Help with Silverado tune

[nfs13b]

Hello all,

I have a 2003 Silverado 2500hd with the 6.0L engine running on LPG... I am getting faults picked up with the EFI Live tool which are:

P0172 "Fuel Trim System Rich Bank 1" EPA ($10) "Powertrain Control Module (PCM)"
P0175 "Fuel Trim System Rich Bank 2" EPA ($10) "Powertrain Control Module (PCM)"
P0302 "Cylinder 2 Misfire Detected" EPA ($10) "Powertrain Control Module (PCM)"
P0300 "Engine Misfire Detected" H ($10) "Powertrain Control Module (PCM)"
C0201 "Antilock Brake System (ABS) Enable Relay Contact Circuit" H ($29) "Electronic Brake Control Module (EBCM)"

In regards to the misfire the car still runs fine and the only reason I know its there is because the programmer tells me it is....

Is anyone able to have a look at my logs and tell me if I have a problem with the tune...

12993
12994

This is just a normal driving log on my way home and also when I was towing my car trailer...

Now I dont know much at all about the tuning side of it so any help would be really appreciated..

I have recently changed the spark plugs too and put in new iridiums hoping it might fix up the miss but it didnt..

[schwoch1]

Are you using a multiport setup with your LPG fueling, or are you using a vaporizer before the throttle body and just shutting the petrol fuel injectors off?

Mike

[nfs13b]

Are you using a multiport setup with your LPG fueling, or are you using a vaporizer before the throttle body and just shutting the petrol fuel injectors off?

Mike

Hey Mike,

I am using a liquid LPG setup with 8 LPG injectors installed and the signal piggybacking off the petrol injectors signals

[schwoch1]

Hey Mike,

I am using a liquid LPG setup with 8 LPG injectors installed and the signal piggybacking off the petrol injectors signals
Ok, I was hoping you would say that!!! I have seen some horrible conversions where people just jam a vaporizer before the throttle body than wonder why their plastic intake blew off when it backfired!!
Is this conversion strictly LPG only, not bi fuel? If LPG only, it looks as if you are going to have to do some tuning on the VE tables to get the A/F ratio's back in line so that the engine is not running rich. I also think you are going to have to change the A/F ratio that the computer is commanding/expecting to get everything to work right. Do you have a wideband oxygen sensor? If you are doing a bi fuel that could get a little tricky then. Let me know!!!

Mike

[nfs13b]

Sadly it's dual fuel, it starts on petrol for 30 seconds and then switches to lpg. The only other time I drive it on petrol though is if I run out of lpg then it may go 10km on petrol till I can get some more lpg. If the economy will improve a lot though I am happy to make sure ii fill it up before it runs out.... Although someone once told me there should be a way to make it switch to think it is on liike ethanol and use the flex fuel maps... But didn't say how....

[schwoch1]

Sadly it's dual fuel, it starts on petrol for 30 seconds and then switches to lpg. The only other time I drive it on petrol though is if I run out of lpg then it may go 10km on petrol till I can get some more lpg. If the economy will improve a lot though I am happy to make sure ii fill it up before it runs out.... Although someone once told me there should be a way to make it switch to think it is on liike ethanol and use the flex fuel maps... But didn't say how....

Yah, that throws a little curve in things. Most of the dual fuel systems that I have seen use some sort of control box that is adjustable by the end user to adjust the A/F ration while on LPG. There was a system out there a while back that Moates marketed that had two flash chips in it, so in essence you could have two totally different calibrations inside of one PCM, but I don't belive it was designed for the 1meg PCM's that your truck uses. If you have to be able to have dual fuel options, than I would check with your LPG conversion company and see if somehow they can see what the heck is going on!!!
Mike

[nfs13b]

The conversion company said its all just controlled off the one tune, but they said they put the right size lpg injectors in it to balance against the petrol ones for the different tunes... How about if I drive it on petrol for a bit and log that? And I think I missed a bit earlier, no I don't have wideband o2 sensor in the Silverado :(

[BLK02WS6]

This is just a thought, but could you use a Flex Fuel OS and use a switch connected to the PCM pin for the flex fuel sensor - when you switch fuels, you could use the VE tables for the different fuels provided in the OS?

I don't know anything about LPG, but from the discussion, it seems that any way you could change the fueling would be helpful...

Another thougth was whether or not COS5 was available for your OS - use the nitrous enrichment for your fueling controlled off of a switch??

[schwoch1]

Got a copy of your cal? Maybe flex fuel is an option!!!! I have heard of injectors being 'sized' for the application, but I am not a believer at all in the one size will fit multiple vehicles theory!
Post a copy of your cal if you can!!!

