ptuomov

Both my parents were dentists working full time there and in 1988 the were able to barely afford two Toyota Corollas. That's something that is hard to explain to anyone here in the US, given the GDP per capita being relatively high there. Regardless, I wasn't very old when I decided that wasn't going to be my life.

No need. I've got all the cams that I need for any possible experiment that I can conceive of. Here's one modified GTS set, with our common friend's adjustable sprocket:



I'll pass your comment on to John, who's kit the screen belongs to.

John's kit is designed to work with the minimum amount of modifications to the stock car. Only the parts that must be modified have been modified.

When the stock engine experiments are completed, the low-compression engine goes in. It's the one in my avatar. That engine has a pan spacer and therefore will sport a mini-starter.

ptuomov

Here's a short video that summarizes some of the changes to the car at this development iteration.

Oh, yeah, and the car runs, too! ;-)

ptuomov

Next order of business is to make sure that the revised system can control the boost. The wastegate has a 7psi spring and at this point the solenoid is running boost signal interrupt but not yet the offensive control. Signal interrupt makes it so that the front of the diaphragm sees pressure that is the same or less than the manifold pressure, depending on the solenoid duty cycle. Offensive control makes it so that the back of the diaphragm sees pressure that is between the atmoshperic pressure and the manifold pressure, depending on the solenoid duty cycle. The offensive control is not needed at this point.

The car is strapped on the dyno:



Here are the torque and power curves from a mild shakedown run with retarded ignition and 9 psi constant manifold pressure. (I say mild because the most that this engine has seen so far with the old system is about 25 psi at the midrange rpm and 18 psi at the top end rpm.) The resulting 500 rwhp at 6400 rpm is very close to our expectations. By the way, it's STD corrected, not SAE corrected, if someone gives a **** about it -- I don't:



The turbos were spinning at about 90,000 rpm based on the censors, the engine was consuming about 30 lb/min of air per turbo, and the compressor pressure ratio was about 1.8 after accounting for the intercooler pressure drop. All those measurements have a big standard error at this points. Ideally, we want to compute the mass air flow from the MAF signal and measure the compressor inlet and outlet pressures right at the compressor. Nevertheless, we're close enough for government work on the compressor map:



One thing that is interesting and consistent with our past experience is that the 928's power scales faster than proportionally to the manifold absolute pressure. If the stock engine makes 316 crank hp, the first guess for the boosted crank hp at proportional scaling is 316hp * (boost pressure in psi + 14.7 psi)/14.7 psi. For rwhp, replace 316hp with 280 rwhp. For 9 psi, you get something like 509 crank hp or 451 rwhp. What you get in reality is something more like 550-600 crank hp and 500 rwhp.

The main explanation for this is that it's the gross power that scales about proportionally (minus the little bit of extra energy needed to turn the turbine), while the power lost to friction scales less than proportionally. Turning the cams, the auxiliaries, piston to bore friction, etc. doesn't grow proportionally to the absolute manifold pressure, making the gains from forced induction really nice. A very simple formula that I've calibrated to approximate the friction that doesn't scale with manifold pressure is 80 rwhp, which gives the following predicted rwhp formula for the engine with these turbos: (280 rwhp + 80 rwhp)*(boost in psi+14.7 psi)/(14.7 psi) - 80 rwhp = predicted rwhp at the given boost.

No wonder pretty much all the new cars will be turbocharged now that the governments are cracking down on fuel consumption and emissions.

pdejong

Is that dyno chart with modified cams?
Here's my GT with 9 psi boost. Very similar. (Different dyno)
I am curious to see what you will ultimately get out of your engine.

Peter
Charleston/Houston
'90GT TT
'95 968 Cab

ptuomov

Stock S4 cams. The engine that is currently in the car is still completely stock with absolutely nothing changed in the air flow tract between the throttle body and the exhaust ports.

This system should probably do a little bit more in terms of rwhp than yours, because it has much better flowing turbines and wastegates and thus lower exhaust back pressure at 6400 rpm. I think you've got GT2871R's, this car has GTX3576R's. Under 3000 rpm, I'd expect your turbos to have a big edge. The other difference is that my compression ratio is 9.4:1 with the stock squish distance and combustion chamber shape, which is pretty efficient at these low boost levels.

Vilhuer

Interesting things in it. 1000+ Nm torque from stock egine internals sounds great fun. We might copy single output 944 S2 pressure requlator idea to Niklases S4 SC build.

ptuomov

The 1000+ Nm torque is old news, and it was just an experiment to see where the turbo hit the sonic choke. See this post: https://rennlist.com/forums/10435284-post627.html. The guess in that post turned out to be wrong -- in the end the bottle neck was not the boost leaks but simply the compressor wheel hitting the sonic wall and not flowing any more mass. Hence this next step of moving from GT3071R's to GTX3576R's.

We're running two 944 dampers in the front, one per side. The pressure side of the fuel line flows a lot now with low pressure drop. The rear, including the return-style pressure regulator, is pretty close to stock. The only reason why the rear is large enough is that the second Bosch 044 pump is brought on line only when there's fuel demand and therefore the fuel doesn't have to be returned.

ptuomov

Another low boost dyno run caught on video. There's nothing there in terms of power at 9psi, but some interesting bits about holding the rpms vs. sweeps.

John Speake

Very interesting - a great investment !

Tony

Very cool..what a great tuning tool!

ptuomov

In eight weeks, we're going to get a _very_ interesting data acquisition system. We will know _everything_ what happens inside the engine. I am exaggerating only slightly. More to come.

Hilton

Did you make Ken's modification to the LH binary to enable MAF smoothing?

He updated the XDF about 2 weeks ago linked in the second post.

DKWalser

Very cool. Is that your new and improved set up?

Too bad the dynos don't produce electricity from the power generated by the car.

ptuomov

No. The software is all written by Niklas Kämpe. I am not screwing around with the code myself for three reasons. First, I don't know what I am doing. This alone should be sufficient reason. Second, I want everything we learn with this car to benefit Sharktuner 2 development and thereby the community. Third, if all the features that I need will eventually become features in ST2, then they'll supported down the road.

John Speake

No problem to supply you with modified code when you need it...