Hummer H3 low fuel pressure, high LTFT at idle, rich upstream O2 sensor reading

Hello. I am trying to diagnose a "rough idle" issue.
2010 Hummer H3 3.7 i5, stock.
I have been dealing with a low fuel pressure, high LTFT at idle, rich upstream O2 sensor reading.
- Low Fuel Pressure: Upon starting engine, the reading states 57/58psi, then after 10 to 15 second it slowly drops to 43psi, and stays there, unchanged even if pushing gas. Everyone I have consulted with states that it should be 57psi and stay there.
- LTFT at idle reads (now) around +11 to +14, but the warmer it gets, it seems to climb a bit. I said "now", cause before I did some work, it was going from +10 to +25 and even +30. Now, when I am driving at about 30/40mph it stays at around +7, and highway speeds 65 and above, it drops to 0, or dips into negatives, or goes no higher than +7 or so. The STFT goes normal +/-5.
-Upstream O2 sensor shows a nice wave; however, the range is from high 800 to 400, which my research indicates that it is in the rich side of readings. The downstream sensor seems to read a good cat on a flat line at about 150 + or -.

I had put a new fuel pump due to a misdiagnose and it was putting out 43, I put the old pump back in and reads 43psi as well, so it does not seem to be the pump. I just installed a new fuel pressure sensor, and it did not change things. I also checked fuel lines but everything flows well.

I have an inactive p0420 code, most likely due to a gutted cat when I bought it. I put new exhaust manifold a few months ago.
I was having a p0171 code, but it vanished after I put a new air filter enclosure, the old one was cracked.

A mechanic I have been working with had the Hummer for like three weeks and they "could not figure it out", yet charged me $200 for diagnostic (never going there again). All the said was repeat what I already new.

Tomorrow I'll be checking the FSCM, doing some diagnostics, but since a programmed FSCM is about $250, I just do not want to purchase one if that is not the problem and not after I try other cheaper solutions.

I know that the LTFT sometimes indicates leak post MAF, and mechanic said that they checked for leaks, but I do not beleive they did a good job. Should that be my first action?
I know I might be leaving something out, but my brain is tired from thinking so much about this.
Any help/advice/ideas you can throw my way are greatly appreciated.

- Low Fuel Pressure: Upon starting engine, the reading states 57/58psi, then after 10 to 15 second it slowly drops to 43psi, and stays there, unchanged even if pushing gas.

Are you using a manual gauge to measure fuel pressure, or are you using the fuel pressure sensor on the vehicle?

To know if the pressure behavior you're seeing is normal, you'll need access to a scan tool that can talk to the Fuel Pump Control Module. The FPCM scan data will show actual pressure, desired pressure and pump trims, among other things. If the actual pressure is meeting desired, and that desired pressure happens to be 43 PSI, then I'd say it's all good.

Everyone I have consulted with states that it should be 57psi and stay there.

With mechanical returnless systems, I'd agree! But this being electronic returnless, you can't really say that. It's not about if the pressure is meeting spec, it's if the pressure is meeting what the module is trying to achieve at that time.

Upstream O2 sensor shows a nice wave; however, the range is from high 800 to 400, which my research indicates that it is in the rich side of readings. The downstream sensor seems to read a good cat on a flat line at about 150 + or -.

If you haven't already, you might try narrowing down your PID list to watch just a few PIDs, including the O2 sensors. You might see the upstream O2 going closer to full range that way.

Does the upstream go all the way to zero if you create a large vacuum leak?

- Used both, a regular gauge leased at autoparts store, and the info coming through OBDII scanner.
- Yeah, I do not have a decent scanner, just been working with a bluetooth OBDII scanner and different apps that I downloaded. I'd like to buy a decent scanner tool, or see if I can borrow one since I want to learn to read the data and troubleshoot that way. But hey! Thanks for the suggestion, cause I was unaware of the desired values versus actual values.
- I see. Yeah, every single article or video that I came across of stated that 57psi is the running pressure, even my mechanic stated that the values should be 55 to 57psi.
- Yup. When I look at the data, I try to do two at a time, like stft and ltft, then o2 sensor 1 and 02 sensor 2, then fuel pressure, son on, all separate to not mess with the ranges and graphs.

I am thinking that it is something electrical, bad sensor, bad control module, because after I re-installed the original fuel pump, it was running more smoothly, the ltft had dropped to nearly zero, and had great acceleration, but then sunday I put new spark plugs, and it is all shaky again. The darn thing is possessed!

- Used both, a regular gauge leased at autoparts store, and the info coming through OBDII scanner.

Sweet. Did they both happen to agree on the pressure (more or less)?

Samponiato wrote:

- I see. Yeah, every single article or video that I came across of stated that 57psi is the running pressure, even my mechanic stated that the values should be 55 to 57psi.!

FWIW, I dug this out of service info while looking for a better idea of what fuel pressure is supposed to be:



After seeing this, I'm gonna say that the fuel pressure readings you're seeing are entirely normal. Could be wrong! :silly: But I don't think it's chance that your readings fall right in line with this chart.

Oh man, thank you! This is the first time that I see that fuel pressure table. Is that something open to the general public, or just for techs/shops?
Yes, the regular gauge and the scanner data showed the same psi's, at ignition, then running. So it starts in the 57psi, and after a few seconds it drops to 43psi and stays there.
Everything seems to point that it is running rich for some reason (from the extremely high LTFT). Thinking about purchasing a MDI interface and reprogramming all the modules, cause something tells me that the mechanic that worked on it before I purchased it might have been a crook who did some wonky things. In the state that I bought it, there is no way it was not throwing any codes...

Samponiato wrote:

Oh man, thank you! This is the first time that I see that fuel pressure table. Is that something open to the general public, or just for techs/shops?

That table should be available from most any aftermarket service information provider (Mitchell ProDemand, Identifix, Motologic, ShopKey, whatever). All those are going to be $100+/month.

An alternative is MitchellDIY, which is much cheaper and vehicle specific. You can also get a three day subscription to service information direct from GM for $22:

www.acdelcotds.com/subscriptions

That'd be the way I'd go if I were in your shoes.

Everything seems to point that it is running rich for some reason (from the extremely high LTFT). Thinking about purchasing a MDI interface and reprogramming all the modules, cause something tells me that the mechanic that worked on it before I purchased it might have been a crook who did some wonky things. In the state that I bought it, there is no way it was not throwing any codes...

Before going to the trouble of getting an MDI and all that, you can always check your calibration ID's and CVN's against those listed on GM TIS:

spsinfo.ext.gm.com/sps2info/

You may need to make a GM TDS login to get to that page. It also seems to be broken right now. GM is always messing with a good thing...

If you can make that page work, you can compare the calibration ID's/CVN's against yours. If they match, then it's unlikely anyone has been tuning stuff. If they don't match, red flag. You can use your scan tool to get the CID/CVN info via Mode $09. I have an example of a tuned Suburban here:

www.scannerdanner.com/forum/post-your-re...ough-idle.html#26386

That page never works. Checking calibration ids like it used to. So now I just keep 1 available vin slot to check them if I’m wanting to look.

Cheryl wrote:

That page never works. Checking calibration ids like it used to. So now I just keep 1 available vin slot to check them if I’m wanting to look.

So what happened to the old TIS lookup site? :huh: I used it several times last week. Went to find it today, no go.