's going to look really ugly until it's done. I was bothered by this until I saw a picture of Paul Conner's flying SQ2000.

Another job I've been planning on is adding a K&N air filter to the intake. This has gained priority since Ed Anderson's rotary lost compression from worn and eventually fractured seals. It seems that the damage was caused over several months by dust ingestion, so despite Tracy's filterless example my engine's going to get a filter. Of course, this means providing for a possible blockage with an alternate air path. I'll be making a plenum out of fiberglass with a spring loaded flap for the alternate source.

I have a business trip which I was hoping to do in the Cozy. I can't pospone it any longer, so SouthWest are getting my business yet again. Hopefully, when I get back next week, the new EC2 will be here and I'll be able to just plug it in, fire her up and go flying. Anyone taking bets?

DiggeryDoesn't

When the big aluminum tube delivered me back to Florida on Thursday (5/5/5) the new EC2 was waiting for me. Hmmm. Interesting date. I'm not going anywhere on June 6th next year. Anyway, I heard a rumor that Tracy mailed the EC2, then jumped in his car and headed for Colorado to hide from my phone calls for a week or two.

Friday morning I mounted the new smaller EC2 case, hooked up the EC2 and turned it on. Just as with the previous EC2 before I sent it back, the "NOPE" warning was flashing indicating no communication with the EC2. Thinking this was not a good sign I tried the engine anyway. As before, computer A wont fire it at all and computer B runs it fine. I was in the hangar, so I didnt try flipping to A after getting it running on B. There wasnt any point. As lots of people, and the nagging little b.stard (NLB) on my shoulder, have been telling me, something in the wiring HAS to be causing these odd symptoms. The NLB, and one observant reader, have reminded me that I had a problem with the EC2 wiring to the control board on the panel, and had to put crimps in the wires. These wires go along the top of the heat duct. Somehow a bad crimp or some mechanical problem has to be causing all this, so tomorrow I'll rewire the EC2 from scratch. I'll simply cut all the wires at the PCM (program control module) and at the EC2 connectors and solder in new shielded wires routed along the side of the plane. Another little thought jumped out at me this evening as I read the revised EC2 installation manual. The manual tells you to route the injector wires seperately because they carry sharp pulses and can cause noise in the other wires. I did that. Later the manual tells you to run a wire from pin 17 (rotor 1 primary injector trigger) to the EM2. I did that too - down the same route as my other EC2 and EM2 data wires. Duh! The revised pin 17 connection will have its own shielding grounded at one end.

After installing the EC2 I spent a few more happy hours scraping paint off the plane, then went home to the news that Vicky's Digeridoo arrived from Australia. She says it's beautiful, but unfortunately it was damaged in shipping and it digerydoesn't. An email to the guys in Australia brought a quick response and a promise to replace it if necessary.

Before you ask - How long is the rewiring going to take? - the answer is as long as it takes. This time I'm going to get it right.

And, finally, for anyone who visits this page and hasn't received any useful information recently, here's a bonus. Take a look at the Cozy Girrrls latest invention, then order yourself a little $69 vacuum pump from Alltronics, get a roll of cling film, some paper towels, and change the way you do your layups forever. If you need convincing, take a look at the fiberglass bracket the Girrrls made using this technique in the sixth picture down on their status page. As the TV ads say... "Brilliant!"

Rewiring the EC2

I had a few feet of 3 conductor and a lot of 2 conductor shielded 22g aviation wire I'd obtained at an EAA meeting a long time ago. This seemed like a good time to use it up. By Sunday midday I'd installed seven new wires from the firewall to the panel. Six two conductor and one three conductor. A total of 15 wires. I'd calculated that I needed 13 wires, so this should do it with the EM2 injector pulse detection wire having it's own personal shielding.

I unscrewed the EC2 Program Control Module (PCM) and removed the 15 pin connector. The wiring to the connector was a bit tangled, so I worked to untangle it and get it out of the panel where I could access it. One of the wires was a bit tight, and was holding everything else back, so this was the first candidate for the wire snips.

With the offending wire disconnected I was able to pull the connector out of the panel and get a good look at the wiring behind. What a $#@%ing mess. Butt splice crimps everywhere. Who on earth did this? Obviously some clutz with very little experience in aircraft wiring. Y'know, if this was my airplane I'd ground it immediately, rip all this crap out and rewire the whole thing properly. Errrr. Hold everything. This is my airplane, and the clutz that did the wiring was me, two years ago. I've learned a lot about wiring in the past 2 years, and that's what the NLB has been trying to tell me. During one of the panel reworks I'd needed to move the PCM. I'd used aviation grade crimps, but still, the result was ugly and open to corrosion. About a year ago I read Greg Richter's little paper on aircraft wiring and haven't used a butt splice crimp since. Where you HAVE to have a joint, heat shrinked solder joins are neater and much less prone to failure. The EC2 37 pin connector is too big to feed through any conduit so, with the EC2 in a pusher you have two choices - solder the tiny pins in place, or cut and rejoin the wires. I paid for Tracy to make a harness. Money well spent if you're not very adept with a solder gun, but then had to cut and rejoin the wires at each end to install it. Later I moved the PCM and had to make another cut and join. The result - three sets of crimped joins. That's about 45 crimps in a humid climate. Add the potential of mechanical strain down the heat duct and I'm starting to feel confident that I'm on the right track for solving this problem.

I removed the cover from the 15 pin connector and traced the snipped wire. Pin 13. This is the cold start trigger which gets grounded to double the injector flow rate. It's grounded either by a switch on the PCM, or by the double pole switches for the injectors. This way you get increased flow to the remaining injectors when you disable a set, so the engine can still run smoothly on slightly reduced power if you get a bad injector in flight. I traced the wire back into the harness, cutting wire ties as I went. First it went to a crimped butt splice where it joined three other wires. One went off into the bundle that headed down the heat duct to the EC2. The other two wires went into the main bundle behind the panel. I traced them back, and back, and back through the harness until they came out where I'd expect - at the injector switches. OK. I removed the push connectors from the switches, snipped the other end and pulled the wires out. There was a couple of feet of wire in each of which there were two heat shrinked solder joints and the one crimped butt splice. The joints were fine, but there was a lot of redundant wire here. I replaced all this with one wire with a small loop at the end and a crimped spade connector for each switch. From the other poles of the switches came the source ground wires. I traced these back into the main harness. They seemed to be going all the way across the panel. Where were they going? Eventually they came out at the ignition switch, were joined in another crimped butt spice and hooked to the same terminal as another wire labeled EC2 case ground. The center of the ignition switch is fed a direct ground from the EC2 for coil disable, so reasonably, I'd taken the ground from there. I extracted these wires and replaced them with a short loop with spade connectors to each switch and a single wire direct to the EM2 case which also has a direct ground to the EC2. At the PCM connector I soldered the wire from pin 13 to the one from the injector switches and one of the new shielded wires back to the EC2. Doning my trusty Harbor Freight lighted headset magnifing glass I examined the join carefully, gave it a good pull then heat shrinked it. Back at the EC2 I traced the other end of the wire, snipped the wire from pin 30 and made another heat shrinked solder joint. Finally I hooked the test meter between pin 30 on the EC2 connector and the EC2 ground stud. 0.007 ohms. I reached forward and turned both injector switches on. Open line. Switch off either injector, or flip the PCM cold start switch to on, and I was back to 0.007 ohms. OK. Cool. Interestingly enough, the cold start and injector disable functions were working perfectly, but I've eliminated a bunch of unnecessary crimped connections and replaced the suspect old wiring down the heat duct with new shielded cable and confirmed the function for this part of the EC2 input. All this took 5 hours.

