FAST problem: right side lean, driver's side rich?

[Spud]

I have a LM-1 O2 sensor in the driver's side exhaust and the FAST WB in the passenger side O2 sensor and they read differently.

I've got a 97 Cobra DOHC 4.6 with an HP twin stage 1 turbo kit, and a partridge in a pear tree... Running a FAST B2B with a WB O2 with FORD EDIS (not FAST EDIS) and I also have another WB sensor set up for the opposite side of the engine. The driver's side will show 11.3 AFR while the passenger side shows 13.5 AFR. I swapped the sides that the sensors for the respective guages were on to see if the problem followed the sensor swap, and the same thing occurred: passenger side still read about 2 points leaner than the driver's side.

*I've Done the following thus far (many when I had an ignition gremlin causing rich conditions in #1 and #2 cylinders, which I've solved);
1. Changed all the plugs and checked to see if any one was burning leaner than the rest, and the were all pretty close, except the driver's side had a couple more that were a bit darker, but nothing heavy.
2. Did a leak down test and it checked out really good.
3. Did a compression test also and it checked out real good too.
4. Ran the stock intake, and also tried a short runner and now am running a custom sheet metal intake, and the problem followed all of them.
5. Changed all the lead wires to new FMS 9mm and coil packs also, and dilectric grease was used properly.
6. Installed brand new MSD DIS-4 (without tach adapters!)
7. Installed brand new GM 3bar MAP sensor.
8. Had all injectors out (#60 Siemens) to be inspected and declared perfect! Again when I was determining why #1 and #2 were running rich?
9. Checked for vacuum leaks around the intake by spraying propane around the intake (I was told idle would change if it sucked in propane through a leak..?)
10. Inspected and cleaned all wiring to injectors and their connections.
11. Fuel pressure is at 45psi at idle.
12. Installed a new crank sensor and re-inspected all FAST wiring.
13. Installed a brand new EDIS-8 module

Important details: Engine is fed by an A-1000 with 10an lines up to the T, then 8an to the Steeda fuel rails, and then an Aeromotive boost sensitive FPR return system... The FAST is a B2B (not sequential with individual cylinder control) but I don't want to mask a problem by changing it to sequential unless that's the only solution and all mechanical anomalies are otherwise ruled out..?

The car seems to run pretty good actually, considering one side is way richer than the other. With this as is it won't nearly make the power it's capable of, but seems to run pretty strong in it's current state anyhow. Regardless, I can't and won't run it hard with the lean situation on the passenger side. That would spell disaster, and it seems I've dodged a bullet thus far, but again haven't really beat on it too hard yet. If anyone has any ideas I'd really appreciate them. I am contemplating the following otherwise;

1. Re and re the header and connections to ensure no exhaust leaks.
2. Take the front cover off and degree the cams again (very tricky and very crucial with these DOHC engines). I did have it professionally degreed when engine was out last time.
3. Send the FAST back to have it changed to sequential so I can adjust the right side from the left (wouldn't that be masking an underlying problem though?).

Thanks for any help..

Spud

 

[plumber]

I would take a good close look at the exhaust for a leak or hole in that header or collector. If the car is loud its hard to tell if theres a hole in a collector or leak that will suck air and fool your sensor. if your plugs look the same on both sides Id bet your mixture (from bank to bank) is ok.

earlier this year I was convinced my sensor went south because it would read dead lean and it turned out to be a hole in the colector where a pan evac nipple was welded in at one time.

Doug

 

[Spud]

I would take a good close look at the exhaust for a leak or hole in that header or collector. If the car is loud its hard to tell if theres a hole in a collector or leak that will suck air and fool your sensor. if your plugs look the same on both sides Id bet your mixture (from bank to bank) is ok.

earlier this year I was convinced my sensor went south because it would read dead lean and it turned out to be a hole in the colector where a pan evac nipple was welded in at one time.

Doug

Thanks for that. You've validated the first place I plan on looking which is to redo the header gasket and exhaust connections to the turbo and back on the passenger side to see if an exhaust compromise is the culprit... I'll give that a shot anyhow, and see what it gets me. Hopefully it's the exhaust because that's an easy fix. A bit of a pain in the ass to re and re the header, but not too bad with it having a tubular K-member.., just have to take down the starter. A/C pump is already deleted...

 

[Craig Smith]

Oh man, I wish I would have heard of someone having this problem long before you lost many hours of your life and dollars out of your bank account.

