Fuel Rail Pressure Gauge is here for bombers

[DLeno]

All I'm saying is that we can't trust a very fast high resolution measurement, such as from a DMM or the SPA gauge, until we understand the AC component that may or may not be present on that wire. and the SPA gauge is particularly vulnerable to such noise if you depend on the peak detect feature. So we gotta understand the measured quantity before we claim measurement success.



Since learning that the sender contains an active linearization element I have been less concerned about its output impedance and more concerned about noise on the line. my own measurements with a 10K impedance DC gauge calibrated for to the sender suggests that measurements above 26,100 are not reliable. This is the data behind the conclusions in TDR issue 48, soon to appear. I used shielded twisted pair cable, separate signal and gauge grounds, and eliminated ground loops in the gauge ground by leaving the shield open at the sender.



But I have not played with the SPA pressure channels yet, except for the low side fuel pressure which I found to be very very pulsy even with 6 feet of snubbing hose. 1 second averaging works well for this, but the SPA gauge is so fast and has such granular resolution that its peak detector cannot be trusted to capture either a low or a high value for such a signal. I got peak pressure measurements into the upper 20s psi, which are probably correct, but not of interest or of value for the purpose of understanding the low pressure system. likewise, peak rail pressure measurements using the SPA gauge are suspect for the same reason, at least until we can characterize the signal we are attempting to measure.



--Doug

 

[SRadke]

heck ya, count me in... . the idea of a gauge that does both pressures would be awesome!!!!

Yea, what he said. With a plug and play fuel pressure harness to eliminate tapping and a reliable electric sender for the lift pump pressure I'd be in for a reasonable price.



-Scott

 

[DLeno]

The SPA pressure senders are awesome. At $120 a pop they better be. And they don't (so for, in my application) exhibit any of the problems that cheeper senders have. the low side fuel pressure system is very very pulsy and tends to destroy lesser senders. I hear of mechanical isolators breaking down; I hear of electric senders breaking down, but I don't hear of the SPA electric senders breaking down. and I've had mine on for about 9 months now. As I say, the only problem with the SPA gauge measureing low side pressure is that you must understand that the pressure is very pulsy, approaching 30 psi spikes, and that the SPA guauge is fast enough to follow them. so you cant use the peak detector. but you can use 1 sec averaging and that works well and with which I'm satisfied. we should find out shortly how/if the SPA gauge can be reliably applied to the rail pressure measurement. I'm suspicous that if we control the noise we can use averaging again, but not the peak detect feature. at least not without external signal processing.



the problem in measuring rail pressure with a sender not specified beyond 4. 5V is this: it is the region of interest (above 26,100 psi) that is the least dependable for measurement!



there may also be variation from truck to truck. BTW, anyone verify the calibration data for the 600s and 610s? I expect they use a simliar sender, even though it is a different part number from the dealer...



-- Doug

 

[SRadke]

... ... . the problem in measuring rail pressure with a sender not specified beyond 4. 5V is this: it is the region of interest (above 26,100 psi) that is the least dependable for measurement!

This is the part that I disagree with. I like duration, always have. I have a pressure box but I never use it over 1/2 power, usually less. The reason I would want a rail pressure gauge is to tell me at what power level the CP3/rail can no longer supply the fuel. Max pressure, although interesting and something I would like to know, is not my primary objective. That's my personal view.



Also, thanks Doug for the low pressure sender insight



-Scott

 

[DLeno]

I agree with you scott. I like my rail pressure gauge for what it tells me below 26,100, not for revealing the maximum rail pressure acheived by a given box. thats the part that is "of interest" for some.

 

[Matt400]

DLeno-



I don't understand the noise issue, why is it an issue for the gauge but not for the ECM? Are you concerned with outside induced noise or noise from the SPA?

I could understand RF being a concern on a gas application if the wire was near a spark plug wire or something like that. . I just want to understand so it can be addressed.

 

[rbattelle]

Anyone know the sampling frequency on the SPA gauge?



