Cylinder Pressure Testing notes.

McRat

McRat

Diesel Hotrodder
Aug 2, 2006
11,249
26
38
64
Norco CA
www.mcratracing.com
The 502 HP tune made a maximum of 220 bar so far in testing, and confirmed what "Fingers" told me: The Optrand sensor is relative, and not absolute.

With the sensor static (14.7 PSI atmospheric pressure), engine not running, it reads .52 volts. But when the engine is running, the static reading (exhaust valve open) is .48 volts.

Well that can't be right, since with a rating of 0.00131 volts per PSI chamber pressure, that would mean when the engine is running, that would be about -15 PSI absolute, or better than a perfect vacuum. Truth is, it should be at least .56 volts or far higher due to drive pressure.

This means we can't look at things like changes in exhaust scavenging with this sensor, and perhaps means our readings will never be more trustworthy than ±10%.

NEXT, if you want to watch the Injector #1 fire, you need to actually map out Injector #2, if you are getting the pulse from the Bale Connector. Since it's a capacitor based injection driver, you are off 90° in the pulse.

Still haven't hooked up a decent TDC marker, so I'm using the injector pulse, which is advanced most the time.
 
Fingers

Fingers

Village Idiot
Vendor/Sponsor
Apr 1, 2008
1,718
96
48
White Oak, PA
Capacitive discharge would not be keyed to crank angle offset from the input pulse. That is, 90 CAD from the pulse.

I'm home now, so I want to play with unhooking a cylinder's ECM command wire at the bail and seeing if that cylinder goes out or not. I need to map the what wires to what cylinders anyway.

I suspect the wire-cylinder scheme is an artifact of the same ECM being used on the gassers.
 
mde

mde

fuel injection is my life
Mar 17, 2007
396
0
0
Bucky State
McRat said:
The 502 HP tune made a maximum of 220 bar so far in testing, and confirmed what "Fingers" told me: The Optrand sensor is relative, and not absolute.

With the sensor static (14.7 PSI atmospheric pressure), engine not running, it reads .52 volts. But when the engine is running, the static reading (exhaust valve open) is .48 volts.

Well that can't be right, since with a rating of 0.00131 volts per PSI chamber pressure, that would mean when the engine is running, that would be about -15 PSI absolute, or better than a perfect vacuum. Truth is, it should be at least .56 volts or far higher due to drive pressure.

This means we can't look at things like changes in exhaust scavenging with this sensor, and perhaps means our readings will never be more trustworthy than ±10%.

NEXT, if you want to watch the Injector #1 fire, you need to actually map out Injector #2, if you are getting the pulse from the Bale Connector. Since it's a capacitor based injection driver, you are off 90° in the pulse.

Still haven't hooked up a decent TDC marker, so I'm using the injector pulse, which is advanced most the time.
Click to expand...

Pat don't forget the temperature coefficient , when the engine runs, the resistance of the sensor gets higher and changes your baseline. This sensors signal is usual: 0.5V = 0%, 4.5V = 100% of pressure. Also don't forget the accuracy classification of sensors. Example, if you have a class 1, that means +/-1% accuracy of 100%, so when you down by ambient pressure or your 0.48 or 0.52 V from 4V +/- 1% = 0.04V you see you will reach soon your tolerances, with the accuracy and temp's.

Try to get the TDC signal of the Cam Hall sensor, you sould get a nice signal out of them.

V
 
McRat

McRat

Diesel Hotrodder
Aug 2, 2006
11,249
26
38
64
Norco CA
www.mcratracing.com
I grabbed the Cam sensor (and crank) from the Bale connector, and they did not correspond to TDC according to the Cyl Press Sensor.

I'm now using the actual injector pulse and then figuring out what the advance was.

Not right, but better than nothing.
 
Last edited:
Fingers

Fingers

Village Idiot
Vendor/Sponsor
Apr 1, 2008
1,718
96
48
White Oak, PA
Lots to learn. Maybe I can help.

The begining of the "Gap" in the crank position sensor roughly lines up with cyl # 8 and #3. Number 3 also appears to be the cylinder that the #1 command wire controls.