You can pretty much count on the LM-1 reading leaner than the FAST. I had done plenty of experimenting with the LM-1 vs. the FAST as well as LM-1 vs. DFI Gen 7, and the LM-1 read SUBSTANTIALLY richer than both. The difference between the LM-1 and either unit was about the same. Having also conducted tests of FAST and DFI systems against specialized test gas samples, other lambda meters, and even very high-end lab-grade equipment, I feel rather confident in saying that they are both quite accurate, and they track each other rather closely.

I hate to trash products on the internet, but I don't put any faith in the accuracy of the LM-1 system. Sorry to hear of your troubles, but I think you need to trust the system running your car and don't be concerned with any other data that clouds the issue. If you stick your FAST sensor in the other pipe, I would tend to believe that it would read very close to the one it's in now.

 

[Otto J]

I agree with Craig ,But also dont completley rely on the 02,When you make a pass pull a few plugs out of both sides and look at them. Make sure they look consistent on both sides of the motor

 

[plumber]

Craig, long time no talk to.

spud said in his first post that he did reverse the sensors and the same side still read lean.

doug

 

[Buzzard]

Lean condition

I would also check the exhaust system for leaks, and the injector harness to be sure that it is wired correctly for the firing order of the engine. I have seen many people just throw the harness on the engine without checking to see if it is wired correctly. Yes, FAST does make mistakes at times, I know it is hard to believe.

Let us know what you find out.

 

[Spud]

Craig, long time no talk to.

spud said in his first post that he did reverse the sensors and the same side still read lean.

doug

I just finished late last night re and re-ing the header and piping to the turbo and back to the O2 sensor I have downstream from the turbo. I didn't detect any signs at all of carbon tracking or leaks otherwise, but I re-sealed it all with a new header gasket. I fired it up this morning and let it run for a bit and it still showed leaner on the passenger side by about a full point to a point and 1/2 (11.4 driver's side & 12.8-13 passenger side).

Again, in switching the LC-1 sensor to the driver's side, the FAST wb is on the passenger side, and it's indicating still the same leaner reading that the LC-1 Bosch LSU wb had indicated. Good news is the LC-1 & XD1 guage on the dash seem to be working right because the switcheroo produced the same readings on the different sides of the engine.

Since I've been through just about every other "chase and replace" scenario from the obscure to the obvious (as aforementioned), I am thinking it has to be the FAST computer itself. In speaking to a few knowlegable modular personalities down south, they suggested I speak to FAST's R&D. What sounds like a viabel option is to switch it to sequential and while I'm at it have indiv. cylinder control added. They mentioned they've never heard of my anomaly before, but stated that most people don't run two wide bands to know any different... After all I did purchase the FAST unit used and so it's possible the injector drivers are wonky..?

I'll keep yawl posted when I get it back. Meanwhile, any other suggestions?

Thanks,

Spud

 

[Otto J]

may not be a bad idea to have the injectors tested

 

[Craig Smith]

Are the differences you are describing present at idle, only under load, or all the time?

Do the spark plugs seem to tell the same story as the wideband sensors?

While I will stand behind what I said, my first post on this regarding the LM-1 was probably not the cause in this case.

If you have access to an oscilloscope and know how to use one, you could check injector pulsewidth values yourself. I've not personally seen a situation where one bank was being commanded to a different pulsewidth than the other, but I suppose anything is possible.

I can probably help out with a SEFI and ICC upgrade if you intend to go that route - PM me on that if you'd like.

 

[Spud]

Thanks for the replies. I've already sent the unit back to FAST for upgrading (SEFI & ICC), but once I get it back I could sure use some help manipulating it.

To answer a couple questions or suggestions: I've had all the injectors tested and flowed and they all came up perfect (cost me almost $250!). The difference in AFR readings was sort of consistent with the plugs, but not really. There weren't any real obvious differences, but they were slightly more tanned on the driver's side than the passenger and so I guess you could say the plug readings were somewhat consistent. I would have thought the driver's side would have been almost black and the passenger stark white by the readings though? I don't have access to a scope to check the injectors, but I have a mechanic/friend that probably has one he'd be willing to lend me. Also, the different AFR readings are definately more pronounced at idle, as the gap narrows under load, but only narrows to within 1 point AFR.

