Peak Cylinder Pressures?

[TDI4evah]

I thought I'd heard somewhere about a pressure gauge of sorts that screws into the glow plug socket. I like to tweak and tune, and both power and efficiency depend a lot on PCP's.

Does anyone know if such a thing exists and, if so, where I might find one?

TIA

 

[Fix_Until_Broke]

This has been discussed a few times in the past.

You're going to have a minimum of $5k for one cylinder and another $2-3k for each additional cylinder.

www.kistler.com
www.pcb.com

You need a data acquisition that can sample and store at a minimum of 10kHz and can plot.

www.dewesoft.com makes a nice product with very capable acquisition and analysis software (though it's difficult to learn)
www.dataq.com makes an inexpensive product, however with very limited software

Probably the cheapest thing would be an USB oscilloscope (www.propscope.com) with a crank trigger in the 2nd channel of the scope and delay the trigger so you see the compression/power stroke. No real ability to save data, but you'll get a visual of cylinder pressures and maybe injection timing if your trigger is very stable and you're zoomed in. You might be able to do this for a couple thousand.

The sensors are the most expensive part. You might be able to use the new Common Rail Glow Plugs since they're also pressure sensors. They're ~$125 each but you'll probably have to do your own bridge completion and temperature compensation to get good data.

 

[TDI4evah]

Wow, I'm impressed. Very comprehensive response. Thanks FUB.

I've got some homework to do now. Will the CR glow plug screw into an ALH head OK? That's the only choice I could afford.

Thanks,

--- David

 

[JFettig]

I am using a dataq DAQ for doing boost, emp, and n75 vac measurements.

PCP measurement would be very helpful for tuning but the # of variables that can effect PCP are endless.

Hope you're a test engineer of some sort

 

[Vince Waldon]

Hope you're a test engineer of some sort

Yup...of the electronics type for starters, because of this:

have to do your own bridge completion and temperature compensation to get good data.

 

[flee]

FUB, why do I get the idea that you've already considered doing this?

 

[Fix_Until_Broke]

I don't know if the CR glow plug will screw in to the ALH head or not - I'm told that they are the same thread, but don't know about the protrusion into the cylinder or the seat/seal.

Like I said, it's been discussed before and I have considered doing this along with a few hundred dozen other things that I have neither the time or money for...

 

[brum]

I think this is the Beru PSG used by VAG (the picture below). I've thought about this a few months ago and it seems that creating an adapter is possible. But that was all .

For useful data logging you'll need pretty fast logger. Something in the range of 100 000 logs per second. So most car data logging systems are useless here. Also I'm not sure that the PCP will be sufficient for best SOI adjustment. I think you'll also need precise position sensor to log the cam position. On PD I think you can use the crankshaft position sensor. But on VP engine it has only 4 positions and is useless for that (still wandering how the ECU guesstimates the correct SOI angle with it ).


http://motoplakor.rdl.pl/SWIECE/03L905061E.jpg_4.jpg

 

[Digital Corpus]

Those are too long for a 1Z

Goes back to his data sheet for a PIC32, model families 5XX/6XX/7XX, to see how fast A/D sampling is again...

Up to 1Msps conversion speed

A closer look, Section 17.9 in the A/D data sheet, reveals that it can do this while running at 60 MHz.

</random thought>

 

[TDIMeister]

http://forums.tdiclub.com/showthread.php?t=207265

 

[brum]

Ha, this is what I've researched back then .

 

[Fix_Until_Broke]

I think those beru's are too long for an ALH as well - at least the distance between the threads and seat is too far for the ALH head.

10k samples/sec will get you pretty good resolution for PCP, but will be a little coarse for SOI.

1000 RPM = 6000 deg/sec = 0.6 degree resolution at 10kHz just for a point of reference. A pickup on the flywheel (137 teeth) will get reasonably accurate crank position on a digital input into a DAQ (Cheap) - you can interpolate pretty well between teeth if necessary as well as rising and falling edge of the teeth. Best case is a 16 bit encoder on the crank, but those are real $$

I believe the ALH determines SOI from the time between injector needle lift sensor and the crank position sensor at the given RPM to determine degrees without needing super fast sample rate ECU.

I've looked into TFX, but have not had good luck getting any replies to my inquiries...?

 

[Digital Corpus]

If you can do 100Ksps then you can have a resolution of 0.3˚ @ 5000 RPM, or is idle the only desirable area wanting to be observed?

 

[Fix_Until_Broke]

Yes, you would need that if you were to measure the crank position analog - this is why these measurements are typically digital or with an encoder and interpolation is done so the sample rates don't need to be silly fast. 50k gets you just a little more than half a degree which I'm guessing is plenty resolution.

The 1000 RPM point of reference is just that, a point of reference so you can easily calculate what you get at 2000 RPM or 5000 RPM or whatever sample rate without having to break out a calculator.

 

[DubNation]

jesus christ thats expensive

 

[Syncrogreg]

ka boom!

http://www.idparts.com/catalog/product_info.php?products_id=2466

Greg