P0171 Driving me Insane!!!

[Ecomike]

What I'm interested in is what the "fuel trim" #s are. There will be two numbers the 'short term' (many times condensed into STFT) and long term (LTFT). They may give us a clue as to whether the lean condition is intermittant or constant.

The freeze frame data is a helpful tool because it gives you many of the operating parameters necessary to recreate the fault in the shop.(engine temp,engine speed, fuel trims,vehicle speed,things like that)

From what I gather, your software doesn't provide this. Hopefully I'm wrong.

As for test driving, Yes drive it with the laptop connected. If it's available use the software's "snap shot" or "movie" or whatever data logging functions it has to look at the failure after the fact. Otherwise you'll need someone to drive while you watch. Good luck and keep us in the loop.

I agree, the STFT and LTFT will be the one to watch here.

I think the ECU/PCM records the freeze frame data when it generates a code, so it may not show up on his software until a new code is generated. But I would rather see live dynamic data at various operating conditions, than just freeze frame data. I find freeze fram data alone very limiting.

The easyobdII software is what we were, are using here IIRC. I will check later to be sure.

I am curious what E-header you installed? OEM or something special?

 

[XJSteven]

I finally have some live data for you to help me interpret:

At idle: STFT - 33.59% (lean), LTFT - 33.59% (lean), 02 sensor - 0.880V constant.

At 1500rpm: STFT - 33.59% (lean), LTFT - 33.59% (lean), 02 sensor - 0.840V constant.

At 2500rpm: STFT - 32.03% (rich), LTFT - 32.81% (rich), 02 sensor - fluctuates between 0.1 and 1.0V.

From what I gather, your software doesn't provide this. Hopefully I'm wrong.

Hi four_shot, yes I'm fairly certain my software does provide Freeze Frame but being an intermittent problem it hasn't set the code off yet so no Freeze Frame data to report - for the moment.

I am curious what E-header you installed? OEM or something special?

I replaced the OEM exhaust header with Wildcat headers shown here:
http://www.bestmufflers.com/bshop/popup_image.php?pID=526&osCsid=89a2d9b6d9c0b38c664a5d577925ac0e

 

[XJSteven]

air filter

:idea:

Thank you Alex from China. I have no doubt a Chinese air filter will fix all my problems.

 

[Ecomike]

I am wondering if a cracked header, or gasket leak can show up just at idle and low RPM. The only other thing that makes sense is a cold O2 sensor at idle.

Get it hot at say 3000 rpm for about 2 minutes, then drop it to idle, and see if the O2 sensor voltage climbs slowly, or right away. Slowly would be a cooling sensor (heater in the O2 not working), fast would manifold leak, cracked header IMHO.

 

[casm]

:idea:

Thank you Alex from China. I have no doubt a Chinese air filter will fix all my problems.

Might want to edit that to remove the link - no sense in giving the spammer any business. He's been reported, BTW.

 

[BrokenSockets]

I don't think Alex is a spammer - I just got a nice email from him with a great business oportunity involving a Nigerian prince and a large sum of money he needs to get out of the country. He wants to hide the cash in boxes of Chinese air filters.

 

[XJSteven]

I don't think Alex is a spammer - I just got a nice email from him with a great business oportunity involving a Nigerian prince and a large sum of money he needs to get out of the country. He wants to hide the cash in boxes of Chinese air filters.

Spammers..scammers...what's the difference hey? I normally wouldn't trust a Nigerian with a Chinese air filter but if he's a prince it's got to be the real deal. Awesome!

Get it hot at say 3000 rpm for about 2 minutes, then drop it to idle, and see if the O2 sensor voltage climbs slowly, or right away.

I'll give this a go and report back.

 

[XJSteven]

Get it hot at say 3000 rpm for about 2 minutes, then drop it to idle, and see if the O2 sensor voltage climbs slowly, or right away. Slowly would be a cooling sensor (heater in the O2 not working), fast would manifold leak, cracked header IMHO.

The 02 sensor voltage climbs instantly as soon as I drop the revs but I don't think I have a manifold leak or cracked header.