Mike

[nfs13b]

Is the cal the tune? If so here it is attached.... or is it this calz file perhaps?

Or is it:


VIN 1GCHC23U13F222472
Engine 6.0L
Transmission 4L80E
PCM flash 1MB
Operating System 03150003
Calibration N/A
OS ID V3
Cal ID
BCC MOD3
OS Level
Cal Level
Cal file name 03150003.calz
Cal file version 7.94
Cal file date 07 April, 2009
Operating System 03150003 $4F66
Engine Calibration 12577909 $58D5
Engine Diagnostics 12580650 $F69F
Transmission Calibration 12572453 $2A0C
Transmission Diagnostics 12572475 $40FE
Fuel System 12578355 $B3C0
System 12572587 $1B2D
Speedometer 12572529 $CE74

[nfs13b]

was one of those files the right one to post up?

[joecar]

Yes, the .tun file.

[schwoch1]

Sorry for the long time before I replied. Life for some reason seems to keep getting in the way of things! I took a look at your cal, and it has the flex fuel option in it, although it is turned off and the tables are not set up. If you are looking for a quick and easy way of getting your truck to run better, enabling flex fuel is not going to be an option. If you are willing to do some experimentation, it looks to me that you can enable flex fuel mode by inputting a specific frequency to a pin on the PCM to emulate the flex fuel sensor and the PCM will use a different target A/F ratio table/ timing tables. This in theory would allow you to run a different timing/A-f ratio table when running on autogas/LPG/propane.
What does anyone else think???? If you are serious on trying this, I can provide you with the pin on the PCM and the target frequency that you need to enable flex fuel.
It would be quicker to take the truck back to the company who installed the autogas/LPG conversion kit and tell them it don't run right and have them repair their mess up. If they are unwilling/incapable, than I think this would be the next logical step!!!

Mike

[nfs13b]

Hey Mike,

The company that installed the LPG told me that to get the better economy/performance I would need to take the car elsewhere and have it tuned :( Im probably not smart enough to set up the flex fuel on my own either unfortunately

[BLK02WS6]

I would think the flex fuel route would work, but it may take a bit of experimentation... I'm sure we could help on here. Starting with the ethanol tables the same as the gas tables would put it right where it is now, then just make small changes to the ethanol tables to bring it around on LPG... When you start the truck on gas, is there a switch to change it over to LPG, or does it work off of a timer or something (automatic)?

Mike - do you know what the stoich is for LPG? I've worked with alcohol, but don't know anything about LPG... I think if he can hook up a way to force the flex fuel tables, we can help him on here with the tune...

Another thing to consider is that there is a flex fuel OS that works without an in tank sensor as well - it is somewhat complicated, but basically, it uses LTFTs and fuel level changes to tell the PCM to move to the ethanol tables. If you know the values for LPG, we can just substitute them for the ethanol values and work from there... I was looking into this because I have a non-flex fuel Tahoe and wanted to know what it would take to convert it over to flex fuel - I was very surprised to find out that some models don't use an in tank sensor... Here is my thread with some good info in it: http://forum.efilive.com/showthread.php?18862-Ethanol-questions-converting-non-flex-fuel-to-ethanol

I can't look at the tune here at work, but I will look at it tonight and also check to see what the equivalent tune for a flex fuel truck of the same year looks like... if it is like my Tahoe and uses no in tank sensor, maybe we can help him out with changes only to the tuning... but will need to know the value of stoich for LPG...


Edit: I did some searches on LPG conversions - it seems that there are many different ways that the system is integrated into the existing fuel system and PCM - it also appears that the idea I mentioned above for the OS without the in tank sensor won't work - it is dependent on the level in the gas tank changing and now I see that doesn't happen with LPG... so, we are back to using the pin on the PCM - but still workable...

nfs13b - do you have the install instructions, or know the brand of conversion? That would let us know how your particular setup is supposed to "play" with the existing system...

Bret

[nfs13b]

Hey Bret,

The car starts on petrol and then after 30 seconds it switches over to lpg.. I don't have any install instruction but there is no separate computer like their is for some setups. It is a JTG setup and the brand is icom. I think the stoich for lpg is around 15.1 but not 100% sure... If it worked, for the petrol side couldn't I use an existing tune file floating around...

