Fuel Rail Pressure Sensor Calibration?

[rbattelle]

I know this is a long shot, but does anyone out there know the calibration for the rail pressure sensor? Or the OEM supplier? I asked Cummins and they said "ask Bosch". I have emailed Bosch 4 times over the past several months, and they completely ignore me.



I ask because I'd like to make a rail pressure gauge.

 

[kstehnacs]

rail pressure?

Fuel leaves the lift pump at 9. 5 to 11. 5 psi, goes to the hi pressure pump, and is sent to the rail/accumulator between 4351 and 23,206 psi. I wouldn't even think about putting a gauge on the hi side.

 

[HoleshotHolset]

Doesn't the ECM monitor rail pressure?



Perhaps you could view the pressure with an OBD-II scanner...



I would never even dream of installing a non-OEM transducer for pressure that high.



Matt

 

[dan.poitras]

Rbattelle,

I can tell from my cummins wiring diagram

(fuel rail press. sensor specs)

0 psi = . 50 voltage return (VDC)

5801 psi = 1. 39 volts

10153 psi = 2. 06 volts

14504 psi = 2. 72 volts

20305 psi = 3. 61 volts

26107 psi = 4. 50 volts

 

[rbattelle]

Originally posted by dan. poitras

Rbattelle,

I can tell from my cummins wiring diagram

(fuel rail press. sensor specs)

0 psi = . 50 voltage return (VDC)

5801 psi = 1. 39 volts

10153 psi = 2. 06 volts

14504 psi = 2. 72 volts

20305 psi = 3. 61 volts

26107 psi = 4. 50 volts

THANKS DAN!!!!! I have all the service manuals... where did you find that?



For you others... I don't intend to install a pressure transducer in the rail, because there's already one there. That's why I asked for the calibration (which Dan provided).



BTW, 30000 psi is not particularly difficult to measure. I know of at least one place that has lots of very accurate transducers available with up to 30000 psi. And you can get them in all sorts of configurations (gage, absolute, bar, psi, etc. ).

 

[JNutter]

This reminds me of the trick guys used to do their 7. 3 power strokes. They'd put a resistor (10,000 ohm?) across the oil pressure sensor on the head (probably equivalent to the sensor on our rails) and this would fool the computer into thinking the pressure was low - which would make it compensate by boosting the the fuel pressure. I had a friend that did this on his 2000 PSD and he said it gave a very noticeable boost in power.



I really don't want to experiment on a truck with less than 5000 miles on it, but all of the sudden making my own power module seems possible, for a fraction of the cost. Looking at the chart, a resistor in series with the sensor would reduce the voltage seen by the computer. This would make the vehicle computer think the rail pressure was low. The vehicle computer would raise the rail pressure to compensate. That's pretty much what all the pressure boxes do, right?



Another simple circuit could be put on the boost sensor to prevent voltages higher than a certain value to be output to the trucks computer. The circuit could still let through the voltage if it was lower than the pre-set value, which would leave a signature in the computer that appeared to be normal boost values - with a flat plateau at max boost when you were all the way into the throttle. Does anyone know how high of a manifold pressure it takes to trigger an over-boost code? How about the calibration of the MAP? Is the same information about the pressure/voltage relation available for the MAP?



Could there be more to the pressure boxes than I imagine? Is what I outlined really all there is?



A duration box, on the other hand... My hat is off to TST as I contemplate everything that must have went into that box. That box must be 100 times as complicated as the pressure boxes...

 

[rbattelle]

JNutter,



Everything you say certainly sounds sane, but like you I'd be unwilling to be the guinea pig on it!

 

[JNutter]

What do you have in mind for your rail pressure sensor? A high input impedance A to D converter of some sort, with a cheap micro and an LCD display? You probably know electronics, or you wouldn't even bother with this.



How about the connector on the rail sensor? Have you identified who makes it yet? I'd guess that you'd want a male and female to make a pass-through connector for your sensor - to keep it weather tight.

 

[dan.poitras]

rbattelle

I got these specs from the wiring diagram sheets they gave out when I took the cummins course. Cummins bulletin#3666483-01

It has all sensor test specs etc. Mind you the diagram is for a cummins ecm but the sensors are the same.

Dan

 

[JWerkheiser]

Just be really careful raising the rail pressure. Too much for too long will blow out the fuel rail. I don't know what "too much" or "too long" is, but I don't think I would want to experiment.

 

[rbattelle]

JNutter,



Well, I'm not an electrical engineer, but I've done a pretty fair amount of data acquisition and some electronics design. My idea is much simpler than you describe. It's clear the transducer is 0-5v (pretty typical for transducers), and it's pretty linear (of course). So I just want to make up a voltmeter with a custom background that sweeps from 0 to 30000 psi as the voltage climbs to 5v.



One trick, as you mention, will be to find a way to splice into the existing connector without disturbing the signal to the ECM. A very high impedence device (such as a voltmeter) should keep from disturbing the ECM signal too much, without the need for any external resistors (I hope?). But how do I splice my wiring into the sensor in parallel? I have not yet looked closely at the connector (too cold here!). Ideally, I'd find the connector manufacturer and get an adaptor to go between the sensor and the wiring harness. Is that how fueling boxes connect to the sensor? If so, I could ask them to sell me one of their connectors...



It's pretty apparent that you've a background in electronics, so any opinions/suggestions are much appreciated.

 

[JNutter]

I am probably not as experienced as you are with electronics, and I've been steadily forgetting what I knew over the past year becuase I've switched to mechanical design full time.



