PV tune, lost the O2 sensors

[05Train]

Ok, any thoughts here?  Bike's a 2012 CVO Ultra with a Jackpot headpipe.

I put the Jackpot mufflers back on today (which are WAY too freaking loud on this bike) and did a tuning run.  1st run, no problem except that the VEs exceeded 127' so it automatically "scaled" the displacement up to bring the VEs down into range.  Once I did that, the O2 sensors stopped reading.  I pulled the codes, and both of them were offline (P01 32 & P01 54).  Also, it was showing that the O2 was "shorted high" (P0032).  I cleared the codes, and nothing changed.  I stopped tuning and rode for a few hours, got home, cleared the same codes again, and started the bike in the garage.   P0032 showed up again (I only idled it for a minute). 

I checked the O2 leads, they're fine and properly plugged in.   Tomorrow, I'll put the Fatshotz on, restore back to a tune I know works, and try again.  I'm thinking that maybe by increasing the displacement I boned something up.  Weird thing is that the bike ran fine (43 mpg), though it was clear that there was still more tuning to do with those mufflers. 

The only changes to the bike are a new air cleaner ( I got the Chisel one that matches the wheels), and the dealer replaced the stereo and installed a hidden antenna.  I can't remember if I checked to see if there were codes before, but if I did, there weren't any. 

Oh, and FWIW, the tuning I did with the stock mufflers worked great, the bike runs like a quiet beast with them. 

[mayor]

does the ve tables reduce when you selected the auto scale?  it sounds like the ve's may be to high, which could flood the sensors and prevent them from getting good readings. 

[mayor]

can you post both the before and after maps from when you used the scaled function? 

[05Train]

does the ve tables reduce when you selected the auto scale?  it sounds like the ve's may be to high, which could flood the sensors and prevent them from getting good readings. 

They're supposed to.  Since I'm resigned to not using these mufflers, I'm going to mount a (hopefully) quieter pair and reload a map I know works.  Hopefully that does it.  If not, I'll get the maps posted this evening.

[05Train]

My laptop's acting up, so no maps at the moment (my kingdom for a tuner that runs on a Mac). 

I restored the stock tune.  Same codes.  I restored to a tune that wasn't scaled.  Same codes.  I'm gonna unhook the negative battery terminal overnight and see if that changes anything.   If not, I'm thinking that the ECM somehow got corrupted and shut the sensors down. 

Luckily the FM map I have seems to work well enough with the Fatshotz, so it's not killing me running in open-loop. 


Sent from my iPad, probably while I'm pooping.

[ViennaHog]

The P0032 code refers o the new style sensor heating circuit. If they short out, the sensor is pretty much out of service. I ASSume you produced a short somewhere in the circuit for the sensor when fiddling with the headpipes. You can check for this code in the EDM. Let me do a search and I post the routine

[ViennaHog]

deleted

[ViennaHog]

wiring


[attachment removed after 60 days by system]

[strokerjlk]

does the ve tables reduce when you selected the auto scale?  it sounds like the ve's may be to high, which could flood the sensors and prevent them from getting good readings.

if there is a 126.5 in the ve table, pv will ask if you want to scale or cap.
if you scale ,it increases the ci 5%. so yes it will add fuel everywhere.
when the pv asks "scale or cap" you need to look at the ve table and see whats going on.
is it just 1 cell?
is it several cells,and where at? it's a judgement (tuning) call.
if it's just 1 or 2 cells . i would cap and then see where it goes next session.
if it is several cells in the sweet spot , then scale and re tune.

[rbabos]

I have a base call with plenty of 127s in an area that I will never hit. These force the scale option to be displayed but not necessarily warranted.  I could have changed the 127s to 120 without any ill effects and never saw the scale option. I chose to just ignore them being unused and capped the whole tune, never even coming close to those 127 cells during autotuning.  Otherwise it would have kept scaling like forever and needlessly throwing fuel at it. See, If I never hit those unused areas in autotune, it would never stop asking me to scale. Never.
Now if the 126+ are in the actual operating areas, then scaling is a good idea since autotune will hit them and adjust. The requirement to scale will stop eventually.  PV don't that these different locations into account. It sees 126+ anywhere in the cal and it bitches to scale and in my case it would have been the wrong choice with never ending ci increases.
Ron

[05Train]

OK, finally got the tune out of my POS Windows laptop..

https://www.box.com/s/6r1aq03ymuake9dam0cu

That's the one I created right before the O2 sensors stopped talking to the PV or the bike.  Since then, I've unplugged the ECM, restored to the stock calibration, and tried everything again...No joy.  The PV shows the O2 sensor voltage, and that's still reading (IOW, the O2 sensors are still functioning).  Jamie at FM is going to forward everything to Dynojet to see if they can make sense of it, but for whatever reason the bike seems to be stuck in open-loop.