One down, 36 to go. This is going to take a while....

Disappointment

One by one I replaced every connection between the EC2, the EC2 control panel and the EM2. I was 3 wires short with my 15 wires, so another run of three conductor was needed. After a week of soldering, heat-shrinking and testing I finally got everything buttoned up late Saturday afternoon. I switched on the master.... and the EM2 still flashes "nop", just as it did before. There's power at the EC2 and I tested every wire as I connected it, so I'm pretty sure the serial-in and serial-out wires are good. The batteries are a little low, so tomorrow I'll charge them, retest the data connections, then panic. Well, perhaps not panic, but I don't know where to go from here. I noticed that the EC2 manual refer's to "earlier EM2" data connections. Tracy couldn't have changed the way the two units communicate and send me the wrong one, could he? I wonder if he's back from Colorado yet...

Next day I got to the hangar early determined to find the problem. While the batteries charged I changed the plugs. The old plugs seemed clean and dry, but I changed them anyway. Next I confirmed the continuity on the serial in and serial out wires to the EM2. I also checked that there were no shorts at either connector. No continuity beteen the pins and ground or adjacent pins. I scraped some more paint off the canopy for a while to give the batteries a good charge, then wheeled the plane out and tried a start. Nothing. Not a glimmer on either computer. I checked for a spark - no spark on any plugs. I checked for 12v at the coils. Yep.

OK. What's going on here. There's power at pins 33 /34 on the EC2. I even removed the cover and checked for power on the back of the pins. The EC2 ground pin, case and ground stud all test as a good ground. There's no spark or sign of fuel on the plugs after cranking. The EM2 isnt getting communication with the EC2. It's almost as if, dare I say it, the EC2 is dead. I just don't see any other possibilities. Tomorrow I'll recheck all the connections to the panel making sure that there are no cross connecting pins. After that I'm stumped.

Next day I checked all the remaining items on the list which I'd compiled from helpful sugegstions of other builders. I checked the crank angle sensor wiring, the coil wiring and the harness to the control panel. Nothing worked. I sent an email to Tracy asking for any other suggestions he might have, and threatening to tow the plane up to Shady Bend and camp out until we have a solution.

A Tribute to Paul Conner - May he fly forever with excellent temps and strong tailwinds

Monday afternoon, May 16th, while I was busy trying to get my rotary engine running, Paul Conner, who taught me how to fly a canard airplane was airborne in his rotary powered SQ2000 dealing with an unknown problem.

I'm so close to what he was doing that I feel like I know exactly what was in his mind at that second. You do your research, listen with an open mind to all the opinions, then do everything humanly possible to make your airplane perfect. You take every precaution you can reasonably take, but then there comes a point where you just have to kick the tires, light the fires and see if all your theories work in the real world. If you don't, then all your work is for nothing. You know there is risk involved, you minimize it as much as possible, then accept what's left as unavoidable. In doing so you advance the art.

In any flight, in any airplane, there can be a point during take-off where there's just no way out if things go to hell in a handbasket. It's what I call the "Oh sh.t" point. A little higher and faster and you have a chance of dealing with it. A little slower and lower and maybe, if you're good, you can pull it off. Paul must have been at exactly the "oh sh.t" point. No way out. On a previous flight Paul had a vapor lock problem which must have occurred very close to that point. He pulled off a truly amazing piece of flying and got the bird back on the ground undamaged. This time he didn't make it. Having flown with him, the one thing I'm sure of is that, if he couldnt get out of whatever went wrong, no-one could have.

On March 17th. the Mobile Register printed a well written initial report on the accident. Since the aircraft was destroyed we may never know precisely what bit Paul on Monday afternoon, but efforts are already underway to find out. Speculation is worse than useless. If we find any facts, I'll report them here.

Paul's contribution to experimental builders and pilots world-wide will live on. He proved that an SQ2000 can be powered effectively with a normally aspirated rotary. He proved, beyond a doubt that the engine can be cooled effectively with a NACA scoop. He proved the Microtech ECU and a number of other ancilliary systems. In all of this he contributed a great deal to the safety of those who follow. He contributed tremendously to my own safety with his wise teachings which I have tried to pass on in Chapter 28.

Paul Conner was a highly skilled and experienced pilot, a superb instructor and a very concientious and able builder. Beyond all that, he was Gentleman. I feel privileged to have known him and to have flown with him. He will be greatly missed.

Moving on

After a couple of days moping and thinking about Paul, I had a brainwave. Why not take my EC2 to Bulent's hangar and install it in his airplane. Then I'll know for sure if it's the wiring or the computer. Friday afternoon, 5/20 I headed down to Ft. Lauderdale. We connected my EC2 to Buly's harness, made sure it was properly grounded, then turned on the master. Buly's EM2 immediately flashed NOP. We tried a start anyway, but got no sign of ignition. Next we reinstalled Buly's EC2. When we turned on the master the NOP message didn't flash, and the engine started on first crank. So. There we have it. My brand new replacement EC2 is dead, and has to go back to Tracy once again.

The next question is Why is it dead?Of course, we could have proved my wiring by bringing Buly's EC2 back to my hangar and trying it out. He didn't offer, so I didn't ask. I'm pretty confident that Tracy would never ship a dead unit, so my suspition is that the old wiring had a short in it somewhere which was enough to fry the EC2 in the few seconds I had it installed and powered up. But, then again, the previous EC2 was hooked to that wiring for quite a while, displayed the exact same symptoms, but was declared it to be in perfect condition when returned, so I must be dealing with something very intermittent. Hopefully whatever the wiring problem was, it's now been corrected, the repaired EC2 will work perfectly, and I'll be able to fly this airplane again. I expect it'll take a week to get the EC2 repaired, so I'll be stripping paint and catching up on some other jobs for a while while I wait. I've had a few suggestions, such as checking for voltage on EC2 ground returns that I'll also be looking into in the meantime.

On a side note - Bulent's plane is nearing completion. He just needs to add a few instruments, finish a few minor items on the engine then bolt everything together and get it inspected. Bulent's engine installation is very similar to mine - just more shiney. His is a 3rd gen Cosmo engine with a rebuilt stock turbo. He's just finished an augmenter similar to mine but made out of fiberglass with a heat-proof wrap. It doesnt butt up to the turbo heat shield which means it'll takes its air from the cowl rather than from around the turbo. Buly's intake is completely different. He has a large P51 scoop under the lower cowl which feeds his radiator at an angle. The stock 2nd gen oil cooler gets its air from a small scoop inside the main scoop, and the intake (now with K&N filter) bleeds off the main plenum to a secondary plenum containing the intercooler. This way his alternate air source is cowl air via the intercooler. Very clever. The intercooler itself is at an angle in the starboard wing root. He did a very nice job with his over-the-top of the engine intake system which is ceramic coated in reflective chrome, as is the turbo heat shield. Very pretty. Almost a pity to put the cowl over it, even if it does have camlocks. The other very noticable thing about Buly's plane is the back seat, which is tiny, and the front seat which is supersized. It isnt any wider, but it LOOKS wider because of the depth and the front opening canopy. The panel is nicely done in a matt silver but, like mine, it's a bit lacking in avionics. Just a com radio, transponder and three of the basic six flight gauges so far. There's a neat row of 17 small breakers just above the passenger leg space that I'm almost tempted to copy. Unlike my plane, Buly's paintwork is perrrrrfect right down to his massive home-built three blade prop. Also, unlike mine, Buly's engine RUNS when you turn the key.