Here are some more results:

FUEL TRIM ON LPG

At Idle - STFT: -33.59% (lean) / LTFT: -33.59% (lean)
At 1500rpm - STFT: -33.59% (lean) / LTFT: -33.59% (lean)
At 2500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)

The above readings were taken in PARK but if I go for a DRIVE with the vehicle moving I get the following results (regardless of rpm):

At 1500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)
At 2500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)


FUEL TRIM ON GASOLINE/PETROL

At Idle - STFT: +/-5% / LTFT: +15% (rich)
At 1500rpm - STFT: +/-5% / LTFT: +15% (rich)
At 2500rpm - STFT: +/-5% / LTFT: +18% (rich)

The above readings were taken in in PARK but the Long Term Fuel Trim was slightly different each time I checked (I think it was slowly increasing). I didn't check the Fuel Trim while Driving but did notice something odd:

On gasoline/petrol, both stationary and driving, I was unable to keep the revs steady at 1300-1400rpm. Instead it fluctuated rapidly up and down between 1250rpm and 1500rpm once every second. This doesn't happen on LPG.

02 SENSOR ON LPG

At Idle - 0.880V constant
At 1500rpm - Normal swing (0.1-1.0V)
At 2500rpm - Normal swing (0.1-1.0V)

02 SENSOR ON GASOLINE/PETROL

At Idle - Normal swing (0.1-1.0V)
At 1500rpm - Normal swing (0.1-1.0V)
At 2500rpm - 0.7V to 0.8V range

Can any of you experts please tell me what this means? It's all very confusing to me.

 

[XJSteven]

On a similar topic, I understand the PCM uses 02 sensor data to control the air/fuel ratio, for example- by keeping the fuel injector open longer. I asked an LPG specialist today how it works on LPG (when the injectors aren't in use) and he said it uses the Converter at idle and then a Stepper Motor at higher revs using data from the TPS and 02 sensor. I don't know if this bit of info helps with my problem?

 

[Ecomike]

On a similar topic, I understand the PCM uses 02 sensor data to control the air/fuel ratio, for example- by keeping the fuel injector open longer. I asked an LPG specialist today how it works on LPG (when the injectors aren't in use) and he said it uses the Converter at idle and then a Stepper Motor at higher revs using data from the TPS and 02 sensor. I don't know if this bit of info helps with my problem?

Converter at idle? Please explain, what converter?

I see one interesting item, the LPG is running rich at idle and the gasoline is running rich at 2500 rpm, according to the O2 sensor, and then more normal otherwise. Complete opposites!

Is the 2 fuel system confusing the PCM stored operating memory as you switch fuels, or is smart enough to know you switch fuels and used separate memory?

Gasoline seems to be OK (less bad, or problematic) on both fuel trim and O2 readings, except O2 says it is running too rich at 2500 rpm, which is very puzzling???????? That throws out my cold O2 sensor at idle O2 sensor idea, if you ran the idle test at extended idle long enough for the O2 sensor to get cold, if the heater was not working properly at idle????????????


The LPG data on the fuel trim versus O2 data is totally out of wack by itself, no obvious correlation there at all unless you ignore the 1500 rpm data in park!!!!!!!!!!!!!!

LPG idle data says the O2 sensor is not working right, or the PCM is getting bad data meaning a wire problem, but most other data says PCM and O2 sensor and wires are OK.

This one may take a while to figure out!!!! I would rerun all the test, make sure the test data is all repeatable!!!!! Also, let idle for 5-10 minutes on each fuel before taking the idle data!!! Hopefully some of the data is bad, or not repeatable, to help us get a logical picture!!!