[05Train]

Here's the calibration I was tuning from....

https://www.box.com/s/3u7hmlvmy437pne8a57s

[FLTRI]

When the software realized and helped you correct the "running out of VE" issue it took the entire fueling calibration and enrichened everywhere.
This caused areas of the calibration to become so rich it doused the O2s.
This is common so the areas around the corrected maxed area must be reduced so it doesn't foul the O2s.

After more than 15 years EFI tuning Harley's I can say with great confidence:
There are no 100% auto tune systems...period.

Biggest issue closed loop systems have is exhaust market who provides us with unlimited choices, whether properly designed or not. So when O2 sensors are relied upon for fueling corrections and they get mixed messages or too rich or lean to read, the system can't work correctly.
Bob

[05Train]

When the software realized and helped you correct the "running out of VE" issue it took the entire fueling calibration and enrichened everywhere.
This caused areas of the calibration to become so rich it doused the O2s.
This is common so the areas around the corrected maxed area must be reduced so it doesn't foul the O2s.

After more than 15 years EFI tuning Harley's I can say with great confidence:
There are no 100% auto tune systems...period.

Biggest issue closed loop systems have is exhaust market who provides us with unlimited choices, whether properly designed or not. So when O2 sensors are relied upon for fueling corrections and they get mixed messages or too rich or lean to read, the system can't work correctly.
Bob

Ummm, I never stated that this was an auto tune system, and it didn't adjust anything, I did.  When the ECM was reflashed, by coincidence or not, the ECM stopped reading the O2 sensors.  The exhaust is set up just fine for the O2s.

If, in fact, the sensors were "doused" then what?  My gas mileage is still in the low 40s (this is on a 110), I'm still getting varying voltage readings as they're reacting to the exhaust gas mix, and I've returned to a leaner map.  Are you saying that once they've been "doused", they'll continue to send voltage readings to the ECM, but the ECM won't do anything with the information?

I'm confused.

[rbabos]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

[05Train]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

[rbabos]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Heater failure for both is what I'm reading out of this. P0132 & P0134 took too long to heat up and function so the ecm gave up on them=code , which sorta point back to power for both heaters not available. Having two faulty sensors would be pretty slim.
The B1006 is overvoltage generally caused from faulty voltage regulator. Have you checked charging voltage with a multimeter? This one would concern me the most.  ViennaHog's schematic would be a good way to track down where in the heater circuit the problem is.
Ron

[05Train]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Heater failure for both is what I'm reading out of this. P0132 & P0134 took too long to heat up and function so the ecm gave up on them=code , which sorta point back to power for both heaters not available. Having two faulty sensors would be pretty slim.
The B1006 is overvoltage generally caused from faulty voltage regulator. Have you checked charging voltage with a multimeter? This one would concern me the most.  ViennaHog's schematic would be a good way to track down where in the heater circuit the problem is.
Ron

OK well that's screwed up, since the voltage regulator was just replaced under the recall.

I'll see what FM and DJ come back with, after that it's to the dealer.  This sucker's still under warranty.

[Jamie Long]

When the software realized and helped you correct the "running out of VE" issue it took the entire fueling calibration and enrichened everywhere.
This caused areas of the calibration to become so rich it doused the O2s.
This is common so the areas around the corrected maxed area must be reduced so it doesn't foul the O2s.

Bob, you are absolutely incorrect with this comment. When PV scales the VE tables it levelizes them, meaning it not only changes the displacement constant and lowers the VE in an area where it has reached the high limit of 127.5,  it also does the math and applies a correction to the rest of the VE table as well. 

[Jamie Long]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Based on the info provided and the tunes you emailed me it scaled the VE table slighly due to O2 correction on the rear cylinder. It simply scaled VE 4% so VE was within the high limit and leveled the rest of both tables. I have taken a look at everything and have also forwarded info to Dynojet, I do not like to speculate without having seen the bike first hand, however from what I have seen you may likely have an electrical issue affecting the O2 heater circuit, in this regard it is also somewhat interesting that the voltage regulator was just replaced. When I have more info I will be in touch and you are also welcome to contact me with any new info you find.

[05Train]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Based on the info provided and the tunes you emailed me it scaled the VE table slighly due to O2 correction on the rear cylinder. It simply scaled VE 4% so VE was within the high limit and leveled the rest of both tables. I have taken a look at everything and have also forwarded info to Dynojet, I do not like to speculate without having seen the bike first hand, however from what I have seen you may likely have an electrical issue affecting the O2 heater circuit, in this regard it is also somewhat interesting that the voltage regulator was just replaced. When I have more info I will be in touch and you are also welcome to contact me with any new info you find.