Christmas in May

On Sunday, 5/22, I saw a post on the Canard forum. Howardt offered to buy my old gauges, which I'd offered for sale at $200, and fly down from Tampa to collect them. He also mentioned that he was experienced in debugging electronics problems AND had an EC2 he was willing to "hazard" while testing my harness. My answer was simply "There IS a Santa Claus!" I contacted Howard and we arranged for him to fly down the next day.

While waiting for Howard to arrive I decided to test something which had been bothering me. I have been suspicious of my coil disable circuit. I discussed this plan with Tracy by phone many moons ago when I was installing it. Perhaps he didn't follow what I was planning to do. After testing that there were no unexpected cross connections in the switch I connected the center pin on my standard Left, Right, Both ignition switch to the EC2 ground, then connected the coil disable pins on the EC2 to the left and right mag terminals. This way "off" would disable both coils, Left would disable leading, etc. The setup has been like this from day one. I disable the coils during my runup as you would check mags in a Cessna. Two years after wiring it this way I put a volt meter between the center pin of the ignition switch and ground, and cranked the engine. Would you believe 1.5 volts? Second time I tried it the reading was just a few millivolts. Next time it was 1.5 volts again, then I couldn't duplicate it. 1.5 volts down the coil disable ground wire would be a bad thing - right?

When Howard arrived the first thing he produced, without any prompting, was a schematic for the standard ignition switch produced by Bob Nuckolls which carries, in very small print, a warning not to connect the center ground pin to any other ground in the aircraft. Oops. I have the same schematic in my folder of circuit diagrams, with the same warning. I guess I missed it. Apparantly there's a possibility of leaked voltage from the starter solenoid switching function. Don't ask me why. We tested with the computer logging equipment but couldn't repeat the problem. In any case, I have separate switching for the leading and trailing coils and could disable a coil set during runup by switching the power. The coil disable function is therefore redundant. Howard also handed me a copy of Bob's notes on grounds which I'd read previously but perhaps not taken enough notice of. Finally, the last present out of Santa's sack was a load testing gizmo as described in the above link.

Howard has a relatively simple and inexpensive computer setup which made me think of the line from Independance Day - "I've got to get me one of these". It's a $50 multimeter with a serial port. It comes with software and, when hooked to a laptop, will show you a graph of what the multimeter was reading during an operation like cranking the engine. We hooked the multimeter to the starter bolt and to the battery ground terminal then cranked the engine. I don't have the detailed data, but the graph showed that voltage had flowed backwards through the ground path. To be precise, current flowed backwards. Voltage was negative. Apparantly that's not good.

Ground Bounce

I'd always thought that ground bounce was what you got on a heavy landing. Not to electronics guys. The way I understand it, this electricity is slippery stuff. When you send a bunch of electrons down a wire, like the starter cable, they're really not interested in turning the starter. All they want to do is get home to the battery. Like pigeons they always get home, but you're never quite sure which route they took. Also like pigeons, they have a nasty tendancy of leaving a trail of crap along the way. If the path you planned for them isn't easy enough, or there are too many of them and that path gets crowded, a bunch of them will be perfectly happy to jump from one wire to another, go backwards through you're delicate gizmos letting the magic smoke out for fun as they go, or whistle up and down wires that were intended for other things giving birth to more electrons in places they're not supposed to be. Something else I learned was that there's way too much wiring in a plane to test and correct everything in one day.

We found significant ground bounce in the starter ground circuit, so I removed the engine ground strap and cleaned the connections. The coil ground was also unacceptibly weak. Howard made an interesting point here. I went to great lengths to ensure redundancy. Leading and Trailing coils have their own circuits, fuses, switches and coils, and there's a backup computer too. However, according to the wiring diagram, all four coils share one ground wire to the engine block. Duh! My single 16g ground wire had a solder connection to each coil, then a crimped ring on the engine block. Howard suggested that we remake this with an 18g wire from each coil going into two crimped rings. When the crimps were done he showed me how remove the plastic, solder the wire into the crimped fitting, then double shrink wrap everything. My coil ground circuit improved dramatically when we did this, and it's a lot more durable and fault resistant.

Ground Loops, and How to Fry Sensitive Electronics

I'd always thought a ground loop was what happens when you land a tail dragger badly. Not to electronics guys. A ground loop is when the electrons take another, unexpected path home to the battery. Howard and I spent about nine hours in the hot hangar working on the plane wiring. Some of it went over my head, but I learned a lot during that 9 hours. I made notes and came away with a number of suggestions to implement. Stuff like moving sensor wires away from the starter cable and the coil leads. I also have a lot more circuits and connections to check with the load test gizmo. Eventually, it was 9pm. We'd corrected quite a few weak grounds and Howard thought it was time to try hooking up his EC2. It's an older version, so it wont talk to the EM2, but it should tell us if the circuitry is ok.

I climbed in the back and started to hook up the EC2. At this point Howard walked around the plane and asked me to stop. I insist that you connect the ground first, he said. OK. The master was still off, but still, that makes sense. I pulled the ground wires together and hooked them up to the EC2 ground stud, then plugged in the connector. We plugged in the PCM at the panel and turned on the master. With the computer hooked to the primary injector circuit we cranked the engine looking for the pulse width voltage. Nothing. That was when I noticed that the screen on the EM2 was blank. To cut a long story short, here's what happened. It was a little dark in the back of the plane. It was hot and sweat was dripping in my eyes. When fishing up the dangling ground connectors I'd missed one. The one I missed just happened to be the main ground for the EC2. I'd invertantly created a massive ground loop. When we powered the EC2 the electrons had no way home. Annoyed, they quickly fried something in the EC2, then headed off down the serial cables and fried the engine monitor as well. Disappointed, to say the least, I collected all my old gauges in a bag and gave them to Howard as thanks for his trouble.

A Sleepless Night

It was too late for Howard to fly back, so we got some steaks and crashed. At 3 am I was up again. I'd hoped to end the day with a solution. Instead I'd fried the EM2 once again. I was not happy. After some pacing I checked my email. Tracy had found the problem with my EC2 - a burned out choke filter which had been subject to voltage overload. That's great, I thought. Now all I have to do is head down to fedEx in the morning, wait a couple of weeks, and I'll be back where I was 6 weeks ago. It was then that I had a brainwave. Howard's flying back to Tampa in the morning. All I have to do is rip the EM2 out, jump aboard the 172 and I'll be in Tampa by 9 am. I'd rent a car, drive to Tracy's, watch him fix the EM2, drive back to Tampa and pickup a SouthWest flight into West Palm. I'd be back by 6pm with everything fixed. What a plan! I bought the SW ticket on-line, then waited for Howard to wake up.

