Roe SC and erratic A/F

[KenH]

I have been having a problem getting a consistent A/F on my setup. I can tune it in on a dyno or with my WBO in the car and it seems to run OK, then the next time I punch it, it will run pig rich, burying the WBO gauge and barely run, then it is fine again. Car also tends to vary at idle as well. Usually it will idle around 14.0 - 14.7, then it will drop down to 13.0 or so for no apparent reason and you can smell the gas.

I have found a number of issues that could cause this and fixed them, but I still am having the problem. Here's what I have done so far:

- Fixed a mismatch between the low impedance injectors and VEC2 with a load box from Sean.

- Car was not going into open loop mode at WOT. I fixed this by redrilling the throttle mounting bracket and moving it back to get full TPS range.

- Took off the fuel rails and replaced the fuel injector connectors with MSD type.

- I have checked fuel pressure on the dyno and it reads a consistent 55psi, so I don't think it is uneven fuel delivery.

- I replaced the PCM with a '96 one.

- I added supplemental grounds at the battery temp sensor, TPS sensor, MAP sensor and Air temp sensor.

Any other ideals to try? Mods are as per my signature.

 

[Streak]

try turning off the water/**** kit and installing the old card for the vec2. It could be a programming problem but you need to find out if the programming is doing something different when the water/**** is activated....

 

[Schulmann]

I had some similar issues but in different contexts.
On the driver side cylinders turn very rich sometimes.
My AFR trouble can happen in two situations:

- I start the car when cold. Everything is perfect. Drive 200mi. Then stop for a break. I start the car and driver side cylinders turn into very rich. Between 10-13 AFR and stay rich. If I stop and restart the engin 3-4 times the problem goes away.

I think this problem is related to Wideband oxygen sensor connection, the way how I linked up the 5v output from the oxygen sensor to the VEC2. I have 3 inputs from other devices connected to the VEC2. I was forced to use Master Analog input 2 on Vec2 to connect my LM-1. Actually I have just removed this connexion and reconnected the LM-1 to Slave Analog In 1 (I temporarely removed the other input from this place.)


The second situation is the following:
- I start the car hot. Engin runs perfectly. Withing 5 seconds I do a hard acceleration in 1st gear. Again cylinders on driver side turn rich 10-13AFR and stay rich. This is an older issue since awhile.

I presume that maybe I have an electrical issue. I think, I have connected too many electrical devices onto the same fuse. My next mod will be the rebuilding of the electrical connections. Maybe this week-end ...

I had another friend who had a problem to connect input to the VEC2. Actually he had an issue with Slave input 2. For some reason Slave input 2 was producing 0-5v output ... ?! We wanted to connect the stock oxygen sensor to this place and once the connection was made we got a check engin light indicating a rich condition. We spent 2 hours to locate this trouble.

 

[Sean Roe]

Hi Ken,
You need to log and watch the what the injectors are doing. Specifically, start by looking at the "add fuel" value when it goes richer.
The program will do what you tell it to do for fuel enrichment based on RPM and manifold pressure. Maybe it's getting into a vacuum area where there's some fuel enrichment?
You can decrease the value in vacuum, between 12" vacuum and 0 psi, then test the results. I know I've had to reduce these values on some cars. Some have lower vacuum at idle (closer to about 8"), pushing the fuel load table into an area with a larger value. Some need it adjusted in order to get better tip in throttle response (some cars need more enrichment, some need less). You can also change the fuel RPM values to 0 at 750-1,000 RPM.

Andras,
It sounds like your friend plugged into analog 2 output, not input. Analog 2 output is 5v.

Regards,
Sean

 

[Mr Hemi Head]

try turning off the water/**** kit and installing the old card for the vec2. It could be a programming problem but you need to find out if the programming is doing something different when the water/**** is activated....

this would work with a 5lb pulley, but not with the 6.5lb pulley that Ken is running.

 

[KenH]

Hi Sean,
I looked back at some logs I had taken previously and here is one that I think may have caught a hint at the issue. It looks to me like the stock injector signal coming into the VEC2 is suddenly max'ing out for some reason causing the VEC2 to similarly max out the injectors. Any ideas on what could cause this?

In your previous post you mentioned "between 12" vacuum and 0 psi". What is the conversion factor between 12" vacuum and psi? Is it basically 2:1 i.e. -24psi?

