EcmSpy/DDFI Experts - Please Help!

[CBi]

I have been chasing down a horrible misfire problem with my 06 Firebolt (Race Exhaust, Race ECM) that has developed over the last few months. I threw a few parts at it (sensors) and checked the grounds. I did find a not-so-clean ground (some paint behind it) but it did not fix my problem.

I decided to get my ecmspy datalogging up and running in an effort to better diagnose my problem. While I have yet to fix the problem, I know exactly what is causing it....quickly soaring EGO correction while in closed loop. With the bike warm, I can hold it at a high idle, watch it go into closed loop and start jacking up the injector firing time until the bike starts to gag on the overly rich mixture. It has the same behavior at cruising speeds.

One thing I have noticed during the datalogs is the 02 voltage never goes very far past 0.58v (even when the EGO correction starts to climb).

I read about another Bueller on this board with almost the exact same problem, who traced it down to a wiring issue. Suspect of the wiring in my bike, I decided to diag the 02 wiring. First, I tapped into the 02 wire with my DMM, and watched the voltage, which was showing a rich mixture (fluctuating around 0.8v). Then I checked again right at the ECM connector....same readings, although the reading in ecmspy would fluctuate in unison with the dmm, just never going past 0.58v. I decided to run a new wire for the 02 sensor into the ECM (eliminating the harness in the bike). No change in behavior.

So now I'm left with two culprits to the problem. Some sort of programming/setting issue, or a malfunction with the ECM itself.

Before I buy a new ECM, is there anything I can look at in ecmspy? (yes I have reflashed to the original EEPROM settings)

I ran across virtually this same issue with a BMW a few years ago (02 voltage in ECM not reading correctly, but behaving well enough not to throw any codes).

I realize I can't fix the ECM, but would a failing A-D converter cause something like this? I'm just curious....it seems as though the 02 voltage is just not scaled correctly.

I really hope some fellow buellers can help out here. I just want to get back to riding!

Edit - I'm currently running the bike with the 02 turned off...no issues, except not running nearly as well as it should.

 

[SoCalXB12R]

I think we may have the same problem with our XBs.

My bike is ruuning very rich, fouling plugs quickly, and running like crap at constant cruising speed.

I get an engine light sometimes, at cruising speed. It blinks a couple times then goes away. I've got it hooked up to VDST standard software, which is telling me "engine trouble" but no engine code. Hmmm...

Noticed this weekend that the AFV is at 143. its suppossed to be at like 100. Can't change the AFV settings with this software, only view it (worthless crap!). if you have ECMspy I heard you can make those adjustments. You may want to look into that, and tell me how it goes.

I haven't changed out my O2 sensor yet (cuz it looks like a painful process) but will look into that more now.

I'm running a Micron exhaust w/ EBR race ECM. Just installed the EBR, hoping that would solve the problem (had an OEM race ECM, bought off EBay, on there for about a year, operating great, then one day this problem happened. HD techs suggested replacing the ECM. Didn't solve the problem. Now I have 2 race ECMs and the problem still exists...).

I'm curious to see how you solve your XB issue, cuz it will probably solve mine as well. Good luck and keep me posted!

 

[dallasb]

@CBi: Do you have a stock ecm to see if that fixes your problem? I know this sounds simple but have you checked your spark plugs and wires? Also what year is your bike? Also where did you get the eeprom? If it was downloaded maybe the file is corrupt.

@SoCal: What year/model of bike do you have? I think you can use ecmspy to reset your AFV's back to 100. Do that and go for a ride and see if they climb back up to 143. If they do then I would suspect a faulty sensor.

 

[Warlock]

+1 on what dallasb said.

 

[SoCalXB12R]

dallasb:

04 XB12R. I posted another question (ECMspy / Direct Link???) that addressed the ECMspy issue. using VDST standard and not able to reset AFV, only view it. Trying to get ECMspy to work with the same cable, if possible.

Are you referring to the O2 sensor?

 

[dallasb]

Sorry didn't realize it was the same person...we can just use your thread...

 

[SoCalXB12R]

I give up with the ECMspy! I was able to reset the AFV w/ my VDSTS, with the default setting be 100. I ran it for a while (on a stand) going through all the gears, and watched the AFV go as high as 150! It was bouncing all over the place, BUT never went below 100. Engine light started blinking a few times, but still no toruble code.

