The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support JNZ Tuning
Please Support Morrison Fabrication

Logging Fuel Pressure

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

We're on Boost

15+ Year Contributor
2,113
852
Aug 25, 2007
Seattle area, Washington
I've been wanting to log fuel pressure lately, so I started poking around on the subject about a week ago. I would be logging with my ecu+.
I found out some worthwhile info by sending an email to Prosport gauges and would like to pass it on, here it is:

Me: I want to buy an electric fuel pressure gage with a sender that puts out a signal I could log on one of the 0 to 5 volt analog inputs in my logger. Will the prosport electric fuel pressure gage sender give me a logable signal?

Prosport: You are going to have to use the premium series Fuel Pressure gauge as that has a 0-3v output signal that you can tap into.

Me: Do you mean I would tap into the green wire that goes to the sender? Or is there an actual port on the back of the gage that is made to be an analog output to a logger?

Prosport: You would have to tap into it, no normal output signal o the gauge.

Me: I can do that, no problem. Wouldn’t I be able to do the same thing with the less expensive gages though, I mean the “performance series” gages?

Prosport: no, because that sender on the performance series reads Ohms.

Me: Ok, that makes sense. Thanks!


So there you have it, one possible way to log fuel pressure. You would prefer the input on your logger to be a high impedence input so it doesn't throw off the reading to your gauge. Logging just plain old voltage doesn't tell you what the fuel pressure is directly. So before I install the gage on the car I will set it up on the bench with a fitting to an air compressor and read the voltage at several different air pressures. That will be my "calibration" table to convert volts to pressure. The "premium" gage with sender is about a hundred bucks.
BTW what got me thinking about this is a problem with a lean condition under load ever since installing my Walbro. I'm thinking the problem is with the Walbro install. So I might just do that different to see if it "fixes" the problem. Still, I've always thought it would be nice to be able to log fuel pressure, even before this problem happened.

Would appreciate comments if anybody has tried this or done it a different way!

Gary
 
You don't need th gauge, and ECMLink does indeed convert to signal to PSI.
 
You don't need th gauge, and ECMLink does indeed convert to signal to PSI.

I'm looking at the autometer install instructions for a fuel gauge. Looks like there are 3 wires to the sender but they don't say which wire is which.
Is it right that there are 3 wires to the sender? Are they +12v, ground, and the 3rd wire would go directly to the input on your link?
Are there better instructions somewhere that are just for the sender?
 
Where would such a thing be hooked up to?

Well the 1/8" pipe threads will screw right into pretty much any aftermarket fuel pressure regulator gauge port. But if you are talking about the electrical end of it and if you are talking about DSMLink, I don't have a clue because I don't know anything about the hardware aspect of DSMLink. All I know is it is listed as a loggable parameter ( AMtrFuelPress ) here:
dsmlinkloggableparams [ECMTuning - wiki]
If you have a hardware logger like me (ecu plus) then you would have some inputs that are made for taking a 0 to 5 volt input, hopefully with a drop-down list of senders that it can translate from volts to the appropriate unit like psi. I've got one input where you can set up your own translater function, as long as it is linear. You key in the 2 endpoints of the line and it figures out everything in-between

By the way, Deftracing.com got some more info for me on the Prosport fuel pressure sender (the "premium" series one). They gave me a table of numbers for translation from volts to psi:

0.5v=14.5 psi
1.0v=29.0psi
1.5v=43.5psi
2.0v=58.0psi
2.5v=72.5psi
3.0v=87.0psi
3.5v=101.5psi
4.0v=116.0psi
4.5v=130.5psi
5.0v=145.0psi

As you can see, this is a linear function all the way through.
This Prosport sender is available by itself without the gage or any install kit for about $34. The sender with gauge and wire harness is only about $100. The autometer sender is $100 all by itself.

Gary
 
Last edited:
Well I hooked up a Prosport premium sender and got nothing but gibberish voltages from the output.
The "tech" people at Prosport told me to power it with 12 volts. I asked them twice if they were sure about that, because the other pressure senders on the market for the most part are 5 volt powered. They insisted 12 volts. So I hooked it up to 12 volts and got just stupid erratic voltages out of it.
I have no real conclusion as to whether these are good senders or not because for all I know, they might work just fine when powered by 5 volts. The one I have is a dead potato now, it doesn't work on any voltage.
Another possibility - it might have been a defective sender. I'll never know because they don't want to replace it. I would have to buy another one and I'm not going to do that. First they said it would work without the gauge and told me to hook it up to 12 volts, then they said no replacement because they have never tried to use one without the gauge before and maybe it won't work that way. Great.
The other possibility I suppose is that they might have told me a wrong pin-out. All they really told me was how to hook it up by wire color, but the colors they are thinking are the wires that go from your car to the gauge, not to the sender. It could be that the harness they sold me has a different color code than the wires to the gauge and they wouldn't even know.