When I get home today, I'm going to pull the codes from the instrument panel rather than the PV.  I'm curious to see if they're there, especially considering the fact that the engine light isn't lit.

Regardless, I'll give the dealer a holler when I get home. 

I suspect y'all are right and it's an unrelated issue that's taken out the O2s.

[rbabos]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Based on the info provided and the tunes you emailed me it scaled the VE table slighly due to O2 correction on the rear cylinder. It simply scaled VE 4% so VE was within the high limit and leveled the rest of both tables. I have taken a look at everything and have also forwarded info to Dynojet, I do not like to speculate without having seen the bike first hand, however from what I have seen you may likely have an electrical issue affecting the O2 heater circuit, in this regard it is also somewhat interesting that the voltage regulator was just replaced. When I have more info I will be in touch and you are also welcome to contact me with any new info you find.

When I get home today, I'm going to pull the codes from the instrument panel rather than the PV.  I'm curious to see if they're there, especially considering the fact that the engine light isn't lit.

Regardless, I'll give the dealer a holler when I get home. 

I suspect y'all are right and it's an unrelated issue that's taken out the O2s.

That is interesting that no current and historic code shown in the ip with engine light.
Ron

[FLTRI]

When the software realized and helped you correct the "running out of VE" issue it took the entire fueling calibration and enrichened everywhere.
This caused areas of the calibration to become so rich it doused the O2s.
This is common so the areas around the corrected maxed area must be reduced so it doesn't foul the O2s.

Bob, you are absolutely incorrect with this comment. When PV scales the VE tables it levelizes them, meaning it not only changes the displacement constant and lowers the VE in an area where it has reached the high limit of 127.5,  it also does the math and applies a correction to the rest of the VE table as well.

Sorry if I offered misinformation here.

Can you offer a logical reason for why reflashing a VE limit-corrected PV calibration causes O2 system failure?
Bob

[05Train]

Weird. It should never have asked you to scale with that cal as I see it.
As for the codes. It's possible there's a power failure to the sensors themselves. Ecm tries to light them but can't throwing the code. PV sees them working but ecm won't read them if the heaters are not active. Don't think it's the sensors themselves but power supply for the heaters. Ecm plug, wiring or whatever.
Ron

The first link is the scaled one, displacement was bumped to 114ci.  I'm not familiar enough with the PV to surmise why it did so.

The heater code is thrown first.  If I just start the bike and let it idle in the garage, it only throws the P0032 code (and bizarrely B1006).  The P0132 & P0134 only show after I ride.

Based on the info provided and the tunes you emailed me it scaled the VE table slighly due to O2 correction on the rear cylinder. It simply scaled VE 4% so VE was within the high limit and leveled the rest of both tables. I have taken a look at everything and have also forwarded info to Dynojet, I do not like to speculate without having seen the bike first hand, however from what I have seen you may likely have an electrical issue affecting the O2 heater circuit, in this regard it is also somewhat interesting that the voltage regulator was just replaced. When I have more info I will be in touch and you are also welcome to contact me with any new info you find.

When I get home today, I'm going to pull the codes from the instrument panel rather than the PV.  I'm curious to see if they're there, especially considering the fact that the engine light isn't lit.

Regardless, I'll give the dealer a holler when I get home. 

I suspect y'all are right and it's an unrelated issue that's taken out the O2s.

That is interesting that no current and historic code shown in the ip with engine light.
Ron

Well, I'm not sure yet, though the engine light isn't on.  I know on the Night Train, the light comes on for any code.

Bob - It now seems rather apparent that this isn't an issue with the PV, it was just a coincidentally-timed failure.

[Coyote]

I thought you did not get a check engine light for an O2 sensor error. At least not on newer bikes. I know I can unplug mine and get a diagnostic error code and not have the check engine light come on.

[rbabos]

I thought you did not get a check engine light for an O2 sensor error. At least not on newer bikes. I know I can unplug mine and get a diagnostic error code and not have the check engine light come on.

Quite possible but the overvoltage should have lit it up, I'd think
Ron

[05Train]

I thought you did not get a check engine light for an O2 sensor error. At least not on newer bikes. I know I can unplug mine and get a diagnostic error code and not have the check engine light come on.

Quite possible but the overvoltage should have lit it up, I'd think
Ron

Well, we'll find out when I get home.  I can't imagine that the PV would be picking up codes that the bike doesn't see, but stranger things have happened.

[Jamie Long]

Sorry if I offered misinformation here.