#1 cylinder TDC is 25 teeth past the "Gap". But I have found the accuracy of the crank pulse to vary from engine to engine by as much as 6*. I Don't know where the variation comes from. Build, sensor position or reluctor position.

Here is a plot without an injection event and TDC marked. This is a hot engine, so the pulse is slightly shifted.

attachment.php


You can verify by looking at the pressure pulse on a cold engine while cranking without combustion. A warm engine shifts the pulse to the right because the air charge absorbs heat from the cylinder.
 
Last edited:
Fingers

Fingers

Village Idiot
Vendor/Sponsor
Apr 1, 2008
1,718
96
48
White Oak, PA
mde said:
Pat don't forget the temperature coefficient , when the engine runs, the resistance of the sensor gets higher and changes your baseline. This sensors signal is usual: 0.5V = 0%, 4.5V = 100% of pressure. Also don't forget the accuracy classification of sensors. Example, if you have a class 1, that means +/-1% accuracy of 100%, so when you down by ambient pressure or your 0.48 or 0.52 V from 4V +/- 1% = 0.04V you see you will reach soon your tolerances, with the accuracy and temp's.

Try to get the TDC signal of the Cam Hall sensor, you sould get a nice signal out of them.

V
Click to expand...

Volker,

These sensors are more like accelerometers than the typical pressure sensor. Probably a form of charge coupled device. They are intended to measure fast pressure pulses but sacrifice the static reference to do so. The pressure drop overshoot is an artifact of the amplifier not the sensor. It comes from the sensor needing to be recharged and the amp falling behind on the demand. It can be reduced with a better amp, but will always be there to some degree. The good news is the affect is predictable and can be corrected with software.

Both the Optrand and the sensor I use are temperature corrected devices. But even so, the temp swing of the sensor per combustion event is insignificant.


But, what do I know.....
 
McRat

McRat

Diesel Hotrodder
Aug 2, 2006
11,249
26
38
64
Norco CA
www.mcratracing.com
Primitive, but you are watching 3000 RPM with pilot on and a 3000us tune.

See the pilot, then the main. The "bulge" is the cylinder pressure.
 

Attachments

  • ex2.jpg
    ex2.jpg
    31.4 KB · Views: 55
Fingers

Fingers

Village Idiot
Vendor/Sponsor
Apr 1, 2008
1,718
96
48
White Oak, PA
Just under a 2400 PSI pressure pulse. That is in line with what Nick (DuramaxTuner) was seeing, though I would wager that his is coming in closer to TDC than yours. Really need the Crank Position Pulse.
 
Fingers

Fingers

Village Idiot
Vendor/Sponsor
Apr 1, 2008
1,718
96
48
White Oak, PA
Pat,

Do you have a graph just after a shift point when the fuel comes back in? That is when I see the highest pressures so far. Probably not as high as yours.

Just curious.
 
McRat

McRat

Diesel Hotrodder
Aug 2, 2006
11,249
26
38
64
Norco CA
www.mcratracing.com
Fingers said:
Pat,

Do you have a graph just after a shift point when the fuel comes back in? That is when I see the highest pressures so far. Probably not as high as yours.

Just curious.
Click to expand...

Wish I did.

Perhaps because that's the highest torque we see during a pass.

I tried to collect that info on a dyno, but the dyno screwed up.
 
V

vortecfcar

No longer a member
Jun 28, 2007
167
0
0
Keeping in mind this is preliminary, but the highest cylinder pressure I've seen so far has been at about 2900 RPM, I don't see over 2800 psi no matter if the truck is seeing 2800 uS or 3200 uS. It's clearly overfueled past 3000 uS (dual fueler - 23K). It's difficult to get readings exactly on the shift because cylinder pressure plots aren't read constantly (woud be an astronomical amount of data I'm sure). I'll keep plugging away at it.

keep in mind, my truck is running thick (C) head gaskets and de-lipped pistons so it's decompressed a bit.

Nick
 
You must log in or register to reply here.
Top Bottom