Cozy

Chapter 23 - Engine - Intake

Getting an intake for the 13B was one of the most difficult parts of the entire engine. I went through a lot of options before I got a completed intake. I started searching for the intake solution around summer 2001. As of November 2003 it's almost done. If I'd known what I know now I'd have got a TIG welding course and made my own. I got as far as buying the welding machine, but backed out an let an "expert" do the work. This cost me a lot of time, money and inconvienience. The following is the story of my intake....

Initial intake Woes

I'd chatted with Bruce Turrentine on and off about an intake. He's been working on a high temp composite design complete with rails, injectors and throttle body. He has a cut off 3rd gen intake which would join to his unit. We agreed a price and I sent a check as incentive to get it finished. A few months later I got the check back, with a letter saying he'd been delayed by more lucrative work, was still working on the intake, but that holding the check was "burning him up". You don't come across this kind of honesty much outside the homebuilt aviation world. Hmmm. Now what am I going to do for an intake?

A care package from Bruce

this is the intake primary injector rail which fits directly into the block Here is the rail in place with the cut off manifold flange in place The 2nd gen throttle body I asked Bruce to send me all the intake parts, and a 3rd gen oil filter block. The box arrived and it was a very small box. No wonder - the intake manifold was cut back alright - cut back to 1/2 inch thick. I've seen the intake work Paul Lamar's been doing. The work seems to be progressing, but it isnt there yet. There's no sign of the throttle body, and If I'm going to have stuff welded together, I might as well use the parts I got from Bruce, plus a couple of J bends from Burns stainless. My box also included a pair of injectors and the rail which fits on top of the block. These are the primary injectors, and they fit nicey right into the top of the engine. Now - what do I do for secondary injectors? Bruce tells me they normally fit into the intake manifold - the bit I dont have.

The intake

I went up to 2003 Sun & Fun hoping to see Bruce's progress on the intake. He'd left before I got there, and apparantly hadn't brought anything to show. I looked at Chrissi's modified Lamar intake. Not bad, but it doesnt solve the whole problem. Someone has to weld J pipes onto it. Then the Lamar runners would fit to the J pipes, but still there's no plenum or throttle body. At the Cozy dinner Ken Knevel showed me his 20B intake design in 3D rhino on his PDA. I discussed the design with Tracy Crook and Bruce Turrentine, both of whom thought it would work. Tracy felt that the injectors might be better off on the runners rather than firing across the plenum.

Finally I could visualize what was needed. If only I could make something like that. I dropped into the welding tent and had a go at TIG welding. After a few minutes I was welding happily. I'd done a bit of Ox/ac welding many years ago and the technique isnt much different. I've already spend more than the price of a TIG welder on custom welding work, and theres a lot left to do. I decided to spring for the Lincoln 175 welding machine and do my own welding. I got a discount on the demo model, plus a ride to the car in their electric cart. After a day walking Sun & Fun the cart ride was worth the price of the machine. My logic was this - I've pretty much mastered the epoxy art. I can make just about anything with fiberglass. Metal, however, has me stumped. My friend, Bill, has a complete machine shop in my back yard shed - lathes, milling machines, the whole nine yards. It's time I learned how to use these machines and TIG weld. Once I break through the haze of ignorance I'll be able to make anything I need in metal as well.

Can you spot the error in this picture? I'm wearing shorts. Read below for why this wasnt a good idea Next day I installed 230v 60amps on my patio. Hmmm. What do I do for a bench. Ideally I need a metal bench for grounding. I wonder how much a welding table costs. Hey - wait a minute - I have a welder. I'll make my own damn welding table. I went to the local steel supply company and picked out a 48 * 30 sheet of 3/16 steel, 50 feet of 1 inch square rod and some expanded steel for a shelf. The next trip was to the welding supply house for some mild steel welding rods. I used Bills cutting machine to cut the tube into lengths for the frame and legs, then stick welded it all togehter. In a couple of hours I had a presentable table. Cool. Well not exactly - you see I did the welding on the concrete patio floor wearing shorts and a short sleeved shirt. When I was done I had a pretty good (read somewhat painful) suntan where my skin had been exposed to the torch radiation. Oops. There'll be no naked welding for me. Next time I'll wear jeans and a long shirt like it says in the instructions. Duh!

My welding table finished. Note how I welded the wall while tacking the top in place. Oops. One thing it didnt mention in the instructions was the acetone rag I left on the steel bench. You need acetone to wipe the parts clean of any oils. Once you have the welding hat down and the torch going the world kinda goes away, and you don't notice anything on the "outside". I finished a practice weld, lifted the hat and noticed that my acetone rag was merrily burning right in front of me. I knocked it on the floor and stamped it out. Another lesson learned. I moved the acedtone can well out of harms way and started in on some aluminum TIG practice. My first weld of two 1/16 al sheets went pretty well. A little clean -up with a grinder and even the first weld would be ok.