--- Ken

 

[Schulmann]

Kenh,

InFuel DC is your "Fuel Duty Cycle" in %.
I have also 100% duty cycle all over in 4th gear.

Show us your graph with:
- PSI
- RPM
- AFR
- In Fuel [ms]
- Tot Fuel [ms]
- Add Fuel [ms]

 

[KenH]

Schulmann, yes it is in %. It is the IN Fuel DC % line. In this graph it went from 70% to 100% I don't have the file here at work, but the Add Fuel and Total Fuel also spikes with the Total Fuel going to 100% as well.

Here is a graph I found at work. If I remember right, the car was running pretty crappy this whole run and the RPM line seems to reflect that. On this one the %DC is 100% throughout the run. Here is it with the info you requested.

--- Ken

 

[Schulmann]

Kenh, this chart looks "normal".
Try to post the previous one with the same information.

 

[Mr Hemi Head]

Ken here is my latest run with a 5lb pulley. The car runs great. Why the difference in the fuel values? There is a fuel spike at the beginning of my chart but the car does not bog.UBBCode:

 

[KenH]

Schulmann, I don't agree that it looks normal. It doesn't make sense that the injector pulse-width would be an inverse slope of the RPMs.

Ken here is my latest run with a 5lb pulley. The car runs great. Why the difference in the fuel values? There is a fuel spike at the beginning of my chart but the car does not bog.

I agree, your fuel curve is what I would think looks normal. On runs that seem to be OK, my log looks much like yours. In the case above, as far as I can tell, the injectors have gone static (running at 100%). If I add the TOT Fuel DC % line, it shows 100% for the entire WOT run. What doesn't quite make sense is that my WBO only shows about a 11.1:1 A/F in the log. Sometimes when this happens, I have noticed the gauge ******* on the rich side (<10), so perhaps the logging output is offset for some reason from the WBO (its a Wideband Commander with quite a few miles on it, so it may be due for a new sensor).

 

[Mr Hemi Head]

Schulmann, I don't agree that it looks normal. It doesn't make sense that the injector pulse-width would be an inverse slope of the RPMs.

Ken here is my latest run with a 5lb pulley. The car runs great. Why the difference in the fuel values? There is a fuel spike at the beginning of my chart but the car does not bog.

I agree, your fuel curve is what I would think looks normal. On runs that seem to be OK, my log looks much like yours. In the case above, as far as I can tell, the injectors have gone static (running at 100%). If I add the TOT Fuel DC % line, it shows 100% for the entire WOT run. What doesn't quite make sense is that my WBO only shows about a 11.1:1 A/F in the log. Sometimes when this happens, I have noticed the gauge ******* on the rich side (<10), so perhaps the logging output is offset for some reason from the WBO (its a Wideband Commander with quite a few miles on it, so it may be due for a new sensor).

Here is the same run with TOT Fuel % displayed. There is a spike in the uper RPM range but it does not effect driveability.UBBCode:

BTW I'm using the WB Commander, I question the response time. Sometimes the W/M light is on, under part throttle, but there is no change in A/F. Rick

 

[Schulmann]

The WBC is a little toy. Not very accurate under 12 AFR. I have even remplaced the oxygen sensor and the AFR is still the same kind. Maybe there is a bug in their unit or the software. I have just installed an LM-1 unit and that one runs much better.


Kenh, you are right your chart doesn't look normal.
Sean will give you a better answer than me.
I am very interested in the solution.

Where did you connect your WBC ?
Is there any other input into vec2 ? If yes, where did you connect them ?

 

[Schulmann]

On driver side I have a WideBandCommander.
On passanger side I have an XD-1 unit (same as LM-1).

I presume that LM-1 gives the right AFR. I even did a switch and LM-1 always gives a lower AFR espetially under boost and under 12AFR.

I think Kenh that you chart is not so bad. But I agree that the spike is not nice.

 

[Joseph Dell]

I really need to learn the datalogging stuff on these things. Is there an easy way to set it up?

JD

 

[Mr Hemi Head]

Schulmann the graph with the both A/F monitors is giving me a headache. Could you just show the A/F values with RPMs?

 

[Mr Hemi Head]

I really need to learn the datalogging stuff on these things. Is there an easy way to set it up?

JD

Joe logging requires a cable from the VEC2 to a computer running the VEC2 program. I use my Sony VAIO on the passenger seat and an external cable routed over the steering column out the window to the engine compartment.
That way if I get pulled over during logging it looks legit.LOL

There is a wireless RS232 system available but it's hard to justify the expense for limited use.