Can you offer a logical reason for why reflashing a VE limit-corrected PV calibration causes O2 system failure?
Bob

The specific cause if his O2 system issue has not yet been pinned down, however your assumption that "a VE limit-corrected PV calibration causes O2 system failure" is certainly premature. Again, while I do not like to speculate without seeing the bike I feel there may be an underlying electrical issue. During a phone conversation yesterday with the OP he noted that while logging it dropped O2 feedback near the end of his tuning session, when in AT Basic mode the ECM is doing business and usual. It did not actually happen when it was reflashed.

One other consideration is because late model ECM's do not set a CEL for O2 codes there is a possibility that this issue may have been happening previously, you would not have known unless you actively read back the codes. It would however showed up as an error when running AT Basic. 

[FLTRI]

I thought you did not get a check engine light for an O2 sensor error. At least not on newer bikes. I know I can unplug mine and get a diagnostic error code and not have the check engine light come on.

 
Bob

[FLTRI]

The specific cause if his O2 system issue has not yet been pinned down, however your assumption that "a VE limit-corrected PV calibration causes O2 system failure" is certainly premature. Again, while I do not like to speculate without seeing the bike I feel there may be an underlying electrical issue. During a phone conversation yesterday with the OP he noted that while logging it dropped O2 feedback near the end of his tuning session, when in AT Basic mode the ECM is doing business and usual. It did not actually happen when it was reflashed.

One other consideration is because late model ECM's do not set a CEL for O2 codes there is a possibility that this issue may have been happening previously, you would not have known unless you actively read back the codes. It would however showed up as an error when running AT Basic.

Just tossing out the issue the OP stated O2 operation was stable until he reflashed the modded calibration.
If that is not correct

When the software realized and helped you correct the "running out of VE" issue it took the entire fueling calibration and enrichened everywhere.
This caused areas of the calibration to become so rich it doused the O2s.
This is common so the areas around the corrected maxed area must be reduced so it doesn't foul the O2s.

After more than 15 years EFI tuning Harley's I can say with great confidence:
There are no 100% auto tune systems...period.

Biggest issue closed loop systems have is exhaust market who provides us with unlimited choices, whether properly designed or not. So when O2 sensors are relied upon for fueling corrections and they get mixed messages or too rich or lean to read, the system can't work correctly.
Bob

Ummm, I never stated that this was an auto tune system, and it didn't adjust anything, I did.  When the ECM was reflashed, by coincidence or not, the ECM stopped reading the O2 sensors.  The exhaust is set up just fine for the O2s.

If, in fact, the sensors were "doused" then what?...

I believe you have an PV AT system?
Just noting that when the modified VE-adjusted calibration was reloaded the O2 system ceased to operate properly.
Try reloading the original, non limit adjusted and see if the sensors come back to life. If they've been run too rich for a while it will take a while to clean them back up.
Bob

[Jamie Long]

The OP is running Power Vision with AT Basic app using the factory O2 sensors. First thing we had him do was go to Restore Original on the PV which restores the complete original binary which is automatically backed up as a mirror image to the same state as before PV was married, he also flashed the earlier PV calibration where the bike was operating properly. The O2 sensors do not respond with either of them. Also no reason the bike would have been running rich as you continue to suggest, especially considering it was in closed loop right up until this issue presented itself

[strokerjlk]

I believe you have an PV AT system?
Just noting that when the modified VE-adjusted calibration was reloaded the O2 system ceased to operate properly.
Try reloading the original, non limit adjusted and see if the sensors come back to life. If they've been run too rich for a while it will take a while to clean them back up.
Bob

closed loop Lambda range was set to allow it to toggle lean/rich,

[05Train]

At the dealer having the charging system checked.  Results as I get them.

[05Train]

No news today.  Hopefully tomorrow. 


Sent from my iPad, probably while I'm pooping.

[05Train]

Quick update....The bike's still at the dealer.  They think they've narrowed the overvoltage code to a faulty sensor in the speedometer.  New speedometer is on the way, hopefully I'll have a concrete answer early next week.

[joe_lyons]

I have seen a few times that the speed sensor was failing and causing it to throw some weird codes that affected other sensors using a 5 volt reference signal.  It was easy to test unplug sensor and see if problem persists.

[05Train]

Finally got some movement on this.

They still can't figure out the overvoltage, but they're thinking it might be the Custom Dynamics LED turn signals and load equalizer (I can't buy that since they worked fine on my Limited).  But they swapped out the ECM and the O2 codes disappeared.  So, new ECM coming.

[05Train]

And finally, resolution.   

New ECM installed, no more codes.  Best I can figure, something (or some things) got really wet when I rode through TS Andrea and hosed up the ECM.  Dynojet was kind enough to issue me a new license for the PV so it won't cost me any more to set it up with the new ECM.   Jamie at FM was a huge help through this. 

[joe_lyons]

Awesome glad to hear you got taken care of.