So there I was, a few hours later, driving up route 75 past Ocala on 2 hours sleep with 100 miles still to drive thinking What the hell am I doing?"I hadn't called ahead, I just set off. Now I'm wondering what I'm going to say to Tracy when I get there, and I'm practicing my most pitiful "help me" expression in the mirror. I didn't even know if he was going to be there. This is crazy.

I eventually arrived on his doorstep clutching the remains of my EM2. He greeted me with a puzzeled look and a half smile. I said Have you ever had a brilliant idea in the middle of the night, that got worse and worse as the day goes on?" Good natured soul that he is, Tracy laughed, took my EM2 and headed off to his workshop while I hit the head. By the time I came out he'd found the problem. The main CPU was dead. He'd already swapped it for another one and was busy testing. I watched as he tested the functions on my EC2 and EM2 on the ocilloscope one by one. It's neat to see the square wave injector pulse visually, then watch it shrink and grow as you invoke the cold start and mixture controls. In a couple of hours everything was fixed, tested and packed in the box.

I took the opportunity to pick Tracy's brains about the issues I've been having. This helped me put a lot of what Howard had been saying in context. We talked about how my system is setup. Do it top-down, Tracy said. Start with the heavy amp circuits like starter, coils and strobes, then work down to the smaller stuff. Try to eliminate long runs. For example, he had me talk through my coil wiring. It starts at the battery, goes to the E-bus, through a few connections and a fuse, then heads to the switch on the panel. After the switch it heads to the back, through the firewall to the coils. From the coils it goes to the ground strap on the block, back to the firewall ground and on to the battery. This is duplicated for two sets of coils. How could we simplify that? Why not take power direct from the battery contactor about 18 inches, direct to the coils. This would cut out 60 feet of wire and a gazillion joints. You want the coils on when the master's on anyway. That's why we have a disable function, Tracy explained. Use the disable option, but power the coils with short wires and avoid all the ground loops. Brilliant. Something similar could be done with the injector circuits. Just as Howard had suggested, I could power them with relays at the firewall, just like the one inside the EC2, and eliminate another 60 feet of ground-loop prone wire. I thanked Tracy profusely for dropping whatever he was doing and fixing my stuff, and promised to try not to fry the stuff again, at least not both at once. Laura gave me an ice cream while Tracy laughed again and said Just try to fry the EM2 every other time instead of everytime. What wonderful people! Wanting to take up as little of their time as possible, I thanked them again, headed back toward Tampa, caught the last flight out and was home by 11pm with all my gizmos fixed.

On the way back, I did some thinking. Three years ago, when I started planning the wiring for the plane, I decided that this stuff was important and I was going to do it right. I got Nuckoll's AeroElectric book, bought all the recommended tools and materials from his web site and followed his teachings to the letter. The wiring worked fine two years ago when I finished it. Since then it's been subjected to South Florida humidity. Little by little all my joints started gaining oxidation and resistance, and maybe a few things worked a bit loose with the vibration. While Bob's theories make sense, and he obviously knows a lot about electrical stuff, they didn't work for me. I learned more from Howard in one day than I did from Bob's book. Most of what Howard and Tracy had to say lined up with Greg Richter's paper on aircraft wiring. There's lots of knowledge in Bob's book, but there are not-so-good ideas in there too. For example, I wish I'd used breakers instead of fuses. I may change these over later. The general rules seem to be, Keep it simple. Solder, crimp and shrink-wrap everything. Use RTV inside the heat-shrink to make it totally airtight. Never use a crimped butt-splice. Avoid long runs. Test all circuits under load. Keep the power wires seperated. I could go on, but Greg's paper says it all.

A word About Urgency

Monday night, after letting the magic smoke out of $1900 worth of electronics, I was pretty frustrated and upset. I couldn't just sit and wait. I had to DO something. The trip up to Tracy's wasn't about urgency, it was about frustration. I've tried injecting urgency into the process of airplane building at various stages, and it doesnt seem to work. The project seems to have a timeline of its own, and rushing it doesn't help. On the other hand, you do have to give it some level of priority or it'll never get done. When I got back with the fixed gizmos you'd think I'd be off to the hangar first thing next morning. Nope. I'd just invested two days and had to catch up on other things. I'm hoping to get started on the plane again over the weekend, but before I do, I'll be picking up a decent soldering iron and one of those serial interface multimeter things. I have a lot of rewiring, resoldering, and testing to do. I'd like to get the little buggers a bit more disiplined and under control before I let electrons loose inside my gizmos again.

A little input helps

The day after posting the above I got two emails. Chrissi mentioned that RTV contains acetic acid, and I'd be better with the heatshrink from Digikey that comes with a hot glue coating inside. Scott (long-ez builder) gave me a wake up call. Tie the coils to the contactor, and now you're a glider if the contactor fails. Duh! I'd like to think I'd have realized that before implementing the idea. That was yet another of those brilliant ideas that doesn't hold up in the morning. I'm beginning to think that the bottom line here is weak connections. The Nuckolls Z-13 schematic I have works, and has lots of good redundancy built-in. I just need to implement it properly so that the electrons don't get a chance to stray off the intended path. There's no need to reinvent this wheel. Howard may well have already found the main culprit in that badly grounded start circuit. There were a lot of stray amps there. If I eliminate all crimped butt-splices, solder and heat-shrink all connections ensuring that they're airtight, and keep high current wires away from other wires, I'll probably be in good shape. If that doesnt do it, then I'll put relay's over the spar to control coils, injectors etc.

Now, if I could just find some time to get down to the hangar (when it's not over 100F inside) and spend some time there, I could make some progress. Meanwhile, Char's on vacation and wants to go out & play (together), the gutter on the side of the house needs replaced, the pool needs resurfacing (anyone have a large supply of Diamond-Brite they don't need?), the patio floor is getting a skim of concrete and epoxy paint to cover up all the MGS drips, and I have a client that wants some work done over the weekend....

Perfect Weather

There was a time when perfect weather meant clear skies and light winds down the runway. These days perfect weather is rain, and lots of it. Without rain the temps in the hangar are unbearable after 11 am or so. Each day over memorial weekend I got up early and did 3 or 4 hours in the hangar before it got too hot. While I'd like to get this work finished and go fly, I'm determined to go slow and do it right. Once the hangar hits 90F, I go home to cool off and come at it fresh again in the morning.

I started with the EM2 and some of the wiring in the front. Every time I came a cross a butt splice I replaced it with a soldered joint added a blob of silicone insulator gel before heat shrinking. All crimped fast-ons and ring terminals were recrimped, soldered and double shrinked. There were 6 ground wires going to the EM2 ground stud. One for the display, one legitimate ground to the buss, one pigtail that went nowhere, and three that had no right to be there. As I'd added a few extra's, like the GPS and fuel crossover warning light, I'd used the EM2 ground stud as a convienient place to get a ground. I ran these direct to the panel ground, and replaced the EM2 ground with it's own personal crimped, soldered and heat-shrinked ground. Before connecting the EM2 I tested the ground resistance with my new under-load testing kit to ensure that the joins were good. Finally I powered up the EM2 and was relieved to see the screen fill with data. I need to reenter all the calibration data which, thankfully, I have written down.