[BLK02WS6]

What is JTG a brand? Sorry, not too familiar with these conversions. I'm trying to figure out how the system works to determine if you even need an alternate fuel map... from what I read about one system that is on a fuel injected car, they use an emulator to "fool" the computer while on the LPG to keep from getting the codes - but, the fuel is cut off while on the LPG... if it works like that, there is no need for an alternate fuel map because there is no gas being burned while on the LPG... seems there are several variations of LPG conversions out there and they are vastly different. I will search and try to learn about your particular one so that maybe we can figure out how to help you...

Bret

[schwoch1]

I looked up what this truck would of used if it were flex fuel and it would use a composition sensor that measures the alcohol content in real time. The sensor than puts out a specific output that changes frequency when more or less alcohol runs thruough it. Like has been said, I think it can be done, it just will take some experimentation!!!

Mike

[schwoch1]

Hey Bret,

The car starts on petrol and then after 30 seconds it switches over to lpg.. I don't have any install instruction but there is no separate computer like their is for some setups. It is a JTG setup and the brand is icom. I think the stoich for lpg is around 15.1 but not 100% sure... If it worked, for the petrol side couldn't I use an existing tune file floating around...
I just checked, the stoich. ratio for propane/LPG is 15.7:1. They routinely run it leaner though because LPG doesn't have the same lean problems that regular gas/petrol does. I have heard of ICOM, they are a fairly large company that does LPG conversions. Here in the states, LPG conversions are kinda unheard of, everyone is afraid that their car is going to go up in a ball of flames. No one has any problem with it to heat their homes or cook their food, but put it in a car and everyone gets really paranoid for some reason. LPG is actually safer than petrol/gasoline, but no one seens to belive you!!!
Mike

[BLK02WS6]

I looked at the icom website and they only list a JTG II and JTG HP... They use a stand-alone computer, so it sounds like you may have an earlier version or something... To be able to figure out how to make it work, we would have to know exactly how the kit interacts with the existing components - and it will take some experimentation as Mike said.

What I would do would be to find out the frequency that it would take to force the ethanol tables - hook up either a separate switch or somehow into the 30 second timer circuit to put that frequency on the correct pin on the PCM. At first, I would make the ethanol tables duplicate the gas tables (that effectively starts it off right where it is now) - then make small changes to the ethanol tables to dial it in on the LPG. Make sense?

[BLK02WS6]

nfs13b - may want to talk to these guys: http://apsperformancetuning.com.au/performance-tuning/performance-lpg/ They have done what you are trying to do and might be a whole lot more helpful than we can be on the internet... Not sure where they are located, but hopefully this is helpful to you...

One thing I'll say - this post has me thinking about alternative fuels and how I wish they were more available here in my area... I'm going to learn all I can now, so if they do, I'll be ahead of the curve!

[nfs13b]

Hey Guys,

Here is the link to the manufacturer/importer of the kit I have: http://www.alpgw.com.au/Products/Systems/JTG-Liquid-Injection.aspx

It uses no petrol while running on LPG

I will speak with APS, thanks for the link, the issue will be though that they are about 1000km away from me :(

[BLK02WS6]

I went to that site last night during my google searches - thing is, it doesn't go into much detail...

Maybe APS can give you details about how they tied into the PCM pin to switch to the ethanol tables - that is the hurdle - tuning the tables once you have activated them is easy...

[schwoch1]

I went to that site last night during my google searches - thing is, it doesn't go into much detail...

Maybe APS can give you details about how they tied into the PCM pin to switch to the ethanol tables - that is the hurdle - tuning the tables once you have activated them is easy...
It looks fairly easy if someone can make a signal generator to feed the PCM what it needs to switch over to E85, here is a description of how the fuel composition sensor works:

FUEL COMPOSITION SENSOR DESCRIPTION

The fuel composition sensor (FCS), or flex fuel sensor (service parts term), measures the ethanol-gasoline ratio of the fuel being used in a flexible fuel vehicle. Flexible fuel vehicles can be operated with a blend of ethanol and gasoline, up to 85 percent ethanol. In order to adjust the ignition timing and the fuel quantity to be injected, the engine management system requires information about the percentage of ethanol in the fuel.

The FCS uses quick-connect style fuel connections, an incoming fuel connection, and an outgoing fuel connection. The two connections have different diameters, to prevent incorrect attachment of the fuel lines. All fuel passes through the fuel composition sensor before continuing on to the fuel rail. The fuel composition sensor measures two different fuel related parameters, and sends an electrical signal to the powertrain control module (PCM) to indicate ethanol percentage, and fuel temperature.