The good news is that they hired my cousin to be the tech when I stepped up, and he actually went to electronics school. I had a couple years of practical electronics in college, but it wasn't as much as what he got in tech school. He's a lot better at circuit design than I ever was. He also likes Jeeps and diesels. Between the two of us, we stand a reasonable chance of figuring out what the connectors are. Hopefully they are available from Digikey or some other place with a reasonable minimum purchase. We'll try to take a look at them over the weekend.



The best way to go would be to get a male and female version of that exact connector and make a pass through, attaching pin 1 to pin 1. Then you'd just need to crimp an extra wire into the ground and signal pins. It should be pretty to figure out which pins thoose are by backprobing with the engine on.



I would avoid cutting into the wires at all costs. If I was a Dodge tech and I saw that, I'd have to cancel the warranty.



I mentioned the pressure box idea to my cousin (he called to make sure my truck was OK after the dealer fixed my PS hose), and right away he wanted to tie it into the EGT thermocouple so it would de-fuel if the EGT gets too high His idea is to put an amplifier with a gain of less than one on the pressure side so that the rail pressure would read the correct voltage at idle but less than actual pressure at the higher levels, with a very gradual change in between. He thought that might work for the MAP too, but I'm worried that it might make the computer think the boost was low enough to de-fuel. This all sounds good, but I really don't think I'll do it. It's fun to think about though



John Nutter

 

[rbattelle]

John,



You and I are now on slightly different tracks. I don't want to raise fuel pressure at all, I just want to monitor it with a gauge in the cab (because I'm obsessed with monitoring things).



However, if you happen to find a source for those connectors, let me know. And I agree 100% with not cutting the wiring for the sensor, not only because of warranty reasons, but also because I figure splicing directly into the wiring might introduce enough change in resistance to have an effect on the ECM's ability to accurately read rail pressure.



I forgot about Digikey! I really need to pull the electrical connector and take some pictures of it and start searching through some catalogs.

 

[boonsur]

I'd worry about simply lowering the signal, trying to make the computer raise the fuel pressure. I'd guess that with the pressure low, and either a high boost or EGT reading, it would sense a problem and shut-down, throwing codes along the way. The boxes also intercept the other signals, and output them so the computer thinks everything's fine. One of the boxes, allegedly, gives a constant reading to the computer (say boost of 8 psi, always). A dealer can look at these parameters and figure out that the truck's had an "enhancement".

 

[KBennett]

It's interesting that this subject came up.

This last Saturday the local RMTDR chapter had a meeting at the ATS shop. A guy from Cummins was there and he said that the factory engineers were able to raise the rail pressure by putting a resistor on the rail pressure sensor. He did not say whether it was in series or parallel or what value they used and what pressure level they safely achieved. If you are interested in a little digital gauge check out: Red Lion

I am a distributor for these guys and they make a good product. PM me for pricing.



Not to get off the subject but the Cummins guy said that he liked paper air filter elements rather than foam/oil. He had seen turbo failures from the oil being sucked in by the turbo and getting on the blades creating an imbalance. In his opinion the air cleaners on our trucks should be about 4 times bigger than what they are.

 

[silver03]

boonsur, a dealer cannot check for "enhancements" with a scan tool. as far as the computer knows, the sensors and the signals are good regardless of what is actually going on. (unless it exceeds some predetermined parameter like boost too high)



also, giving a reading of 8psi always will not work since the gauge is an abosulte sensor not gauge. what ever it is before starting is stored and boost "adds" voltage to the signal.



putting a resistor across two of the pins will give you the result of adding more fuel pressure.



i would not worry too much about the hype about blowing fuel pressure systems, these systems are designed to handle a lot of malfunctions, tolerances, etc and i do not see how a motor that could potentially be stuck on (pwm=100%) can blow a motor. personally i feel it is dealers scaring customers into there products.



don't bother looking in digikey (its not in there), the connector manufacturer is right under your nose.

 

[rbattelle]

Originally posted by silver03



don't bother looking in digikey (its not in there), the connector manufacturer is right under your nose.

What do you mean by this?



KBennett-



Does Red Lion carry analog panel meters? I want one that I can disassemble and make a new backing that reads in PSI instead of volts...



I want something like this: panel meters

 

[rbattelle]

Probably more info than anyone wants...

I think I may have found a good panel meter in 0-5v range. It's made by Simpson, and can be found in their DV Voltmeter catalog, "Round Style", DC Voltage. Part number is 08870.



This Company carries it for $72.



The question is: is it possible to disassemble this meter and replace the background without damaging anything?



Attached is an example photo of the meter.

 

[KBennett]

Sorry for the delay in getting back to this thread, I kinda lost track of it.



Red Lion does not make any analog meters, only digital. Simpson makes good meters but the movements are meant more for sitting still in a panel, not the bouncing an automobile or truck would give it.

If you have to go analog I would get a low end amp or volt meter from the auto parts store. To modify a ammeter you will need to take out or cut the shunt. The coil in the movement may still put too much load on the circuit. You could measure the resistance with an ohmeter to find out the impedance. I don't know what the typical resistance is on the coil but if it is too low, around 1-5 Kohms, it might throw off the reading to ECM. This is where amplifiers come into play and if you notice on the Red Lion data sheet the input impedance of their voltmeters are really high at 1 megohm putting a very small load on the device it is sensing.



Another alternative is to give Autometer a call. They may have something that will be plug and play.

 

[KBennett]

One more:



If you still have that onboard computer you could add a data acqusition card or serial converter that can convert voltage signals to something the computer could read. This value can be read by software and displayed on your screen in any format including analog.