At the back of the plane I tested the coil power leads, and found one with more than double the resistence of the other three. I traced the wires back and came to more butt splices where the pigtails were joined to the EC2 cables. I removed them all, redid the joins and tested. Resistance on all four was now lower than the lowest previous reading. While in the back I replaced the short ground lead with a new one and removed a bunch of other butt splices from sensor wires. After three days at this I'm making good progress. I still have a bunch of stuff to do at the back before reconnecting the EC2. I'm taking my time and trying to implement as much of what Howard and Tracy recommended as practicable without ripping all the wiring out and starting from scratch.

Going no-where Slowly

Finally, after about 20 hours of rewiring, soldering and testing, I got ready to hook up the EC2. I rechecked every pin on the connector for continuity to the panel module and for cross connection to other pins. Next I turned on the master with the PCM hooked up, but the EC2 connector off, and checked for voltages that shouldn't be there. I found 12v on pins 33 and 34, where it should be for power input, and on the four injector pins. That made sense, but there was also 4.9v on the serial out pin and 1.9v on the serial in. I called Tracy to check. He said this was to be expected, so I connected the EC2, checked the ground one more time, and hit the master.

NOP

Essentially I'm at exactly the same point I was at 3 months ago. Either there's a problem with the wiring or a problem with the EC2. There's no (easy) way to tell if the EC2 is operational except by running the engine or by seeing the NOP message on the EM2 go away. The message was still there, so I tried the other way. She cranked fast, but made no attempt to fire. When I took the plugs out they looked wet, and cranking with the plugs out produced a cloud of vapor. I can't be sure, but looks like the injectors are firing, and the plugs are not. The symptoms are identical to what I had the last two or three times the EC2 came back from repair. I'm beginning to think that the wiring was ok from the get go, and something else is wrong here. Sure, I corrected a few imperfections, but there was no smoking gun.

Looking back through my own log I see that the EC2 was working fine, except for occasionally scrambling its A chip data. B was running perfectly. I sent the EC2 back, Tracy made some changes and upgrades, then it didnt work at all when I plugged it back in. This is when the trouble started. When he changed the B chip back to the original it worked on B. Next time it didnt work at all. He tested it, said it was fine, but it didnt work on return. It didnt work on Buly's plane either, so I sent it back and he found a burned component. I watched him fix it and test it, but it still doesnt work in my plane. Could there be something different between the schematic in the manual and the test bench at Tracy's? For example, lets say that power on pin 34 fries the board, and Tracy's bench only puts power on pin 33. Buly and I have power on pins 33 and 34 per the manual. That would explain it. Not at all likely, but it WOULD explain it. I'm wondering if its something along these lines, but more complex. Whatever it is, it's hard to know where to go from here. Well, actually it's not. The next step is obvious. One thought I've had is that the main difference between Tracy's test bench and my installation is the length of the cables. Maybe I should make a 4 foot harness from scratch and see if that works.

Unless someone comes up with a better idea, I have one test remaining. In the morning I'm heading down to Ft. Lauderdale to test my EC2 in Buly's plane again. If it doesnt work there either, then it goes back to Tracy for checking. If he finds something wrong, I don't know what caused it or how to correct the wiring so it doesnt happen again. If he finds nothing wrong I'm equally stumped. The other alternative, of course, is that it works in Buly's plane. I still don't know where to go from there. My logic tree has three possibilities left, and I have no idea what action to take whichever one we end up at. I'll take a couple of asprin. Perhaps I'll feel better in the morning.

A Frog, or a Prince?

Saturday morning I did the drive down to FEX thinking how much quicker the trip had been in the Cozy. Buly installed my EC2 in place of his own, and the NOP flashed. So, it looks as though I've fried the computer once again (but read on).

Buly's getting close now. Mounting the wings and finalizing the panel. By the way, his longerons are 43 inches inside to inside at the shoulder, and its 47 inches from the inside fuse side to the other at the seatback. That's about 5 inches wider than my plans Cozy. His bird is starting to look beautiful as it finally comes together. Still....I feel sorry for Buly.

[RANT]
Buly's plane is like a magnet, and his hangar has become the hang-out place for all the old and bold pilots, experts and self-proclaimed gurus of aviation. It's like kids outside the 7-Eleven except that the crowd is a lot older and slightly better behaved. He has a yellow "keep out" rope across his hangar door, but it doesnt work. I counted eight people just hanging out and BS'ing the day away, leaning on cars and setting off the alarms (really). He has to navigate around them just to get a tool off the bench. I have a few occasional visitors like this, but nothing like what Buly has to tolerate. What are these people thinking? Don't they realize that we're not here to shoot the breeze and socialize. We're busy, trying to get things done, and are concentrating on fine details upon which our lives will ultimately depend. At best these "hangar rats", or, as I call them, "Velcro Men", are an annoying distraction living out their dreams through someone else's action. Don't read me wrong here - people are welcome to visit, look over the plane and ask questions. Once. Maybe twice or three times as they and I progress over months and years. If the visitor is actually building, or planning to build, or has a specific question, I'll be happy to stop what I'm doing and discuss the details for as long as it takes. Dropping in, or calling every day, even twice a day, (when I'm in the back of the plane and the cell phone's just out of reach on the wing), just to say "Hi" and catch up on my progress drives me crazy. That's what this web site is for. For the FEX crowd, who probably don't read this, Buly has a web site too. Go read it every afternoon, and let the guy get on with what he's doing. When it comes to the point that he's taxi testing and preparing for first flight have the courtesy to stay TF away and let him THINK.
Of course, this is just my opinion. I could be wrong.
[/RANT]

Eureka

I did a RAA (Rotary Aviation Association) inspection on Buly's installation and pointed out a couple of minor things. At this point, no-one knows all the failure modes better than me. In general his turbo Cosmo 13B is very nicely done. It's certainly pretty with chrome and ceramic coating all over the place. Following Paul's fuel issues I'm a little unsure about his duplex valve fuel set-up with all those long 3/8 pipes, and I wouldn't have put my fuel pumps in the engine compartment, even if they are low on the cold side. I wonder how well his P51 scoop and double plenum will get air to the oil cooler. We'll know soon. Overall its a very, very pretty airplane. After a pleasant, if disappointing, visit I headed back to my own hangar. It was rent day, and before sending the EC2 back I wanted to check whether it was really dead. Maybe it was just a communication problem with EC2, I thought. Wishful thinking, probably, but I'll get someone to crank it while I hold a plug, just to be sure.