I do see one thing that makes some sense by itself, it looks like the PCM is seeing rich O2 data at idle in park for LPG, and has fuel trim set to try and lean the LPG out, but the O2 sensor is telling it is not having any success, which means it is getting too much LPG at idle in spite of PCM efforts to lean it at idle in park! Which sounds like a fuel feed control, on minimum control rate problem on LPG

 

[gradon]

Hmph, on the lm-2 obd2 scanner/wideband o2 meter/datalogger I have, - fuel trims mean the computer is pulling a % of fuel out from it being too rich and + fuel trims mean the computer is adding a % of fuel because it is running lean(so - means rich and + means lean). That +/-30ish% number is the max the computer can compensate, so things are bad, and it can be forced into Open loop. The fuel trims are only active in Closed loop, where the o2 sensor voltages should be oscillating around a certain voltage that is the stoich 14.7:1 ratio(.45V comes to mind for the ho+ years).

 

[Ecomike]

Hmph, on the lm-2 obd2 scanner/wideband o2 meter/datalogger I have, - fuel trims mean the computer is pulling a % of fuel out from it being too rich and + fuel trims mean the computer is adding a % of fuel because it is running lean(so - means rich and + means lean). That +/-30ish% number is the max the computer can compensate, so things are bad, and it can be forced into Open loop. The fuel trims are only active in Closed loop, where the o2 sensor voltages should be oscillating around a certain voltage that is the stoich 14.7:1 ratio(.45V comes to mind for the ho+ years).

Thanks, I probably was reading that fuel trim data backwards, so I need to start over on my thoughts regarding anything to do with the fuel trim data. I rarely ever had fuel trim data (but that is changing), have only seen it 2 or 3 times now, as most of my junk is Renix, just DMM and AMMs to read volts directly.

 

[XJSteven]

Converter at idle? Please explain, what converter?

Sorry I don't completely understand the LPG set-up myself but I know there is a Converter with an adjustable 'Idle Mixture' bolt on it.

Is the 2 fuel system confusing the PCM stored operating memory as you switch fuels, or is smart enough to know you switch fuels and used separate memory?

I wondered this myself. Both readings completely change when I switch the fuel over, including the LTFT, so it possibly is smart enough?

I would rerun all the test, make sure the test data is all repeatable!!!!! Also, let idle for 5-10 minutes on each fuel before taking the idle data!!! Hopefully some of the data is bad, or not repeatable, to help us get a logical picture!!!

I repeated this on more than one occasion before posting the results and did let it idle for at least 5mins on each fuel first. I will try again today to confirm data.

 

[XJSteven]

Data Confirmed - Updated

Same results today except for the changes in red:

FUEL TRIM ON LPG

At Idle - STFT: -33.59% (lean) / LTFT: -33.59% (lean)
At 1500rpm - STFT: -33.59% (lean) / LTFT: -33.59% (lean)
At 2500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)

The above readings were taken in PARK but if I go for a DRIVE with the vehicle moving I get the following results (regardless of rpm):

At 1500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)
At 2500rpm - STFT: +32.81% (rich) / LTFT: +32.03% (rich)


FUEL TRIM ON GASOLINE/PETROL

At Idle - STFT: +/-5% / LTFT: +15% (rich)
At 1500rpm - Unable to get an accurate reading due to rpm jumping up and down every second between 1500rpm-1600rpm, LTFT constantly changed from -33.59% (lean) to +26.56% (rich) - with a few figures in between.
At 2500rpm - STFT: +/-5% / LTFT: +18% (rich)

The above readings were taken in PARK but if I go for a DRIVE with the vehicle moving the LTFT constantly jumps around between +15% (rich) and +32.03% (rich).


02 SENSOR ON LPG

At Idle - 0.880V constant
At 1500rpm - 0.880V constant, and stayed constant up until 2000rpm
At 2500rpm - Normal swing (0.1-1.0V)

02 SENSOR ON GASOLINE/PETROL

At Idle - Normal swing (0.1-1.0V)
At 1500rpm - Normal swing (0.1-1.0V)
At 2500rpm - 0.7V to 0.8V range


Thoughts anyone?

 

[Ecomike]

One conclusion I have is that the computer is trying to control LPG fuel flow rate, but is not having any success! Control problem with converter?

Does the TPS have a bad spot, noisy spot at 1500 rpm (looking at the gasoline idle problem at 1500 rpm).