The Cummins service manual specifies that maximum rail pressure should be not more than around 20305 psi for a "normal" engine. Are we actually running higher than that stock?



Doug - have you checked the signal with an oscilloscope? Seems to me that would pretty much clear up all questions about the sending unit.



-Ryan

 

[Matt400]

What are you looking for in the signal? I could graph the signal with a graphing meter.

 

[rbattelle]

What are you looking for in the signal? I could graph the signal with a graphing meter.

Please don't tell me you own a handheld scopemeter. The last time someone told me they had a nice piece of electronic equipment it cost me $400 (DLeno knows what I'm talking about).



-Ryan

 

[DLeno]

hee hee hee. ok, so I'm waiting on my fluke 189 so I can actually make this measurement myself. But anyone with a scope or fluke 189 can do this. most DMMs cannot, but the 189 can display AC and DC components on dual displays. A (non 189) graphing voltmeter might not be fast enough, as its sampling frequency will be very slow compared to the types of errors we are concerned about.



I'm only trying to be cautious here and not claim to have a succesful measurement before we fully understand what we are measuring and the possible sources for error. hooking up a fast digitally sampled DC gauge onto this line may be a valid approach, but it may not be. I've been measuring rail pressure with an analog gauge for about 9 months now, and this tool has been the basis of TDR issue 47 and 48 articles on fueling enhancments. But my measurement is neither fast nor suspectiable to noise, since I benefit from the mechanical averaging of the meter movement and the noise-elimination of the shielded, twisted pair. I did that not because i thought the SPA method was bad; I just chose not to use it because doing so introduced more uncertainty which I did not want to deal with.



knowing that idle should be less than 5,000 psi and that WOT should be about 23,500 psi, I quickly determined that my gauge was accuate, at least to 100-200 psi or within the paralax error, needle inertia, etc. . good enough for what I needed at the time. But the next step is to produce a plot of DC and AC components from the 189. a scope would do it as well, yes.



Potential sources of error include:



1. external EM noise picked up by the signal wire, ground loops and other measurement errors caused by the way you hook things up. EM noise can come from the high speed CAN bus, the ECM, and most notably the 90 volt injector solenoid control signals spewing high frequency EM energy everytime they fire. in a case like this, you just don't know until/unless you measure or take precautions.



2. naturally pulsy behavior of the signal that an ordinary DC voltmeter would be blind to, and a fast DC gauge would be sensitive to. even if there is zero noise and you measure a squeeky clean DC output of the pressure sender.



3. signal conditioning at the ECM that a guage would not possess naturally (like an analog movement would) especially one as fast as the SPA. . A simple bypass cap at the signal entrance to the ECM would dampen high frequency noise. no I don't know the sampling frequency of the SPA; I just know it is fast enough to respond to the pulses present on the low pressure side, and guess what, those are the same frequency as present on the high pressure side and produced by the same thing (the CP3). low side pressure spikes acheive well over 300% of the average!



All I'm saying is that we dont' have a reliable measurement until we know the answers to these questions. so fluke 189 owners, unite and measure!



Ryan, yes we are running higher than 20305 psi. thats for us HO guys anyway. the SO guys might actually run there. The HO runs 1600 bar at WOT 3000 RPM, or about 23,500 psi.



So Matt, theres a good validation check: run your truck bone stock and see if you measure something close to 23,500 psi at WOT and 3000 RPM. thats assuming you have an HO. again, compare 1 sec averages with peak readings.

 

[Matt400]

Please don't tell me you own a handheld scopemeter. The last time someone told me they had a nice piece of electronic equipment it cost me $400 (DLeno knows what I'm talking about).



-Ryan

I have a Snap-On Vantage Digital Graphing Multi Meter

 

[rbattelle]

I have a Snap-On Vantage Digital Graphing Multi Meter

This website costs me a fortune. :-laf



I guess the pressure in the service manual I have (which is really for industrial ISB's) is not quite applicable to our pressures.



Just to be completely clear, what's the exact goal here? Do you guys want to measure average rail pressure, high spikes, or something in between?



