Originally Posted by
stuartshomepc
This is actually quite true.
The best way to think of these engines is an "airflow processing capability". you want to get a lot in, and a lot out. So one would be wise to look from the point air goes in, to where it comes out (the tailpipe).
The exhaust manifold design sure looks like a piece of crap, designed by a bad plumber (no argument there). We worked on that and even designed a nice long runner tubular header for the T II engine. We were all kinda shocked that it really didn't help. The way that the Turbocharger inherently works produces high back pressure. At times I saw upwards of 45psi in the inlet to the turbine. With those pressures, the goal is to just get the exhaust gas out. Larger runners or a more elegant design would only work with a larger turbo, that had reduced turbine inlet pressure. But a larger turbo gave us more lag at launch, something the executives didn't want.
This high turbine inlet pressure is one reason we went to very low, or no overlap camshafts. As first we applied "naturally Aspirated" thinking and increased overlap, and kept loosing power. We finally realized that we were blowing 12psi boost in, bit were working against 35 to 40psi in the exhaust manifold. So until we went to zero degree overlap cams. or a 5 degree clocked a bit, we picked up power. With any overlap on the cam, we were letting exhaust back into the engine.
We had a restrictive exhaust system too, especially with the mandrel bends behind the catalyst. We fought hard for an improved exhaust system and it just added cost and most did not see the Return on Investment (ROI). I set up one of our executives wit a 2.5" mandrel bent exhaust and a reduced back pressure catalyst and muffler. It still met emissions and noise requirements and it "woke the car up". Even with that car, and a high level executive pushing for change, there was a "why bother" attitude. Our transmissions/clutches were barely holding up and they didn't want more power. And the Torque steer would just get worse. So getting more power beyond the stock Turbo II was not a priority; launch feel and getting a quicker boost rise was. And the suggestion was for a smaller turbo. I had to put on on a car, even though I tried to explain the dynamics and how it was going to become a "choke point". Well, the car launched great and squealed the tires when it launched. Everyone seemed happy, until the engine "tried" to go above 4800rpm, and was "self governed" as the back-pressure was so hihg not enough inlet air could get in. The proverbial "potato in the Tailpipe". So we were kinda stuck as Garret had noting to offer us. And it was their internal oil ring seal that was an issue as it sealed well, but was to tight is caused a HUGE amount of drag on the turbine at idle. And that is when I got permission and started looking at turbo's from KKK, IHI and finally Mitsubishi (MHI). I set up 2 identical cars and took the executives into a small lab where we had equipment to measure turbine speed. At idle, the Garret turbo was sitting there at 0rpm. The MHI was spinning at 13,000prm already. Typical turbine speeds at full output would be ion the 150,000rpm range, so "starting from something" was better than nothing. And the car with the MHI turbo and equivalent a/r and compressor trim launched much better than the Garrett Turbo. Garrett's response was pretty much "this is what we sell, buy it" yet MHI wanted the business and made multiple changes to accommodate us, including a bit lower internal back-pressure.
The .48a/r Garrett turbo was all I had to chose from as the .63a/r (aka:"super 60") introduced too much launch lag. I was always beaten up for "turbo lag". They wanted quick boost, and low back pressure at Wide Open Throttle. But twin scroll turbo's were in their infantries then. so the MHI was the best I could get. The VNT was an interesting attempt, but we had extreme difficulties controlling it with modulated pressure to the vane actuator can. It BEGGED for a linear actuator for control, but none would stand up to the heat. Build variability exceeded what our control system could manage. And when the carbon/unison ring (the ring that turned the internal vanes) started sticking, there was nothing we could do to control it. Some would only make 7psi "out of the box", while other screamed up to 20psi and him the maximum boost fuel shut off. And the program died quickly.
The T III was more of a politically driven desire to get Lotus involved. We could have done it, but were not given the task to develop the head. Public relations and executive agenda's made the decision to build that head with Lotus. A 200hp/liter engine was promised by Lotus, and Chrysler engineering could have done it (given the chance). Within the first week I heard of their peak RPM target of beyond 7000rpm. It brought up an issue that I fought for year of fuel injector minimum flow (needed to be stable and consistent for good idle quality) and the maximum flow (called "turn Down Ratio"). The bottom like was there was not enough open intake valve time to get the fuel in at higher engine speeds, and still get a decent idle quality. I was told I was wrong. Same thing on the way I designed the Charge Air Cooler (intercooler) by routing the hot air into the bottom (the cars of that era had a high pressure air zone below the bumper, so I put the hot air part of the where the most cold air was coming in. Lotus put the hot air in the top, charged Chrysler $250,000 for a wind tunnel study to show it worked better with the hot air going into the bottom of the core (just like I designed it) and I blew a gasket. I was supposed to spearhead the T III program but I knew it was going to be a fight of "we're smarter that you are", and I asked to be removed. In the end, they had to ramp the boost down from 12psi to 9, as gee, they were out of injector flow. And the Turbo they picked was wrong and was turning so fast it was way out of it's efficiency zone, resulting in very high turbine out temperatures. just like I had told then in the original meetings. The engineer that had to do the calibration had an awefull time with it. He kept fighting to get 12psi through the rev range and called me down to the dyno to get my opinion. I was one of the quickest and best WOT fuel/spark mapping people, but there was nothing I could do. Exhaust gas temperatures were too high, it was too lean. After about an hour I said "this will never work. Match the boost to fuel flow". He did it and it brought everything down where the target temps needed to be. But when he had to inform management about the boost reductions at high engine speeds, they didn't like the facts and had him try again. It was a lost cause. I was SO glad that I stepped out early. Did they make their goals? yeh, Could it have been a better performing engine?? Some of your members showed it could have and DID get more power out of it.
BUT, even on that motor, the exhaust manifold was crap, but it did make decent power. The Maserati Turbo exhaust had some good qualities, but the runner diameter was too small.
In the end, we decided that the exhaust manifold design was really not as important as matching the right turbo and getting a good calibration. So we put our available hours there.
So was the manifold good? Nope. Did it matter as much as other things? Nope. So why optimize what had little return on investment? We had to "pick our battles" as we were such a small group.
Stuart