As Char so nicely put it, I went from the first practice right to the final exam (and failed). I got out my cut off manifold flange and some al pipe and tried to weld them together. Like everything else, TIG welding gets tricky when you get out of the practice stage and into the real world. When compared to epoxy, it was a bit like doing the seat-back, then jumping straight to the wing root layup. The complex compound curves of the intake where the pipe joins to the flange were impossible to match with a piece of straight pipe. I managed to get the pipe welded on, but there were huge holes on either side. I decided that the pipe would need some serious preparation before being welded to the flange. For my first experiment I wanted to confirm that I could really join a pipe to the manifold flange. I used the dremel to grind the weld and, sure enough, there was no visible join. The two pieces had become one. OK. I cut my first attempt off and began trying to figure out how to make a decent join. I tried to build a wall of aluminum filler material which I would then grind as needed, but was having trouble getting a good "pool". After experimenting for a while and talking to the support guy at Lincoln I determined that I needed a bigger electrode, bigger filler rods and all the amps I could get. This learned, I was able to build a wall ok, but grinding it was a pain. Grinding the inside didnt seem like it was going to work. After a bit of experimentation and conversation with Chrissi I decided to buy Paul Lamar's cast manifold and runners and fabricate something similar to Ken's design from there. I think I'll put the throttle body in the middle of the plenum instead of the end if I can work out the piping from the intercooler.

Deciding to eat a little crow, I sent Paul Lamar a check for the runners and manifold with a note:

"We might not always see eye to eye, Paul, but we have the same aim in mind. 
The net effect of you’re work is definitely in the right direction and I need to 
keep in touch with it. Please put me back on the list. I promise to try to behave."
While waiting for the Lamar manifold parts I tried a bit more welding on an aluminum avionics frame I'd made. Total disaster. The aluminum just would not stick together and eventually just melted and deformed until it was scrap. I built a new frame and rivited it together. When the manifolt parts came I took them down to Japtrix to be sure of a quality (and fairly expensive) job. Anyone want to buy a TIG welder - Never used ...at least not sucessfully.

Later I read that 2000 grade aluminum, such as that found in home depot, is impossible to weld. I think I'll try TIG welding something simple with 6000 grade aluminum. Maybe I can get my confidence back.

The local mazda guru

I'd been told about Japtrix, a local mazda shop where the guru, Roger, really knew his rotary stuff. They're shop is about 3 miles from my house, so I dropped in and told Roger what I was up to. He agreed to visit and take a look at what I needed done. He showed up the following Sunday with a freind who knows something about aviation and proceeded to pour over my engine for an hour. We discussed the turbo, 3mm seals, high compression rotors, exhaust augmentation, fuel system, ignition computer, coils, controls, AC, alternator, smog pump, seal lubrication and just about everything else related to the engine. Roger obviousy knows his 13B stuff. He's been a "Wankler" since '84, he says. He was happy with my overall plan, but wanted to talk to an aviation expert he knows about the injectors. I offerred to give him the phone numbers for Bruce Turrentine and Tracy Crook as well. Step one would be to get the engine down to his shop, make brackets for the alternator, smog pump and AC compressor and get the surpentine belt fitted. Once that's done we can start hooking up the various hoses. My only concern is the alarm buzzer that was ringing loudly in my wallet.

Mr. P.

Japtrix have all their welding done by a local guy, Mr. P., who comes in when needed. I've seen some of his work and it's beautiful. Unfortunately he has another life which gets in the way of his welding duties, and it took about 3 months to get the manifold tacked together to the point where we could see how it fit under the cowl. It didn't. When I made the cowl I decided to make it the "ideal" shape for airflow and worry about getting everything under it later. Well now was later. Roger came over to the hangar to see how well the prototype intake fitted. The bell for rearmost runner was right up against the cowl, so the intake would have to go down about an inch. This part would be easy. Fitting a 4 inch diameter plenum wouldn't. We decided that the best way would be to put a 45 degree bend in the end of the intake just before the bells to get the plenum down away from the cowl, and go with a 3.5 inch plenum. The throttle body would go at the forward end. Next we talked about where the intercooler and ac condensor could go. The best spots for these seems to be in the wing roots. This way I can try ducting air from the radiator plenum, and if this doesnt work they'll be ideally located for armpit P51 scoops.