-Ryan

P. S. Doug - my 189 shipped today from New Jersey... should be here in a couple days. Can't wait!!

P. P. S. Matt - I'm not even going to go to the Snap-On site and look at that graphing meter... my bank account wouldn't survive.

 

[Matt400]

Just to be completely clear, what's the exact goal here? Do you guys want to measure average rail pressure, high spikes, or something in between?

I think it will depend on the level of modifications. Myself am interested in pressures that could possibly push open the limit valve.

An example is the ramifier box I tested, a super nice strong power delivery that seemed to work real well. However If I were to choose to run this box all the time I would fix a way so the in cab power knob would not rotate to a pressure zone I am not comfortable with. With this gauge set up I could basically defuel based on pressure.

Others with more mods and big HP may be interested to see if the rail is loosing pressure under WOT runs. I remember reading a post a while back where the TST was capable of lowering the pressures below what it should be from long durations. Folks with stacked boxes could use the gauge to dial in duration, then pressure, then duration and pressure again till they find the best settings.

As for the lift side all I want to monitor is how it gets depleted under WOT runs and set a low level alarm in case diagnostics on the lift ckt is needed. Don't want to starve the CP3.



I did do a WOT run tonight on the way home but due to road conditions I had to back out. The SPA registered 19,500 (mine is an SO) with the box off. The gauge is very fast reading of all the pulses and at idle if I stare at it I will see anything from 4,200 to 5,200 but remember this is using a temperature channel where averaging time cannot be changed. The Boost channel should work much better.

 

[AGrubbs]

Been watching this thread with interest! I'd be up for one if it is not out of line in price. .



Arnie

 

[Matt400]

I'm not even going to go to the Snap-On site and look at that graphing meter... my bank account wouldn't survive.

Ahhh. .come on now, how about if I put the link righ here?

 

[Matt400]

Off the transducer I couldn't get any AC but off the SPA I was able to get . 002v



The cable SPA uses is twisted pair shielded with wrapped foil so if we could get them to add the connector splice that would be nice. Also they said the peak detect always runs at 100mS response time regardless of the average time settings, so it always finds the true peak. The average time according to SPA can be set as high as 25. 0 sec so this should help with the pulsey nature of the rail and the super fast gauge and give a more average rail pressure. Guess I will have to pop for a boost gauge to test this out because I cannot do those settings on a temperature channel.

 

[SRadke]

I would also be interested in purchasing a dual pressure gauge marked "boost" pressure and "drive" pressure after I go to twins. I can see real application for the twin guys who really should have 4-6 gauges to keep track of things and not having room for that many single gauges.



Matt, do you have an email address where I can reach SPA with such a request? Thanks.



-Scott

 

[Matt400]

I would also be interested in purchasing a dual pressure gauge marked "boost" pressure and "drive" pressure after I go to twins. I can see real application for the twin guys who really should have 4-6 gauges to keep track of things and not having room for that many single gauges.

Thats a good idea also, SPA had at one time a DG219 that was a Pressure / Pressure gauge. From what I know it never made it to production but maybe that one could be brought back and serve your need along with the Rail / Lift Pressure gauge by simply using a different sensor and setting but the same head. I don't know what kind of transducer could be used for exhaust pressures though on your drive side.



The nice thing about using a Boost channel for rail pressure is that it can be set for an internal high level alarm where a pressure channel currently cannot.

Also I am not sure if a pressure channel is able to increase the averaging time.



John from SPA in the UK is monitoring this thread to help with the possibility of bringing this all together. He is not a TDR member so can't post but will see what everyone has to say and I keep in touch with him.

 

[SRadke]

Thanks Matt. As for drive pressure, I'm sure a length of copper tubeing from the manifold would effectivly isolate the pressure sender from the heat.



Sorry everyone for hyjacking the thread... . back to the rail pressure!



-Scott

 

[Matt400]

Thanks Matt. As for drive pressure, I'm sure a length of copper tubeing from the manifold would effectivly isolate the pressure sender from the heat.

Sure, that could work well. How high of drive pressures do you look for?