Well whatever. I would just say, if anybody wants to try one, try it on a 5 volt power source!
__________________________

While I was looking around at the different senders available I found 3 that all look like they should be very good:

Autometer #2246
Honeywell # MLH100PGB06A
AEM 30-2130-100

The AEM has pretty good documentation, is the most expensive, looks like a good unit.
The Honeywell has excellent documentation, is about $20 cheaper than the AEM, looks like a good unit.
The Autometer has practically no documentation, but has very good tech support by email and even I think by phone. It is a few bucks cheaper than the Honeywell. Looks like a good unit.

I ordered a Honeywell from DigiKey. Hopefully will be testing it out in a few days.

By the way, all these senders take the same electrical plug. It is a GM TPS plug, with the 3 pins in a triangular pattern. You can buy these plugs all day long for under $10 each. They are easy to find online and I think the local parts houses sell them too. It is commonly called a GM Delphi / Packard Metri-Pack P2S Sensor Connector. P2S means "Pull To Seat".

http://www.bmotorsports.com/shop/product_info.php/cPath/109_130_133/products_id/502

http://www.diyautotune.com/catalog/tps-pigtail-gm-pin-style-p-100.html
 
Last edited:
I agree, it's better to skip the gauge and use a pressure sensor connected to the ecu. I used to have a mechanical fuel pressure gauge and there's no way to safely watch it while doing a WOT pull. W T F was I thinking? :)

Yeah, I think I'm going to do a wideband the same way - no gauge. Just the basic LC1-with-sensor-and-software kit which you can buy from MAP for $180.

... how much does that sender cost, and where is a good place to pick one up?

Autometer, usual price is about $100 and lots of shops sell them, Jegs, Summit, etc.

Honeywell price is $106 from Digikey.
AEM sensor prices are a little higher, but they offer a nice kit which comes with the plug with leads so you don't have to mess with that.

I got my Honeywell MLH100PGB06A, tested it in a rig with a Tovic mechanical gauge for comparison, using compressed air from my air compressor which has another gauge on it. That Honeywell sender works great and has never done anything weird, very solid piece of equipment there. I put together my electrical connector with Pre-crimped leads and plug from DIY. I gather they are the megasquirt people. By the way their leads were crimped a lot better than what they show in the picture on their web site :thumb: perfect actually.

Screwed the Honeywell into the gauge port of my Fuellab FP Regulator and wired it up. The 5 volts I got from the stock MAS plug, since I am not using the stock MAS anymore, but the harness is still out there. I have that in my notes as terminal #4 on the harness plug that would normally plug onto the stock MAS. The wire that goes to terminal #7 on that same plug was lying there doing nothing, so I used that wire to take the signal to my logger input. This is cool, I haven't had to make any new holes in my firewall yet! I'm sure that will change.

Did some logging with the sensor in the car, now I know what my fuel pressure is, not only at full boost and everywhere else, but I also can tell how quick the fuel pressure response is to throttle changes. Looks like fuel pressure lags behind throttle transients by only about 1 logger sample, which is 1/25 of a second. Better than I would have thought possible. I'm going to look at that a little closer though, with expanded windows.

When I get a chance I'll post some things about "0 offset" or "0 error" which is the only "error" I found to be of any significance with the Honeywell sensor. Temperature drift and ratiometricity error and all the other possible errors all turned out to be like nothing as near as I can tell.
 
how much does that sender cost, and where is a good place to pick one up? I don't see pricing on that link.

Gee, a google search for "Autometer 2246" brought up a ton of prices.
 
The way you interpret the output of the sensor is basically like this:

First of all there is supposed to be a difference of 4 volts in the output, between the extreme ends of the sensor range, which in this case is 0 to 100 psi gauge pressure (not absolute - so you don't measure vacuum with this sensor). 0 psi is whatever atmospheric pressure is at the moment. There is a tiny little hole in the side of the sensor body that vents it to atmosphere for that purpose.
Anyway, back to the 4 volt span, the 4 volt span is supposed to start at 0.5 volts (0 psi) and go to 4.5 volts (100 psi). You can easily check the voltage output at 0 psi just by hooking the sensor up electrically but just have the sensor out in the air. The voltage you get like that probably won't be exactly 0.5 volts. The difference between 0.5 v and what you actually get is called the "0 error" or "0 offset".
My logger measured the voltage as 0.53 volts during this 0 psi test, so no matter what the book says, 0.53 volts at my logger is 0 psi.
I had previously done a test on the 4 volt span. For that however, you don't have a guaranteed accurate reference pressure at 100 psi or any pressure other than 0. All I could do was compare to my Tovic mechanical fuel pressure gauge and another cheaper gauge that is on my air compressor. It did compare very well though. I measured sensor voltage at about 20 different air pressures, taking the air pressure reading from the Tovic at each point. Then I keyed the 20 or so points into a CAD system and put a best fit line through the points. Point scatter from the line was very small, about .04 volts max, which equates to about 1 psi. The best fit line was almost dead parallel to the theoretical line, only about 0.5 psi off from parallel at the 100 psi end.