 

[Schulmann]

This is my acceleration in 4th gear on highway.
Temp = 45F
Fuel = 94 oct
Puley = 6.5
Injection = 10% methanol, 90% water

 

[Supercharged Vipermad]

I really need to learn the datalogging stuff on these things. Is there an easy way to set it up?

JD

Hey, now I think I can help you!! Look at my POSER thread and you'll see I am READY for all this!! Little taste:

 

[Mr Hemi Head]

Tony all you need is a revolving lite and 2 way radio so you can pass for an undercover agent.

 

[Supercharged Vipermad]

I'll take a picture of the mounting ball....once I take the laptop tray out you can hardly see the rest of the setup! Sorry to hijack...I'll step out now.

 

[Mr Hemi Head]

The WBC is a little toy. Not very accurate under 12 AFR. I have even remplaced the oxygen sensor and the AFR is still the same kind. Maybe there is a bug in their unit or the software. I have just installed an LM-1 unit and that one runs much better.

Schulmann I recently did some pulls on a Dynojet dyno and the A/F values were the same as the Commander. The readings were delayed slightly since a tailpipe sensor was used.Can you do the same to determine which is correct?

 

[KenH]

OK, now that Tony has started his own poser thread, we can get back on track Actually that is a pretty good idea, I have a heck of a time trying to manage the laptop when I am by myself.

The fact that the injectors go to 100% have to be a smoking gun of some kind as they should never be maxing out. In some of the logs there are momentary spikes and in other cases it lasts for the full pull in a gear. When stock, the car didn't have this issue, yet it appears that the VEC2 is just responding to the stock PCM signals to max out the injectors or am I just reading the data wrong? Anyone have any theories on what would cause the stock PCM to intermittently max the injectors. The VEC2 could also just be misreading the stock injector signals which are in fact working properly. Could there still be a grounding issue between the stock PCM and the VEC2 or something?

Schulmann, when you had your viper exhaust catch on fire on the dyno, it sure sounds like it could have been a similar situation as I am seeing here. Injectors all go static and dump max fuel.

 

[KenH]

According the the service manual, the PCM uses RPM and MAP inputs to adjust fuel injector pulsewidth at WOT. The stock MAP sensor as logged by the VEC2 doesn't show any irregularities (runs smooth at around -1 to -1.8psi at WOT when this occurs, so I don't think it is a matter of one of the input signals to the stock PCM being out of whack.

 

[Viper Specialty]

KENH- The OEM PCM works backwards of how it should- it is a warranty writers dream. Instead of pulling some fuel as temp increases to keep the A/F correct, it adds it instead. When the Intake Air Temp sensor hits a certain temp, it will bump the duty cycle to 100%. Now, with stock injectors and a stock engine, this is acceptable. However, on a supercharged car, you are dealing with two issues; first, the injectors flow a LOT MORE fuel when full on, and second, the VEC-2 requires the individual pulses to base its corrected injector pulswidth on. For lack of a better term, it "confuses" the VEC-2. The simple way to fix this is to add a 4.7K ohm resistor in line with the IAT, though I have recently learned of a better way to handle the issue through another tuner on the board. I am not sure if he wants it openly posted, but if you PM me, I can set you in the right direction.

Through my own testing, this issue appears to be mainly limited to Gen-2 cars, I have experienced it once on my SRT. However, it has never occured at a time when it could cause problems. As the car is moving down the road when you nail WOT, the intae is sucking in a rather large amount of air. In the second you let off the gas to shift into the next gear, the intake air can "back up" with air from the engine compartment, more than hot enough to cause a 100% DTC. However, the only times I could duplicate it was when it was 85+ outside, and the car was slammed through 1st or 2nd gear, and then "cruised" into the next gear with minimal throttle. Due to a part throttle situation, the VEC-2 had little work to do, and it was not even noticeable, only showed in logging.

I recomend hooking a Scan Tool up to the car and reading what the PCM is seeing for IAT. See if you can isolate a temp range that causes your 100% DTC condition, as well as the issue itself.

 

[SNKBYT]

Hi Dan,
I am also experiencing A/F problems. At first I though it was bad pluggs. I installed your new champions yesterday but they didn't help. I went to F-1 Dyno last night. The car was running pig rich 9.3-10. We adjusted the A/f to around 12.o. Car ran great on the dyno, then did some street driving. Within 10 minuets the car began running rich again. We did not have time to put it back on the dyno to get the A/F readings. Do you have any idea what is happening? Could my O-2 sensonrs be bad,or is it something else. Thanks Tim
P.S. Thanks for the plugs, I can't wait to get this straighten out to see 705rwhp again.