Is anyone else thinking O2 sensor, or is there something else I may have missed?

 

[dallasb]

Yea replace the O2 sensor. It will cost around 20 bucks probably.

 

[SoCalXB12R]

got one on order< hopefully waiting at the house when I get home today. Now for the big question... drop the engine to get to the sensor or drop the swingarm and go through the fan. Both seem like a pain in the ass to do, but which one will hurt less?

 

[dallasb]

I didn't think that you needed to drop the swing arm to get to the O2 sensor. There was a thread somewhere on here with step by step instructions. I've never done it so can't give any first hand experience with replacement.

 

[SoCalXB12R]

well it wasn't the o2 sensor. replaced it with a brand new (not ebay new...) sensor over the weekend. Reset the AFV, didn't do anything to the TPS (cuz its been set, reset, and re-reset), and same problem continues...

FYI replacing O2 sensor via the fan/swingarm access was so much easier than dropping the engine (which I did a few years back when installing the Micron). My advice is to remove the subframe completely and disconnect only 1 bolt for the shock. I took a few pictures, and will try to get them on soon.

 

[SoCalXB12R]

so what costs less then trailering my bike to and from pensacola, and rolling the dice with a HD mechanic? replacing all the sensors....

next is the engine temp sensor. I read that if it fails it can cause the rear cylinder to cut out when overheated. Perhaps thats whats going on with my bike??? Any suggestions, or someone that has replaced that sensor? looks like a simple task...

 

[ich]

make a log file. the only thing that really matters is what the ecm is sensing, which is then shown in the log file.

 

[CBi]

I ended up having a grounding problem with my bike. I figured it out by using ecmspy.

With the bike warm, I could hold it at around 2k rpm, and watch the bike go into closed loop. The EGO correction would begin to skyrocket, richening the mixture until the bike would start to misfire and studder. A quick snap of the throttle would send the ECM into open loop, and the EGO correction would go back to 100, and the bike would run fine, until the next time it would go into closed loop, at which point it would repeat the whole process.

Looking at the ecmspy datalogs, I noticed the 02 sensor never read above about 0.50v. Based on the above behavior, the 02 sensor should have produced more voltage (a rich mixture). So I began to diag the 02 wiring. First I backprobed the ECM with a DMM, to verify that ecmspy was reporting the actual 02 voltage (it was). Then, I ran a new 02 sensor wire directly to ECM, bypassing the harness. Same readings. Replaced the sensor, no change. Then, I started to trace the ground path of the sensor. I dug through the service manual and researched all of the grounds for the ECM, and discovered Buell did not "daisy chain" the ECM grounds (common in the automotive industry). There are only two grounds for the ECM itself, and they attach to the seat frame rail on Lightnings, and to the headstock area of the frame on Firebolts (which was my problem). In my case, I had enough of a ground to run everything, but it was "dirty" enough to provide enough resistance to insulate the 02 sensor signal voltage, which was causing my engine performance problem.

Since the 02 was producing a "lean" reading, the ECM would continuously "up" the fuel injector pulse time, which SHOULD produce a rich 02 reading, at which point the ECM would respond with a slightly shorter injector pulse. Since the 02 was not reporting correctly, the ECM would keep increasing the injector pulse until it would reach it's maximum correction, which was literally gagging the engine.

If you have the tech skills to setup a datalog with your bike, I can take a look at it and let you know if your issue is similar to mine.

 

[autoprotolysis]

I've got an 05 xb12r with pipe and KN filter. I have the same issue with the misfiring at cruising speed and crap idle. My EGO readout on ECM Spy is from .07 to .8 up and down very fast. Is that supposed to be that way? I think the xb12's seem to all have this issue. Thanks for any input. I will hopefully go for datalog ride this aftertnoon.

 

[CBi]

Wow, necro-thread revival!

Make sure the grounds on the frame near the steering shaft are clean and tight.

ECM Spy is a great diagnostic tool (especially the datalogs), if you have good fundamentals on how an EFI system operates. The best advice I can give you is to read the ECM Spy documentation thoroughly, to get a good grasp of how the Buell system works.