OK, so here's what it boils down to. Whatever your logger measures for voltage at 0 psi is what you go with for 0 psi. Add 4 volts to that to get the 100 psi point. A straight line between those 2 points gives you the psi at any other voltage.

So in my case, I have 0 psi at 0.53 volts, and 100 psi is at 4.53 volts.
25 psi would be at 1.53 volts, 50 psi at 2.53 volts, and so on.

In summary I would say that the "0 error" can be significant, you should measure it if you want to have confidence in the accuracy of your logs. The 4 volt span error however seems to be very small and I just went with 4 volts for the span.

BTW, I forgot to mention that before I ever hooked the sensor output up to my logger I did all this testing on it with my Fluke 87 which is a really accurate voltmeter. When I did the 0 psi test with the Fluke, it measured 0.575 volts. That would equate to about a 2 psi "0 error" from the theoretical spec. That is really why I say that the 0 error can be significant. My logger reading only 0.53 volts under the same conditions, well, I suppose the voltage measuring stuff in my logger is just not as accurate as the Fluke. If the logger inaccuracy was off in the other direction it could be reading 0.62 volts at 0 psi! So really, this correcting the 0 error thing, it is not only correcting the 0 error of the sensor itself, but also for any inaccuracy in the voltage measurement of your logger, and for the voltage drop in your wiring that goes through the firewall.
 
Last edited:
Ever wonder how fast your actual fuel pressure responds to your right foot?

You must be logged in to view this image or video.
[/URL][/IMG]

This was done sitting in the garage in neutral, blip from idle.
In this pic I'm logging fuel pressure in the "Boost" window. Actually the psi numbers in that window are 1/2 of my fuel pressure. I had to scale it like that because the setup for that window only allows up to 50 psi max. So double whatever number you read on the vertical scale in that window to get the actual fuel pressure.
Anyhow, it's fast man! The logger samples each quantity once every .04 seconds. I have the cursor on the very peak of the TPS blip (53.24 seconds). You can see that the fuel pressure ramps up very fast and hits max almost simultaneously with the TPS, within .04 seconds. Then it lags a little on the way down. But it's on the way up that it really counts. This whole TPS blip lasts less than 1/3 of a second.
Just for clarification, the fuel pressure to the left of the blip is 16.8 X 2 = 33.6 psi (my normal idle pressure). The fuel pressure at the peak of the blip is 21.2 X 2 = 42.4 psi.
The FPR is a FuelLab.
 
Reviving this thread because it seems like a decent place for the discussion. I have a 1G with Link, SD (IAT and MAP) and a WB. Which, I think, means that all of the open pins that I might use to log FP are taken up. Is this correct? Are their any other options that I am not thinking of to log FP?
 
Reviving this thread because it seems like a decent place for the discussion. I have a 1G with Link, SD (IAT and MAP) and a WB. Which, I think, means that all of the open pins that I might use to log FP are taken up. Is this correct? Are their any other options that I am not thinking of to log FP?

If I recall correctly, you could use:
Pin 53 - EGR Control Solenoid Valve
Pin 57 - Fuel Pressure Control Valve
Pin 105 - Wastegate Solenoid Valve

I would use Pin 53, as the EGR is pointless.
 
Thank you for replying dexterhollad04. I should have been more clear when I typed my message. Because I am using SD and I have a WB, I am already using those three for WB, IAT, and MAP. So I am looking for another option or suggestions regarding how to set it up differently.
 
Another member gave me a great idea about this. You can always, for example, make the wideband wire detachable (think spade connector or something) so you can log fuel pressure instead of wideband for however long you want. If you feel the need to log fuel pressure, just switch it in for the wideband for a couple minutes.
 
Last edited:
Thank you for replying dexterhollad04. I should have been more clear when I typed my message. Because I am using SD and I have a WB, I am already using those three for WB, IAT, and MAP. So I am looking for another option or suggestions regarding how to set it up differently.

No problem. I was assuming that you were using pins 8 and 16 for IAT and MAP, respectively. If you're using the DSM Link plug-and-play speed density harness, it uses those pin locations. That would leave the aforementioned pins free (sans which ever one you use for logging your wideband).
 
Reviving this thread because it seems like a decent place for the discussion. I have a 1G with Link, SD (IAT and MAP) and a WB. Which, I think, means that all of the open pins that I might use to log FP are taken up. Is this correct? Are their any other options that I am not thinking of to log FP?

Thanks for the question because I have not tried logging fuel pressure with DSMLink yet, but I've wondered about it. The logging I did in 2010 was with ECU+ which was a good and very accurate logger. I've actually thought about partially re-installing it just so I would have a few more channels to log on.

I also noticed that the picture I had posted in post #18 has died, so I'm going to re-post the same pic here.
The horizontal axis is seconds:

You must be logged in to view this image or video.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top