I dropped in at Japtrix to see how much progress Roger and Mr. P. had made in the past two weeks since trail fitting the manifold. Roger proudly showed me how they'd cut the runners in the wrong place, then explained how they would be working on it tomorrow and should have it ready for another test fit by Thursday. Great. We're making progress. Roger is a master at customer relations. I left wondering how he'd made me feel good by showing me a backward step. Motivated, I went down to the hangar, fitted the lower cowling and searched for a good place to fit the intercooler.

Around this time RV wankler, Ed Anderson announced that he'd come up with definitive physics to describe the Dynamic Intake Effect. Not liking the acronym, DIE, the fly-rotary list guys renamed it the "Eddie" effect. Apparantly the intake needs a path for the pulse from one intake closing to bounce back to the other intake and cram the air in. Look carefully at the 3rd gen intake and you'll see the patterns of the two curved areas of the dynamic chamber which allow for this. Ed's not sure how important this is on a turbo installation, and it's a bit too late for me, at least on this intake. For those considering a NA approach, Ed's EDDIE info could be very important.

The fit is so tight that the intake only just goes far enough to the intake side to fit over the studs at this stage the intake weighed 11lb Another couple of weeks went by. I picked up the revised manifold for another trial fit. This time it DID fit under the cowl. Just. The plenum touches the cowl at the rear, and the rearmost injector boss is also touching. OK, so I made the cowl tight. I aint putting any lumps in it. I marked the intake in the offending areas and took it back to Roger. OK, says he, shaping the back of the plenum is a piece of cake. Moving the injector boss isn't. I left him thinking about that one. Later this week (early October 2003) the throttle body and rail blank are supposed to arrive. We're still not sure where the rail and injectors will go. The primaries can stay in the block, but the secondaries are beginning to be a problem. the top of the plenum touches the cowl at the rear. Pencil line is the part that has to be removed. the rearmost injector port is up against the cowl.

The throttle body arrived, and was promptly returned. It was WAY too big. The throat was the right size (65mm), but the overall unit was about 6 inches long and unnecessarily heavy. A few days later the right one arrived together with a fancy (and expensive) fuel pressure regulator. I'd thought that the pressure regulator would mount on the rail, but this one is designed for firewall mounting. I found a spot for it on the cool side, then decided that the fuel return solenoid should also move over to that side. This required revising the fuel pipes so, While I was at it, I converted from aluminum to stainless braided flexible hose. I removed all the pipes from the solenoid, then forgot which was the common, so I switched it on. Nothing. It got warm, but didnt click. I dissassembled it and found that the actuator have become covered with brown crud and had seized. I'd had fuel in the system 6 months ago and fuel had been sitting in there, but I was suprised that these deposits had collected. I freed the actuator up and checked its action again. Click. I'd better keep an eye on that. This failure mode would cause fuel to be returned to the right tank while drawing from the left. If done with anything over half full tanks, this would cause vent overflow from the right tank. I'll put the solenoid valve on my regular mechanical check list. If it clogs up again I'm going to have to rethink the fuel system, again.

the plenum under construction. Interesting how he formed a dome on the end.

Almost Done

The intake fitted. Still needs some final clean-up The aft end has a dome to clear the cowl and help the pulses bounce - Ed says that's good. <IMG SRC = Currently the primaries still use the stock injector mount on the block. This will make the rail fabrication complicated, so we may put the primary bosses parallel to the secondaries on the runners.

Finding a home for the intercooler

I've been wondering where to put the intercooler ever since I got this 3rd gen unit from Bruce about two years ago. Some 13B drivers have gone without an intercooler altogether. All the input I have says you need one. I remember Greg Richter saying he needed MORE intercooling than he had. We don't want detonation, now do we?

The best place for the intercooler seems to be on the left (cool) side below the throttle body. This will make the feed from the turbo quite long up and over the back of the engine, but the connection to the TB will be very short. I balanced the intercooler on blue foam until I had a postion where it didnt interfer with the mount, the rudder cables or anything else critical, but still fit under the cowl. I'd made a 3 inch duct from the plenum to feed it, so the next thing was to duct the air up to the intercooler. Using a hacksaw blade, blue foam and duct tape I made a plug for a duct and glassed it in 2 BID. I figure that the duct and the pipes will hold the intercooler firmly in place, so there wont be much need for additional mounting structure.

Almost done some more

Just before thanksgiving, 2003 we trial fitted the "final" intake with the secondary rail and injectors in place. The injector bosses were angled slightly upwards, so that the rail impacted the intake and it would be impossible get a straight fit on the injectors. The bosses had to move. Mr. P wasn't pleased. Welding isnt too hard (for him). Unwelding is a pain. The final two items on the intake were the braket to hold the secondary rail firmly in place, and the bracket to support the intake itself. We should have the intake back and ready to install on Tuesday. That's December 2nd. Dec 17th is getting awfully close....