While I'm waiting for the return of the puter, I'll re&re the driver's side exhaust to that turbo, and then when I get the computer back with the SEFI and ICC in she goes to see what's what. If the anomaly still exists I won't simply adjust the ICC out of the box as I don't want to overlook or mask a mechanical issue if one is hidden? Regarding mechanical incidence as a possible culprit, I've had the cams degreed by a local and trusted speed shop I know knows what they're doing, and so I'm pretty confident that's kosher, however to err is human... I am running Crower Stage 2 blower cams (Billet not regrinds) that I suppose might possibly be a culprit if there was something awry with them out of the box, but I can't fathom that, because that should have shown up in the degreeing process besides the fact they're very reputable? I ran a compression test and all checked out good, and also did a leak down test to be double sure, and that checked out as well. I replaced all of the ignition components I'm running with the 4.6 DOHC, including new coil packs, new lead wires, new MSD DIS-4 and a new FORD EDIS-8 module, new GM MAP sensor, and crank sensor etc... I've recently put a sheet metal intake in her, but this same lean on the passenger side anomaly existed similarly with the stock and other short runner intake I ran prior to the sheet metal intake, and so that's not it. In fact, the new intake idles stronger. Also, I am running different fuel rails than before, but the same existed with the Aeromotive as now with the Steeda ones.

I'm just very concerned about taking anymore datalogging runs like I've been doing with the AFR out of balance like it appears. What I'm wondering is what stones I'm leaving unturned..? For instance, is it possible that the internal wastegate on the passenger side is functioning different than the driver's side thereby affecting the AFR reading (not that actual AFR in the cylinder)? Also, is it possible or probable in your mind that a cam anomaly might possibly cause a varied reading, and if so wouldn't any anomalies show up when the cams were dialed in..?

I'd like to sweep over any other fixes while I'm waiting for the return of the ECU, including re-degreeing the cams if that appears to be a suggested or otherwise prudent step..?

Thanks for the help...

 

[turbo2nr]

I thought the LM-1 did not work well pre-turbo. Too much pressure and heat?

They have a forum for the LM-1 on the web.

 

[Otto J]

You may also want to check your coils,Maybe you have a cylinder not firing very well on the drivers side.

 

[Spud]

Ignition, injectors and LM-1 (LC-1 actually). I've replaced virtually every coponent of the ignition now except for the ECU which is at FAST to be upgraded to SEFI and ICC. I've had the injectors out, swapped sides and also had them flowed and tested and cleaned professionally, and they were perfect. Lastly, the LM-1 (LC-1) is not pre-turbo. It's downstream and in fact both the Innovative and FAST wide bands are downstream.

The best guess at this point is something to do with the batch firing in the FAST. The gap between sides is wider at idle and narrows as throttle is applied. I am still contemplating switching to COP because that's all I can think of to try completely different than what's in it now, and I understand that the COP can be wired up to run as a "waste spark".

Keep looking for magic clues regarding FAST issues that cause similar anomalies, and any suggestions for other possible culprits.

Thanks

 

[Spud]

I would also check the exhaust system for leaks, and the injector harness to be sure that it is wired correctly for the firing order of the engine. I have seen many people just throw the harness on the engine without checking to see if it is wired correctly. Yes, FAST does make mistakes at times, I know it is hard to believe.

Let us know what you find out.

Turns out you were right about the firing order being wired incorrectly. I have also been communicating with Craig and after he prompted me to double check the firing order, I took a closer look. I bought the FAST unit from a guy who supposedly was running it on a 4.6 Cobra and so I didn't question the firing order of the injector harness and simply wired it by the colors of the wires to the corresponding cylinders and didn't pay any attention to the pinout on the harness assuming it was a 4.6 harness. Well, it turns out I have the standard FAST harness that's for most GMs and SB Chryslers, and the pinouts are of course much different. The really odd thing is though I got my ECU back from FAST with the SEFI and ICC upgrades and it ran great, even with the mismatched harness and wrong firing order on the injector harness. I've now re-wired the harness so that the pinouts ABCDEFGH directly match 65481372 which is the 4.6 wiring according to FAST for the 4.6 (firing order 13726548).

I spoke to Ron at FAST today and he said that the car would have still run with the firing order completely wrong, which I would have never guessed, but it more than likely explains the anomaly with the passenger side running leaner than the driver's side. I still can't believe I've been running it for so long with a GM harness never questioning it and only checking the wire colors not realizing the colors mean something completely different with respect to the pinout of the harness. Looking at the injector plugs, they were even etched with numbers matching the GM and SB chry specs! Oh well, glad I caught it now, but I wouldn't have even looked if not for your promptings as well as Craig, and so thanks very much guys...

Spud

 

[Buzzard]

Glad to help

I am glad that you found your problem, just sorry that it took so much of your time to track it down. Sometimes its the little things that get overlooked with a problem like yours. Happy to help. Now go run the crap out of that Ford and post some times or video up for us.