The fuel composition sensor has a three-wire electrical harness connector. The three wires provide a ground circuit, a power source, and a signal output to the PCM. The power source is vehicle system voltage, +12 volts ), and the ground circuit connects to chassis ground. The signal circuit carries both the ethanol percentage and fuel temperature within the same signal, on the same wire.

The FCS uses a microprocessor inside the sensor to measure the ethanol percentage and fuel temp, and change an output signal accordingly. The electrical characteristic of the FCS signal is a square-wave digital signal. The signal is both variable frequency and variable pulse width. The frequency of the signal indicates the ethanol percentage, and the pulse width indicates the fuel temperature. The PCM provides an internal pull-up to five volts on the signal circuit, and the FCS pulls the 5 volts to ground in pulses. The output frequency is linear to the percentage of ethanol content in the fuel. The normal range of operating frequency is between 50 and 150 Hertz , with 50 Hertz representing 0 percent ethanol, and 150 Hertz representing 100 percent ethanol. The normal pulse width range of the digital pulses is between 1 and 5 milliseconds , with 1 millisecond representing -40°C (-40°F) , and 5 milliseconds representing 125°C (257°F) .

The microprocessor inside the sensor is capable of a certain amount of self-diagnosis. An output frequency of 170 Hertz indicates either that the fuel is contaminated or contains methanol (it should not), or that an internal sensor electrical fault has been detected. Certain substances dissolved in the fuel can cause the fuel to be contaminated, raising the output frequency to be higher than the actual ethanol percentage should indicate. Examples of these substances include water, sodium chloride (salt), and methanol.

It should be noted that it is likely that the FCS will indicate a slightly lower ethanol percentage than what is advertised at the fueling station. This is not a fault of the sensor. The reason has to do with government requirements for alcohol-based motor fuels. Government regulations require that alcohol intended for use as motor fuel be DENATURED. This means that 100 percent pure ethanol is first denatured with approximately 41/2 percent gasoline, before being mixed with anything else. When an ethanol gasoline mixture is advertised as E85, the 85 percent ethanol was denatured before being blended with gasoline, meaning an advertised E85 fuel contains only about 81 percent ethanol. The FCS measures the actual percentage of ethanol in the fuel.

Hope this spurs some idears!!!

Mike

[joecar]

...
The frequency of the signal indicates the ethanol percentage, and the pulse width indicates the fuel temperature.
...
Very clever... it allows you to use a DMM to display both (frequency and duty cycle).

[BLK02WS6]

Mike - is that description from a stock GM sensor? I'm wondering because I don't see where fuel temperature is used by the PCM...

As far as this post - I did some searches for signal generators and found plenty of them that could dial up between 50Hz and 150Hz, but not many that could vary the pulse width as well... some of them were kits, so I suppose some electronics guru could make one... that's not me though...

[nfs13b]

I have another question now.... With the current tune I have I get the misfire on cylinder 2 constantly and a few on misfires on other cylinders.... I have an old mafless tune that I have loaded in this morning to see if it changed the misfires and when I have the loger on without the ignistion on I am getting misfires (only with a value of 2) on a few cylinders (but not cylinder 2).... When the car is running there is no further misfires recorded (but the values stay the same)

Does this mean the misfire issue is tune relatred and not an engine issue?

Also if I was to make the AFR value 14.9 and not drive it on petrol (only start it on petrol and have it idle until it changes over) would that be ok and help at all with better economy on lpg?

[BLK02WS6]

The misfire may be tune related - it might be misfiring due to running too rich...

Yes, if you changed the stoich value to that of lpg, and did not drive it on petrol, you could make that work - You could work with commanded fuel when in open loop table - make the colder ECT areas richer for starting on the gas... that would help your economy a good bit!

Edit - I just downloaded your tune and saw you have converted it to COS3. So, you don't have the same commanded fuel when in open loop table. However, table A0008 will work - you may want to bring the 1.10 down a little more though - that will start it on the gas commanding 14.2. It will not warm up enough in 30 seconds to lean it out much. So, the only other thing you need to do is change B3601 to 15.7 for the lpg... make sense?

If changing to the above doesn't correct your misfires, you may need to look at the plugs and make sure they are fouled from running rich all this time...

Also, if you want to use a wideband and totally tune it, don't forget to work in lambda and EQR - not AFR!

[BLK02WS6]

Here is what I'm talking about...

[nfs13b]

Here is what I'm talking about...

Thanks for that.... When I put it in and drive around and log it, what values in the log am I looking for to see if there are any issues, is it just the misfires and Knock retard I need to be checking?