I reinstalled the EC2, turned on the master and.... the NOP wasnt flashing. Huh? I pressed the button combination to bring up the EC2 data. The data came up. OK. This is wonderful, but I'm not about to put the cowl on and go fly till I find out what changed and why. (sorry. That wasn't supposed to rhyme). The only thing that had moved between last night and this morning was the 37 pin EC2 connector. With the engine monitor happily communicating with the EC2 I climbed in the back and wiggled the big harness going into the EC2 connector. The NOP began flashing. I wiggled it again, and it stopped flashing. Bingo. There's something loose inside the connector, or in the harness. I'm guessing that a wire has broken at the end of a solder joint INSIDE the insulation. When I bend the connector outward to test it the wire makes contact. When I bend it back to plug it in, the contact is lost. Geesh. This could well have been the cause of my EC2 troubles for the past 12 months. An intermittent contact could easily have caused the scrambled EC2 data I had way back when. The funny part about all this is that I didn't make the connectors. I paid for Tracy to do this. I've learned enough since that I could easily do the job myself now. Totally exstatic that I now know what needs fixing, I headed home for some food and a rest before digging in and attacking the problem.

That was when the phone rang. Char said "Could you come home? I need some help". She sounded a bit freaked out. What's up? Another snake? She said I'll find out when I get home. Hmmm. What manner of challenge am I faced with now? When I got home, Char led me to the bathroom. There was a large frog in the toilet bowl. There's only one way he could have got there - up the pipe. I fished him out and asked Char if she'd like to kiss him, just in case he's really a prince. She declined, saying she doesn't need a prince - she already has one. Good answer. I let him go in the backyard while Char gratefully used the bathroom.... at last.

So later today, or tomorrow, I should be able to get the EC2 install fixed and start making some progress. Yes, I am going to load test EVERY cable from the EC2 connector when I'm done.

Better Late than Never

While I was busy discovering my broken wire our new hangar tenant, Mike, arrived. He's bringing his Velocity down on Thursday. Turns out that he's an electronics engineer. I asked if he has an oscilloscope. His look said - "Does a bear sh.t in the woods?" He said simply yes. I'll leave one at the hangar and show you how to use it. Timing is everything, isn't it?

Makin Hole!

I sometimes wonder if it would be possible to have a complete conversation using nothing but old movie lines. Anyway, back to the matter at hand, On Sunday I checked the connections to the board with a magnifying glass and tested between the board and the pins. No problems there. The wires all tested out and I couldn't reproduce the NOP communication error no matter how much I pulled at the wires or wiggled the harness. There's nothing worse than a problem that goes away without explanation except, perhaps, a problem that wont go away.

Finally, I think I've found the problem. The EC2 case is cut out to permit the connector shell to pass through and seat properly. The cut is VERY close. It's possible (if you're me, perhaps even Buly) to have the frame of the connector "miss the hole" and stop at the case. The screws tighten up nicely, but the connector isn't fully seated. Wiggle it a bit and it gets past the edge of the EC2 case and the screws are loose again. Tighten the screws a few more turns and the connector goes in another 1/16 inch or so - just enough to make the pins contact.

I think the ECu was live all the time, just not communicating with the EM2. The engine wouldn't run because the program defaults weren't loaded into the computer. The EM2 showed an equal row of 50% bars for mixture for all MAP readings. I recovered Tracy's own engine settings which are built into the ROM. Now the MAP table was a nice even curve. I turned on the pumps, injectors and coils, shouted clear, and cranked. Vrrrrroooommm. The engine started up right away and ran fairly well on both computers. After almost 12 weeks of silence, I think we have a winner. It needs some fine tuning, but that was enough for inside the hangar. I shut down and went home for breakfast.

So, I think my EC2 problem is solved. Not so, apparantly, the frog in the toilet problem. It seems that the prospect of his reappearance, or worse - we have alligators in Florida - is seriously spoiling my beloved spouse's most relaxed moments. Must be a female thing. We males can keep an eye out for such things and, in the rare moments when we're, shall we say, blind -sided, its not really a problem. After 6 years in experimental aviation I half expect to be bitten in the ass at every turn anyway. Ah well. Better look into it. To cut a long story short, I'd moved the hot tub to skim the patio floor, and thereby displaced the cesspool cover , thus allowing Mr. Frog easy access to the forbidden territory. I replaced the cover, and all is now well with the world.

An end to the Rains - please

Monday was heavy rain and thunderstorms. I guess I asked for it. On Tuesday it let off for a bit in the afternoon, so I went down to the hangar and prepared the plane for some taxi-testing. After a half hour of cleaning up and moving tools, I pushed her back and got ready to start-up. Al dropped in to pat his LongEz. He's heading up North with Mike to collect Mike's Velocity hoping for good weather since the plane's not IFR certified. I'm hoping for better weather too.

Instant start. I taxied out (sans cowl) and opened her up a bit on the taxiway. Kitty wants to fly. The engine was smooth and consistant at all rpm. There were a couple of "pops" from the engine as I accelerated at about 4000 rpm. They sounded like a small backfire sort of noise. Later Al said I blew out a bit of smoke at start-up, but then it cleaned up and sounded good. Perhaps a tuning issue. Maybe the engine was just getting warmed up. I'll keep an ear out for further "pops" in future tests. I did a quick runup, and she immediately went to 4070 which is a little higher than I'm used to. After a couple more runs down the taxiway and another runup to the same rpm coolant was climbing over 200 and it was starting to rain (again), so I put her back in the hangar. My analog fuel pressure was reading zero, but everything else seemed normal. After shutdown I tapped the fuel pressure gauge, and the needle jumped into life. Hmmm. Around the back, everything seemed suprisingly cool and, for the first time ever, there was absolutely no sign of soot on the prop. In all, an excellent test run.

Anyone in World Climate Control reading this - thanks for the rain - I'm all done now, so you can turn it down a little. The weather channel is painting Florida solid green for the rest of the week, so Al may get stuck in the Carolina's and I have to find something else to do. I'm not in a rush to fly her again, but it would be nice to have taxiing weather at least. I need to do quite a few more high speed taxi tests, reenter all the calibration data in the EM2, add some new fuel, put the autotune through it's paces and generally give the plane a shake down because it's been sitting so long. As a start I re-installed the spinner before going home ...in the rain.

Next day, Wednesday, it was great flying weather. Thanks to the WCC guys, but I'm not ready yet. I did some more taxi tests and run-ups. This time I fired up the Autotune feature on the EM2. Despite Tracy's brief instructions I wasn't quite sure what to do. I started out at about 20 MAP and let it do it's thing. The mixture seemed generally high, so I lowered the pulse rate a couple of notches to get it in the ball park, then tried the Autotune again. Now the mixture readings were closer to the center and Autotune seemed to be adjusting things as I moved slowly through the RPM range. It seemed to have problems at low MAP settings. It's working on a specific MAP number, but the MAP starts to surge if the mixture gets too lean, so I think it gets lost. Without quite knowing if the Autotune was doing a good job or not I left it on and did a runup and a high speed taxi test. I still seemed to be getting some surging at idle, but otherwise the engine ran very well. After 15 - 20 minutes the coolant temp was up to 220, so I put her back in the hangar to cool off. As before - no soot, and everything at the back looked good. That is - no leaks, nothing missing and no smoke. Don't laugh - I've had all three at once a couple of times.

