ptuomov

The plan is to take a number of measurements. Any further ideas would be greatly appreciated.

Among the many sensors being installed are:
- Turbocharger speed sensor that measures the rpm of the turbine-compressor shaft. This is useful for optimizing bypass valve spring load for shifts etc.
- Outside temperature sensor inside the IC duct. This is useful in computing the intercooler effectiveness.
- The pre-turbine exhaust back pressure. This is useful in calibrating the simulation software to match the exact turbine (EAP can't accurately match turbos without these data) and in selecting the camshaft profile.

I did independently confirm the presure differential between the wheelwell and the bumper face separately using silicon hoses and a manometer. That was a one-time job on route 2. John had of course done this earlier, but I just had time to kill.

hacker-pschorr

#1 suggestion that comes to mind is test the temp senders before installing them (heat gun / hair dryer etc..)

Two reaons.
1. Avoid installing a broken one
2. Make sure they are sensitive enough.

Most cannot react fast enough for an accurate reading of intake air. Any old water / oil etc... sender will not be accurate.

I'm still playing with a few myself. I just had Todd install two different units above my TB in the plenum. We'll see how they compare. By the time I'm done installing sensors (not until after SITM) I'll have at least 5-6 intake air temp readings.

ptuomov

I'll only go that far if the intercoolers aren't the total overkill that I expect them to be. As you said, no need to get fancy then.

I think you could adapt Garrett's turbocharger speed sensor to the centrifugal supercahrger unit. I don't know much about belt-driven stuff, but I vaguely recall that centrifugal units need a belt that can slip a little. Getting the actual speed reading and comparing that with the engine speed will allow you to compute the belt slippage and also to exactly pinpoint the location on the compressor map.

ptuomov

I am staring to feel good about THE GAPING HOLES IN THE BUMPER...

tveltman

What sensors are you using to monitor EGT and exhaust backpressure? I would be interested in getting a set for myself if you have details. Actually, details on all of your sensors would be delightful. Thanks!

ptuomov

Right now, I just have ports for the exhaust back pressure measurements. After turbo is not a problem, but pre-turbo may require a cooling tube before the sensor. In other words, I am not sure yet what I'll be using and if I may be able to avoid the cooling tube.

The turbocharger speed sensor is from Garrett. It's on their web site. Install's to the compressor housing.

The whole data logging system beyond John's standard product is still work in progress.

tveltman

If it helps, I think I read about Louie Ott's individual cylinder EGT monitoring system. I can't find the post right now, but he used some sensors hooked up right out of the manifold. I would use those, if I could find the link...as for pressure, they must exist somewhere...

ptuomov

EGT monitoring is old school. New school way is to use those ports to monitor lean/rich condition, especially if you have an aftermarket ECU that can do clyinder-specific fuel trim. I think that requires another eight tubes down there, since wideband o2 sensors will not live in the cylinder head exhaust port.

Rob Edwards

Like this?


https://rennlist.com/forums/928-foru...ds-garage.html

tveltman

That is unbediculous...

BC

Have you found an aftermarket ECU that has Fuel trims per cylinder on a load basis? Instead of just on a global percent basis, which does nothing if the intake is not perfectly matched to each cylinder. That does not exist in the S4 intake manifold, and Todd found that some had over 20% difference, up or down, according to the revs and load.

When I found this out, I asked questions of the people involved in the ECU market. I found two, and two only:

Autronic (which Todd uses)
Pro EFI 128*

* This is an untested system as far as I am concerned, but its promising.

SQLGuy

Megasquirt with MS2Extra will do it for a four cylinder running sequential injection (separate percentage cylinder trim map for each injection channel). I would think the logic could be extended to eight cylinders, but you'd need the sequencer/MSIII which is not yet available, or you'd need two Megasquirts running the MS2Extra sequential code... one for each bank; which is weird, but theoretically feasible.

BC

Percentage change in trim does not factor load.

ptuomov

There are four bungs, two pre-turbine and two downstream of it. The exact configuration of what goes where, whether the narrowband is simulated from wideband, etc. is open. Initially what you say is one likely configuration. In the longer run, for tuning / logging one is probably going to be pressure sensor. Que sera sera.

The latest photo should again communicate the perspective a bit.

That's what I had in mind. That's more of a personal R&D lab than a car meant for a customer, though.

Yes, but it's a lot better than nothing.

Lizard928

I would caution against putting any O2 sensor pre-turbine. There is too much heat there and it will cook them in a BIG hurry. That is even with a copper heat sink.