The switch from lean to rich from 1500 rpm to 2500 rpm for LPG shows the excess LPG at idle is caught up with control wise by 2500 rpm and the computer switches from trying to lean out the fuel feed (as fast as it can) to trying to increase it (as fast as it can) which may be due to over control on one side and then a hurry up and change the other way as it passes a point where the excess fuel and feed control system is no longer dumping too much fuel, but suddenly (as RPM climbs) trys to reverse as it finds that it overcompensated earlier. Once again indicating too much LPG at ilde, so much so that it is not under control, until 2500 rpm air flow gets back near stoich, then the PCM overshoots in the other direction.

What ever controls the LPG flow hardware wise (not the PCM), that is comparable to a fuel injector is what I would look at next.

The gasoline system is better off than the LPG from the O2 and FT data.

One other wild card is leaky valves or exhaust leak that changes patterns as you pass from 1500 to 2500 rpm.

I will try and really study the data harder this weekend when I have more time to focus, right now I am just throwing darts quickly.

 

[XJSteven]

I will try and really study the data harder this weekend when I have more time to focus, right now I am just throwing darts quickly.

Thanks Ecomike, I would appreciate that. In the mean time I will swap the TPS back again and see if I still have the same problem at 1500rpm.

 

[XJSteven]

Does the TPS have a bad spot, noisy spot at 1500 rpm (looking at the gasoline idle problem at 1500 rpm).

You were correct Ecomike. Replacing the TPS with my old one has fixed the fluctuating rpm problem. I'm not very impressed with the Standard Motor Products brand from Rockauto.com. This new TPS was a warranty replacement because the first one they sent me only lasted a few weeks before a fault caused my auto transmission to get stuck in 1st gear. I'm back to using the original TPS that was on the vehicle when I bought it 2 years ago (despite attempting to replace it twice now!). Will have to try a better brand next time.

Anyway, the most recent data I posted for STFT and LTFT remains correct because I have confirmed it again with the 'old' TPS re-installed.

 

[McQue]

Sorry I don't completely understand the LPG set-up myself but I know there is a Converter with an adjustable 'Idle Mixture' bolt on it.

Have you considered adjusting this "Idle Mixture" bolt, at idle speed ,while monitoring the 02 sensor live data to see if you can it's voltage output switching back and forth ? Seems like STFT at less than 10% or less would be a good goal...

If you keep track of the turns on the adjustment, you should be able to get it back to where you started if you need too...

 

[XJSteven]

Have you considered adjusting this "Idle Mixture" bolt, at idle speed ,while monitoring the 02 sensor live data to see if you can it's voltage output switching back and forth ? Seems like STFT at less than 10% or less would be a good goal...

If you keep track of the turns on the adjustment, you should be able to get it back to where you started if you need too...

Hi McQue,

I haven't played with this myself but the first thing I did 6 months ago when I noticed the poor fuel economy was take it back to the LPG installer for adjustment. He hooked some sort of 'gas analyser' to the exhaust pipe, took some readings and said the LPG mixture was fine and didn't need adjustment. When I questioned the poor fuel economy and P0171 error code he said he could hear an air leak at the exhaust manifold gasket and that the PCM was probably adding more fuel to compensate (I now know this made no difference because I've since fixed that problem).

I presumed this meant the LPG system was ok and started looking elsewhere for the problem. What also confirmed it for me was when I cleared the codes and drove around on gasoline/petrol the P0171 code reappeared.

I won't be very happy if it does turn out to be a fault with the LPG system because as I said that was the first thing I got checked out months ago! Are you saying it looks like it could be from the fuel trim data I've posted?

 

[gradon]

I agree with Mike's diagnosis in post 135. .8V is rich for gas(have never read up on lpg, so dunno what ratio is stoich or how the o2 reads on it). I'm not sure the o2 should be reading .8V at 2500 with gas under light throttle--my 96 doesn't go into OL until 3100rpm under light throttle(I wish it went in OL earlier to be richer than that 14.7) With LPG, should the o2 be oscillating around the stoich point in closed loop/low rpms?