While the engine cooled off I did some more Top Gloss scraping. It's soooo much fun scratching off the beautiful finish you worked about a year to get. I haven't touched the canard yet, figuring that would cost me some lift so I'll do it last. After a couple of hours of scraping the engine had cooled down, so I pushed her out again. This time I did a pre-flight as I gradually get myself back into the groove for flying. I taxied out to the runway, but my headset is at home so I had to make do with the taxiway. The nose was starting to lift at 60kts when I backed off. No pops or backfires, and no hesitation. Despite the surging while the autotune was running earlier, the idle seems fine at about 1100. I can still richen it up a little with the manual control if it gets too lean. I've probably put an hour on the engine since getting the EC2 fixed. So far no problems, so I think tomorrow it gets the cowling and some fresh fuel. Getting closer...

No. Not yet

Next day I added 20 gallons of 93 octane gas, and 20 oz of 2-stroke to the right tank. I thought about putting the cowl on but the weather was 600 scattered with a 16/20 kt wind, forecasting solid rain from a tropical storm over the next few days. After Ed Anderson's incident with a blown apex seal I've been meaning to install an air filter. Now seemed like a good time. I cut a triangular piece of glass out of the side of the plenum, then picked up a K&N air filter at the autostore. They don't have triangular ones, so I had to modify one to suit. Strangely, they make a 4" * 6" filter that's $54, and a 12" * 16" of the same material that's only $34. The little one is probably for a BMW. I got the big Chevy one, cut the edge strip off, cut out the triangle shape I needed, then reattached the edge with RTV. The result is a perfect sized filter for my installation, and there's enough material left for two more filters.

I cut the intake pipe short and shaped a plug out of blue foam to transition from the filter to the intake pipe. I need a good fit around the edge of the filter and a lip to seal against the side of the plenem, so this seems like a good time to try the Cozy Girrls low-vac technique. I used modelling clay to form a smooth join between the foam plug and the air filter, then carried the clay on over the entire foam plug. I'll glass it tomorrow (in the rain). I did some more paint scraping to round off the day.

While I was working on the air filter modifications, Al and Mike arrived back from New York. They'd done the trip in just under 3 hours at an average speed of 310 kts. Unfortunately the Velocity they went to get couldn't come. It's still 1000 miles away, waiting for better weather. Ah well.

On returning home I learned that, when investigating my earlier EC2 failure, Tracy discovered an incorrect component in the more recent EC2's. Apparantly the choke filter his manufacturers have been using is the wrong spec and is much more susceptible to stray currents than he intended. He's issued an AD recalling all EC2's with that part and suggesting the grounding of all aircraft using them. Interesting. The thing shouldn't have been subjected to whatever blew it anyway, but it looks like I found a weakness that should not have been there. I'm happy to have been of service.

Up to my armpits

Monday 6/13/05 - The next few days were IFR or marginal VFR, so I wasn't too worried that the air filter took a bit longer than intended. The low-vac worked out pretty well. I couldn't get it to seal at first, then I rolled the pipe around the cling-film and off she went. In a few seconds the mold looked like vacuum packed food. As with the food, there were a few minor wrinkles around the sides. I think you're supposed to lay-up the BID in the same direction to avoid this, but I forgot. After cure the part was much better than I could have done any other way. The fiberglass conformed to the shape of the filter edge perfectly. Very nice. Unfortunately I had to "adjust" the shape a bit to get a good fit inside the cowl. I fixed the new filter holder in place against the plenum and attached the intake pipe with 2 BID. Once this cured I added three 4 BID attach tabs, drilled through the plenum and floxed and BID taped three nutplates to hold the attach screws. This is where the heading comes in. The weather this evening was beautiful VFR and a cool 82. I was tempted to leave the air filter off, install the cowl and launch into the sunset, but discretion got the better of me. I really didn't want to install the cowl only to have to take it off again the next day, and I wanted this job out of the way....the nasty bit was still to be done.

So, instead of cruising amoung puffy white clouds picture me, if you will, crouched under the plane with both arms inside the NACA scoop trying to flox a nutplate in place at the very back of the plenum. Sorry. No picture. I didn't have a spare arm. Yes, I did put a spot of grease in the nutplate threads. No. I don't want to do that job again.

So, tomorrow the air filter will be finished, except that I haven't figured out an alternate air source yet. I need some sort of flapper that'll open to inside the cowl if the turbo can't get enough air from through the filter. Once I have the flapper arrangement done I'd like to try one more high speed run with the cowl off to check that the air filter is good, then the cowl can go back on and I can start thinking about flying this bird. Then again, an email from Tracy advises of new EC2 chips with revised code for turbos which he'd "like all turbo users to have". I sent the $20 upgrade fee, so my new chips should arrive any day now. But - the EC2 is working well on both A and B. Do I really want to mess with it? The chips better come soon. I won't be able to resist this nice weather for more than a couple of days. I suppose I could always distract myself with work....

Picking up the Pieces

In other news, Bob Tilley (a rotary aviation enthusiast in Georgia) has arranged for purchase of the salvage of Paul Conner's airplane so we wanklers can do the investigation the NTSB probably won't, and maybe get an idea what really went wrong. I get the impression that Bob paid well over the odds deliberately in order to help Paul's wife and family since there was no insurance to cover the loss. Bob is accepting donations, either toward the salvage cost, or direct to Carol Conner. He can be reached via
his post in the Canard Forum or via paypal to his email - btilley at mchsi dot com. Personally, I'll be sending a check, and will also be visiting Bob to look over the wreckage and discuss whatever findings there may be. I'd also like to have something, anything, from the plane to put in my bird. Paul put a lot of sweat into that plane. It would be nice to think that something from it will live on and continue to tour the skies at 200 mph as he dreamed it would.

Getting Ready to Rumble

Boy. That week went fast. Apart from Monday and Wednesday when I was tied up with work, I did either one or two sessions a day at the hangar. A couple of hours of that heat is about all I can stand in the daytime, so get done what I can, go home to cool off, then come at it again in the evening. I was trying to think of how to install a spring on the alternate air flap when Char suggested that a simple hinge would do the job. The plenum's pressurized - right? That will force the flap shut. If the filter gets blocked there'll be a partial vacuum, and the flap will open. Hmmm. Makes sense to me.

The air filter box needed a bit more adjustment to fit neatly in the space I made for it. I was in the process of making the hinged flap for the alternate air source when Nathan Gifford (proud new owner of Cozy plans) dropped in for a quick visit. His family was vacationing in St. Augustine, so he took a day out and drove 8 hours round trip. Dedication like that tells me that this is one guy who will stay the course and fly his Cozy. (Either that, or he had absolutely no clue how far it was). I was expecting a few hours chatting. I wasn't expecting the Spanish Inquisition. Right. Nobody expects the Spanish Inquisition.

Nathan produced a large box from his trunk and proceeded to set up a complete video recording studio. Professional grade camera with all the trimmings, tripod, lapel mike, the lot. He pinned the lapel mike on me, turned on the camera and started asking questions. About two hours of impromptu, unprepared stand-up, a small layup of a lip on the alternate air source flap, and a complete tour of Kitty later we got some lunch, then wheeled the plane out for a run-up. First Nathan set up behind the plane while I started her up and goosed the throttle. I shut down, he climbed aboard, and we taxied out to the runway, did a runup, then a high speed run until the nose wheel lifted. All this was done with the camera running, and plugged into the intercom for audio. Nathan plans to edit the video and post a small clip for people to see. I'll put a link here when it's available. Eventually we might get it on DVD. What a pity I couldn't take her up. Perhaps a trip to Louisiana is on order later to finish off the video with some in-flight shots.