Also this is the mafless tune that I put back in the car yesterday, its the one that has less misfire's but when I had it in the car I was only getting 280km from a tank compared with 310km with the first tune...

[BLK02WS6]

When you put it in and log it, log LTFTs, knock retard, timing, ECT and misfires... that should be pretty good. Post up the logs so I can check them out...

One other thing - if it has trouble on hot restarts on the gas (because ECT is in the range when it is lean), we can use after start enrichment to richen up those 30 seconds on the gas. Let me know how it does with cold starts, hot restarts, and normal driving...

I wouldn't worry about the misfires this minute - let's see how it is when you get it leaned out where it should be on the lpg... you may need to put a set of plugs in it if it doesn't clear up - or at least pull one and look at it....

[nfs13b]

OK stupid question now...

what PID is the timing...

[BLK02WS6]

"ignition timing advance for #1 cylinder"

[nfs13b]

Here is a log file of my drive to work with the new tune.... Just before I installed the new tune I was running on petrol on that mafless tune and there were no misfires recorded so I am assuming that the misfire issue is purely in the tune and not an engine problem.....

How does it look?

[BLK02WS6]

I should have told you to clear out the LTFTs with the bidirectional controls - sorry, I forgot to tell you that... the LTFTs came down in areas as you drove, but it would be best to clear them out and log again.

It seems that most of the misfires come when it is rich and pulling fuel. However, I think there is something up with #2 cylinder - just becuase the other tune doesn't show the misfires on that cylinder, doesn't mean it isn't engine related - I would need to see a log from that tune to see what is up. Sometimes, conditions created by the tune will make an engine issue more visible... make sense? In other words, a rich tune may bring out a slight misfire that might not be seen with a leaner tune...

An easy thing to do is pull #2 plug and have a look at it. Then swap #2 plug wire with one on the other side of the engine and log again - see if the misfire follows the wire. You can also do the same thing with a coil or injector if you are trying to track down a misfire. The misfire has to be figured out first...

[nfs13b]

I have changed all the wires and plugs on the engine but the misfire(on cylinder 2 and random other ones) was always there (on the old tune)... The only other thing I have left is changing the coil pack around... But what I dont really know is which side of the car is cylinder 2 on... the left or right?

[joecar]

odd numbers 1357 on left (US driver side), even numbers 2468 on right (US passenger side).

[joecar]

There are other reasons for misfire (that you have to also "eliminate")... e.g. injector problem, compression/coolant leak, broken valve spring, etc.

[nfs13b]

Hey Guys,

I had to go out at lunch so I changed the tune to the mafless tune I posted earlier and I reset the fuel learn trim...

There were no misfires and no LTFT info... Is this right or are these reading somehow disabled by this tune perhaps?

[joecar]

Misfire detection is disabled when MAF DTC's are present (which are present when MAF-less).

[nfs13b]

well that sucks... is there a way around that? The maf is still there and connected..... I dont really want to go and get stung by taking it to a mechanic to check if its a serious problem until I know its not the tune (or coil pack)... Or does anyone have a stock 2003 silverado tune that I can put in and see if I get the misfire on petrol perhaps?

Or should I not even be worrying about the misfire?

[joecar]

Post a logs files showing the misfire counters.

Are you getting any misfire DTC's...? Does the MIL flash while you're driving/idling...?

[BLK02WS6]

Hey Joe - There is a log in post 34 that shows the misfires. And in the first post he says he is getting #2 code and 0300.

Here is a stock 2003 tune to try on the gas...

[nfs13b]

Thanks for that..

I am not getting the engine warning light flash up though, only the codes when checked in efilive... If I check the error codes with the mafless tune would I still get the misfire coded come up if there was a problem?

[nfs13b]

to put that tune in am I going to have to load that OS as well or just the calibration?

[BLK02WS6]

You will have to do a full reflash of that tune (operating system and calibration - double red arrow icon)...

[BLK02WS6]

No, you won't get any misfire codes mafless...

[nfs13b]

ok so I put my Maf'd lpg tune back in the car and it was giving me the misfire data in the logs, I then put the stock silverado tune in and started it up on petrol and logged it around the block... attached is the log file..there were basically no misfires recorded...

I then put it on lpg with the same tune in it and logged it just in the driveway not moving and there were quite a few misfires on cylinder 2 (log also attached) so im thinking its either the coil (which will be a weekend job to swap around) or perhaps the lpg injector, which I will see if I can swap around as well...

Does my thinking sound reasonable?

[BLK02WS6]

My money is on the lpg injector in #2 cylinder... yes, your thinking is reasonable!