 

[Schulmann]

Kenh, my dyno fire was purely related to a faulty spark plug. No doubte about that. Since this incident, I switched to NGK spark plugs.


Daniel, could you develop more this issue related to the IAT ? It sounds like no sense but might be true.

 

[KenH]

Daniel,
From the manual, it didn't look like the IAT was used for control at WOT, but I could be wrong. I have noticed that the issue appears to be much more apparent when the air temperature is relatively warm. I have also noticed that on some runs, the first run-up looks OK, then I shift and the injectors max out right after the shift and stay maxed out for the whole second run up, so what you are saying seems to be ringing very true.

Any recommendations on a good scan tool? Does the ScanEase that Roe sells capture this kind of info?

Thanks for the help and PM sent,
--- Ken

 

[KenH]

I figured I should update this post with my final results, as it might help someone else out in the future.

First off, I think there is something about some, but not necessarily all '01-'02 cars that just make them stubborn when you slap a SC on them. I am not convinced that any of my problems were actually related to the basic Roe SC setup and Sean has been great in helping me work through the problems.

After the above stuff went on I tried modifying the AIT with an inline resistor, but that did not help and in fact seemed to make it run worse. It was subsequently pointed out to me that since I am now running a '96 PCM, that might be the case.

Then Sean sent me some beta firmware for the VEC2 that a few people are running now. On my car, for some reason it made it run extremely different, which I don't think was the case with the other cars. The car went very lean and I threw lean bank 1 and bank 2 codes within 20 miles after updating the firmware. I then bought the EASE software so that I could monitor the fuel adaptives and they were max'd out as expected based on the codes I had gotten.

At this point, Sean decided that we should just start from scratch. We pulled all VEC2 programming out of my car and I started from scratch by setting the fuel injector offsets, balancing the left and right cylinder banks and leaning out the rear cylinders based on plug readings and dialing it in until I had the long-term fuel trims near zero at idle. The bizarre part is that the injector % had to be set to 74% rather than the nominal 59% that the 53lb injectors are normally set for. Also, I had to run the right bank richer than the left which is the opposite of normal. This should have made my car run way too rich, but it seems to be what it needed. Idle actually went from 13.5-14.5 to 17.0 -18.0 A/F which is way leaner than normal, but not necessarily bad from what I have read. Logically, I just can't figure out how the fuel trims can be zero without the idle being right around 14.7. If anyone has thoughts on the matter I'd like to hear it.

Sean then built a base program based on my new injector settings. The A/F held relatively steady and changes had the expected affect finally. Drivabilty around town was still sort of poor, so I tried an aggressive program that Sean had sent over that dump lots of fuel in at the lower RPMs and the car ran much better around town, but was way too rich when I got into the throttle. I kept the basic curve and just backed the values down and the car now runs great around town.

I have dialed the car in at WOT using the WBO I have installed in the car and it is running strong. I did have a lingering problem where if I went to WOT at a lower RPM, it would show pig rich and run like crap for the whole run, but if I started at a higher RPM, it was fine. I scratched my head on this for a while thinking it was a fuel delivery issue, but I ended up pulling about 1/2 the water injection out in the 1500-2000 range and that solved the problem. Perhaps excess water was blowing the spark out and causing a rich condition and for some reason it couldn't clear itself as the RPMs increased. In any case it worked, so I'm not complaining!

The car is now running great and has been doing so for a couple of weeks now. I also wanted to thank people like Schulmann, Mr Hemihead and Final GTS for helping to brainstorm on some of the problems, not only here but in *** and email as well. If you don't have a tuner nearby to turn to, it is nice to talk to people and bounce ideas off of them.

BTW, I also tried Sean's new HP/TQ logging feature on the beta software and got numbers of 577RWHP/625RWTQ which is right about what I was expecting with my setup, though I haven't had a chance to put it back on a dyno yet to check those numbers for accuracy. Considering that I was handed a dyno showing 480HP when I first had the SC installed, I'm feeling pretty good about the current state of affairs.

 

[Mr Hemi Head]

Ken, We knew you would get it right. Is HP & Torque graphed during logging? Rick