Finally, on Friday evening, I installed the air filter with it's new duct and took her for a final runup with the cowl off to see if everything held together. I treated this run as though it was going to be a flight. Full preflight, full run-up, belts, canopy locked etc. The first high speed run down the runway was followed immediately I cleared with a very long full power runup. Normally my full power runups are only 15 seconds or so. This time I held it for a couple of minutes to get her good and hot. When the temps hit 200F I taxied back to the hold point did another high speed taxi, then followed it with more full power static running. Temps were sneaking up toward 220 by the time I shut down back at the hangar. There were no hicuups or back fires during the entire run. The engine seems smooth as silk. Just two minor anomolies - the EM2 fuel pressure is reading ridiculously high. Like 150PSI while the analog gauge is showing the normal 35 - 40 PSI. The engine wouldn't run with fuel pressure that high. Either I recorded or reentered the calibration data wrongly, or the sensor is bad. The other thing I noticed was that the coolant pressure went a bit high (20+ PSI) during each take-off run, then reverted back to it's normal 10 - 15 PSI. It does fluctuate with rpm, but I'll keep an eye on this. In general it looks like this puppy, I mean this kitten, is ready to fly. Now if I can just remember how to fly her, we can get the last few hours done. By the time I reinstalled the cowling it was dark, so and I left her to cool off overnight. Depending on the weather, tomorrow may be the day this particular hiatus ends.

Ick Kitty Flies Again - with a Hairball

I'm keeping the letters as stick-on for now. This morning the "S" came off leaving just "ick" and, despite the very ugly paintwork, I took her flying. Apart from the bit when I declared an emergency, it was an uneventful flight.

I got down to the hangar before 8 am to catch the cool (83F) weather. Blue skies and a light wind from the West. Perfect. I had to wait a few minutes for the coolant to get over 135, then I took the active 27 and off we went. As we climbed smoothly through 300 feet I thought of Paul and looked for places ahead that I could put her down. One golf course that'd be pretty messy to get down on. Thankfully the engine didn't miss a beat and I was soon on the downwind at 1000' feeling relaxed and ready for a nice flight. I circled the field at 1000 for a while to check everything out, then called Palm Beach Approach for clearance to climb. The controller asked if I was testing the engine. I told him affirmative and got an open "pilots discretion" clearance. Up to 5000' we went in a nice gentle climb. 32 MAP and 750 fpm. I circled the field at 5000 for over an hour, occasionally taking a detour to fly over the house 5 miles West. Char was working in the yard. She later said she'd heard the engine, but couldn't see me.

After toodling around at 4000 rpm for a while I opened her up a little to maybe 5000 and 36 map showing about 160Kts indicated. Not wanting to test the autotune system in flight yet, I leaned her off a bit to get the mixture bars in the center. Temps were sitting at about 190F, oil pressure was 60. Coolant pressure 10. All was well with the world. I had a large bottle of Pepsi along for the ride and enjoyed sipping at it as I plodded back and forth from the beach to the house and back again. As the hobbs approached 32 hrs I decided I'd like to land and check everything out, then maybe do a longer flight in the evening. I headed back down, decending on the dead side and joined the pattern on the crosswind with a Cessna 172 about to turn base. I was catching him quite quickly so I throttled back a bit and lowered the gear early. It looked like he was going to do his base turn somewhere over the Bahamas, so I stretched my downwind to suit. Then, finally, the 172 made some progress along final and I turned base a couple of miles East of where I'd like to be. I throttled back and the engine cut. Not completely, but enough that I wasn't at all happy about it. It was surging from 900 to 1500 rpm with some really long gaps between the surges. Nudging the throttle didn't help, and, provided it was enough, I didn't want to mess too much with what I had. As I turned toward the runway and set up for best glide I hit the backup fuel pump. No difference. A quick assessment of the situation made me think I could make the field fine, but that 172 would be touching down at the same time as me.

I stopped worrying about the engine for a minute and decided that I needed that Cessna out of the way. Right Now! For the first time in my life I called, Mayday, Mayday, Mayday, Experimental Cozy 96PM on final for 27 at Lantana with engine problems. Cessna on final - Please go around." He acknowledged immediately and started a climb to the right leaving me clear access to the runway. 400'. I had the field and runway 33 made for sure. The active runway - probably. I turned my attention back to the engine which was still "hunting" like a hungry cat. It would come up to maybe 1600 rpm for a second, then die to what seemed like nothing, stay silent for a couple of seconds, then rev up again. With a little time to spare now, I looked around. Mixture seemed like the obvious thing to try next (Duh!). I cranked the mixture knob to the right from it's 10 'o' clock position and the engine picked up and ran smoothly again. Phew. That took a lot longer to explain than to do. I was probably running overlean for 10 seconds. Maybe 12. I advised my 172 friend that the engine problem was solved, thanked him for his help and apologized for the inconvienience. Unfortunately I was now at 200 feet, and getting the engine running smooth put me a little high and fast. No biggy. I dumped the landing brake, checked the gear, hit both rudders and down the elevator we went to the ground floor. Touchdown was a little long and fast, but with bags of room to spare. I taxied back to the hangar thinking that maybe a little more tuning at low rpm was in order here. Strange that it had run smoothly at low rpm and idled fine on the ground, but airborne with airflow it was way too lean.

LATER NOTE: Next day Dave Staten pointed out that there are two map tables in the EC2. The first one, the one I tuned, handles operation where the rpm is low at each MAP reading as in taxiing. The second one handles a decent situation where the prop rpm is high, but the MAP is low as I encountered. You can either tune the second one at altitude in flight, or by removing the prop. You can't even get to it on the ground with the prop on. The in-flight tuning procedure is covered in detail in the manual. The manual also says not to worry about this until you become totally comfortable with the plane and the EC2, and to use the manual mixture control until then. Quoting the manual:

This is a perfectly acceptable way to operate the engine and thousands of pilots flying conventional aircraft engines have used this as their only method of fine tuning the mixture.

Moral of the story: RTFM

So many details. So little brain.

There was a little soot on the prop, but nothing like what it used to be. I'd probably overcompensated a little when cranking the mixture up on final. There was slight oil trail on the lower cowl and some signs of heat on the upper cowl behind the exhaust. Otherwise nothing looked out of place. Underware and seats were fine. Nothing was, bent, dripping, missing, melted, loose or broken. 32.1 hrs on the hobbs and the slick kitten will fly again soon. The "slick" name, while reserved, doesn't suit here right now. I think I can tell the difference between flying in gloss paint and flying in part primer. Just a slight loss of responsiveness. Nothing to write home about, but noticable.

I started and ended this section with an airborne picture. In between there were almost 5 months of delays and electronics issues which, thankfully are now behind me. The image above would have been one of the approach, as requested, but I was a bit busy. Maybe next time.

<<< Back | Index | Next >>>