JasonC SBB

I've done all of the following. I'm a geek, what can I say?


Turbine inlet pressure

Exhaust system backpressure

Crankcase blowby pressure

Crankcase blowby flowrate (waiting for ebay flowmeter)

Intercooler pressure drop

Chassis resonance frequency to measure effect of chassis stiffening mods

Air pressure differential across radiator

water temperature drop across radiator

air temperature drop across radiator

Rotor temperature before and after ducting

Datalog chassis vertical g's


Any other good tests?

miata2fast

iTrader: (2)

You are dedicated.

Interested in the crankcase pressure and flow rate. I think I am going to build an evacuation system, and do not want to go overkill on the pump.

AlexO35

iTrader: (7)

What kind of electronics are you using to data-log all this info?
--Alex

devin mac

iTrader: (2)

yeah, can you provide some details on the setups used to perform all this testing?

JasonC SBB

Ah I forgot,

- intake air filter pressure drop testing

Here are my instruments.

0-30" water Magnehelic pressure gauge (crankcase pressure)
0-2" same (aero testing, air filter drop)

2-channel handheld thermocouple reader

for turbine inlet pressure - 1/8" NPT compression fitting -> 1/8" copper tubing -> vac hose
-> cheapie 0-50 psi pressure gauge from the hardware store

for exh backpressure bolt that fits into O2 sensor bung, drilled, compression fitting, copper tubing -> boost gauge

blowby flowrate Dwyer RMC-106
http://www.lehmanscientific.com/p_dwyer.html

3-axis accelerometer and software borrowed from FatCat Motorsports

kday

That's a good set of data. Have you collected it all somewhere?

I just machined a fitting for a coolant pressure sensor on my in-progress build. That's something you don't hear of very often, but I think it will be better than a float in the expansion tank to detect leaks quickly, and my car doesn't have a float anyway. I'll also have fuel pressure, accurate oil pressure, EGT, compressor wheel speed, and probably more as time goes on. All going into the megasquirt log file.

For my Audi I built a homebrew data logger with embedded flash and wrote a GUI to graph it.

I hate writing GUIs so I am happy that I can use MLV now

JasonC SBB

Good stuff.

For a lot of the stuff I've measured, I didn't datalog em.

However it would be easy to use say a 3 bar MAP sensor to log TIP or exh backpressure.

There are 0-1 psi sensors that one could use.

Jeff_Ciesielski

iTrader: (10)

Jason - Out of curiosity, how much vac were you seeing @ the turbo inlet?

Edit: I might have a reading comprehension fail here, turbine inlet meaning exhaust manifold pressure pre-turbine wheel?

JasonC SBB

Compressor inlet, like 0.3 psi.

I cleaned the air filter and then got .... 0.3 psi.

Joe Perez

iTrader: (8)

I'd like to see a map of the pressure gradient across the compressor wheel while in operation, and the effect of turbulence generated by the nut holding the wheel onto the shaft on the efficiency of the compressor. Ideally, we should compare nuts of varying designs, including socket-cap nuts which are smooth around their entire circumference.

JasonC SBB

Yes I measured turbine inlet pressure (TIP).

The GT2554 spikes up to 30 psi above 6000 RPM, with 10 psi boost, on my 2000. Through the midrange it's about 15-20 psi. This is why on the dyno when I turned up the boost I saw no gains at the top end.

With the GT2560 TIP is around 15 psi through the midrange, ramping up to 22 psi (not spiking) (IIRC).

I have notes somewhere...

Exhaust backpressure is ~ 4 psi with a metal cat. 2.5" exhaust. The cat is responsible for about 2 psi. With a hi flow ceramic cat, its drop was 3 psi for a total of about 5.

kday

I have a handfull of surplus 0-50PSI industrial sensors, one of which is for the coolant pressure. I was thinking of another for exhaust backpressure, especially before I pulled the trigger on a new 3" setup. I think my current cat may have pooped the litterbox, and it would have been interesting to do a before/after. But I am going to put the new exhaust (w/ metal core cat) on at the same time I put in the new motor.

I also have a bunch of thermocouple amplifiers I built, I may use them to measure the temperature drop across the intercooler. But it might be better to just use a second IAT sensor before the intercooler.

The compressor speed sensor is the one I'm looking forward to seeing most. I think that will be pretty cool, particularly for seeing how ignition timing affects spoolup.

AlexO35

iTrader: (7)

How are you guys measuring / planning to measure turbo speed? Using the Garrett setup? Or some other way?
--Alex

JasonC SBB

http://www.atpturbo.com/mm5/merchant...egory_Code=BCS
Yowza, $425!

It would be super cool to integrate it with boost control, and the algorithm can detect a boost leak.

kday

The sensor and signal conditioning hardware can be ordered without the gauge, which reduces the cost a bit. (I like gauges as much as the next guy, but they attract too much attention, and in this case it's pretty useless anyway.) It requires machining of the compressor housing too.

The boost leak detector algorithm is a great idea, I hadn't thought of that. I'm curious how the wheel speed would improve a boost control algorithm, aside from being a failsafe. Please elaborate...

bigwig

Honestly, one of the more important ones not on your list is exhaust manifold pressure compared to intake manifold pressure. 1:1 pressure ratio makes a HUGE difference in making power.

codrus

I've always wanted a tach for my turbo, but $425 is a bit steep. I wonder what it'd take to DIY that...

--Ian

kday

It would take a lot more than $375 (cost of "pro kit" without gauge) worth of R&D to get something working reliably. Given that it's a niche product I think the price is actually pretty reasonable.

JasonC SBB

?? I measured TIP (turbine inlet pressure) and posted the numbers above.

1:1 pressure ratio seems unachievable with most streetable turbos (e.g. those that make say 8 psi by 3000 RPM). Streetable track cars, perhaps.

Do you know of any turbo setups that have that magical 1:1?

JasonC SBB

Given the likely sales volume, I agree it's not unreasonable.

Perhaps if more people buy them, there will be competition and the price will come down.