After a few more days delay the secondary rail support was done - sort of. It only has one bolt, and the rail can twist from side to side. I decided to install a second bolt to stop any movement. I also decided to take the intake to Charlie for the engine support bracket. After prolonged discussion we decided that the best way to do this was for him to make a couple of generic oversized brackets which I would drill and mount to the engine. I'd then grind these away until I had a nice snug fit on the bottom of the runners. I'd mark the spots where they met the runners, then Charlie would weld them up. It would have been a quick and easy job for Mr. P, since he has an engine next to his welding gear, but it wasn't worth the wait. Sometimes you just gotta do it the hard way to get around or over the problem.

December 17th 2003 came and went, which is more than can be said for either the replica Wright (or should I say Wrong) Flyer or my airplane. I didn't get airborne, but Burt's SpaceshipOne got to 900+ mph and 68,000 feet. I decided I was happy not to be in the competion for doing things first on 12/17/03.

Getting new injector O rings from the Mazda dealer took almost 10 days. Now the intake was ready for final installation. Not quite - I had Dave drill and tap the plenum for three 1/8 NPT fittings. Two will be for the EC2 vacuum, and the third id for the air temp sensor. We still have to install a "pop-off" valve to prevent overboosting the turbo. This will consist of an aluminum filler neck welded onto the intake plenum, and a radiator cap of the required PSI rating.

In the meantime, I ordered some silicone turbo pipe connectors from Hose Techniques who seem to have good prices, and I found a Turbonetics tight bend 2 inch aluminum casting at Road Race Engineering, part number ALEL200. This cast bend will be cut off and welded to the turbo outlet to keep the plumbing under the cowl.

I used a grinder to clean up the welds, then painted the intake gold. The gold was too much, so then I repainted it in cast aluminum spray paint. That looks much better. I made brackets to support the plenum side of the intake and had them welded in place. At last I installed the rails, injectors and fuel pipes, inserted the expensive metal gasket from Mazda and tightened up all the bolts. The intake looked great, except that it was now touching the oil return fitting. I'll have to remove it again, grind away a bit of the pipe and reinstall it. Installing the intake is a bit of a puzzel. The fit is so tight that things have to done in the right order, or it wont go on.

Plumbing the Turbo

My 2 inch cast aluminum bend arrived and I tried it under the cowling. Looks like quite a chunck of the bend will have to be cut off to keep it from impacting the cowling. Roger told me the turbo housing would have to come off to do the welding, so I removed the bolts around the housing. At the Cozy Christmas gathering I described how I was planning to control the turbo with a vacuum solenoid. Bulent suggested that the best way to control the turbo would a manual push pull cable on the wastegate. This has the added advantage of allowing me to toss the big vacuum wastegate controller. Two of the bolts on the housing held the vacuum unit, so off it came. When I had the turbo housing off I decided to sandblast it. My hangar mate, Dan has a sandblaster so I cut a deal. Free usage for a bag of beads. I've been sandblasting pretty much everything that'll go in the box. It's amazing how a rusty old smog pump or alternator can look brand new after a few minutes of blasting. Another nice thing about sand blasting everything is that it removes all the old grease, so you don't get grease all over you're hands while working on the engine.

Anyway, back to the turbo. I was working my way around the housing with the blasting gun when something occurred to me. The housing looks totally symetrical, except for the outlet. Sure enough, the bolts just clamp the housing in place, so you can orient it anywhichway you like. I reinstalled it with the outlet pointing down instead of up. This solves the cowl interference problem while also shortening and simplifying the run from the turbo to the intercooler.

I've been looking for aluminum bends, and the only ones I can find are at Burns stainless. The 90 and 45 degree bends range from $35 to $45. Seems like a helluva lot for a bent piece of tube that costs about $6 when straight. Given that I don't really know what I need and will probably have to order extra pieces I resent paying that much for what is probably 30 seconds worth of bending machine time. I searched around some more and came across a "cold air intake kit" for an RX7 on ebay. The picture looks like it contains a 90 bend, a pipe with two 45 bends, four silicone connectors and eight clamps. Bought seperately from hose techniques and Burns these parts would cost about $200+. There were three of them for sale. I got the last of the three for $13. I'll have to wait till it arrives. It may be a waste of money. It may be a discovery. The other "challenge" is going to be plumbing the intercooler (2.25 OD) to the throttle body (3.25 OD). The pipe has to twist a bit around the engine mount. I spotted an intake pipe for a Mitsubishi on ebay that "looked close" and got it for $7.50. I ordered 4 feet of straight 2 inch 6061 pipe and a couple of feet of 2.25 6061 pipe from Wicks. Next week will be fun as I try to make all these parts fit.

<<< Back | Index | Next >>>