scsharp2260
Member
Thanks we have tried all that; I am aware of that notice, but as you say the most that is seen 1000 to 1800 RPM. We see 4000 and once 5000 RPM.
2009 Road Glide Excessive High Idle
- Thread starter scsharp2260
- Start date
Hoople
Account Removed
When you experienced the problem, were you ever able to regain correct rpm and control of the engine without re-booting the ECM? Note that in the OP's description, once the RPM was "run away" at 4000 RPM, the ECM had to be rebooted to correct it. That is of key importance because it means the twistgrip potentiometer has completely lost sync with the TPS and can't reestablish itself, and at the same time the ECM was actually happy with itself even when the twist grip pots did not match the TPS pots. Note that he could instruct the ECM to go higher in RPM but not lower unless he re-booted the system. It was now calling 4000 rpm a base of zero. This shows that the ECM was not just "hung" at a fixed or default RPM and could be told to go even higher so the core CPU was still running. This is not good because even if this is a "1 off case" out of the thousands of bikes made, it means the ECM does not have an emergency set of default values (similar to what Ford uses with their throttle by wire) whenever sync is lost. It also means it won't be an easy issue to correct.
I generally accept and welcome new technology with the exception of throttle by wire. Sure, I know the reasons 'why' they insist on wanting it on vehicles but sometimes the small amount of bad out weighs the large amount of good. My opinion is it has no place or reason on a motorcycle.
Re: When you experienced the problem, were you ever able to regain correct rpm and control of the engine without re-booting the ECM?
No. Once the high idle occurred and manifested itself, the engine would never return to a normal 1000 rpm + 50 or -50 situation So every time it was turned off and then restarted the high idle was learned and thus continued. The only successful solution I ever came up with was to do the 4 on and 4 off "reset" or reboot to relearn TGS position. I remember reading TT 380 (June 12, 2008)where the factory did come up with a reflash for the ECM to aleviate this but it was only as a temporary measure while they worked on a more permanent fix to the software/hardware. It was not to be done unless a customer requested it. I never heard of a more permanent fix from the engineers. Since the reflash was only listed as a temp measure, I never had it done.
I think it is important here for owners to let their engines idle down for 15 seconds or so before shutting them off so that a high idle situation is not introduced into memory. The original poster here is having problems of high idle of 4000 rpms so I am unsure if this is related or not.
No. Once the high idle occurred and manifested itself, the engine would never return to a normal 1000 rpm + 50 or -50 situation So every time it was turned off and then restarted the high idle was learned and thus continued. The only successful solution I ever came up with was to do the 4 on and 4 off "reset" or reboot to relearn TGS position. I remember reading TT 380 (June 12, 2008)where the factory did come up with a reflash for the ECM to aleviate this but it was only as a temporary measure while they worked on a more permanent fix to the software/hardware. It was not to be done unless a customer requested it. I never heard of a more permanent fix from the engineers. Since the reflash was only listed as a temp measure, I never had it done.
I think it is important here for owners to let their engines idle down for 15 seconds or so before shutting them off so that a high idle situation is not introduced into memory. The original poster here is having problems of high idle of 4000 rpms so I am unsure if this is related or not.
Hoople
Account Removed
The actual RPM value of your example and the OP's example are really not what makes them different or similar. The important key is in your example, the ECM will continue on from where it left off and will continue on steeping through lines of code even after a re-boot. After you reboot the idle remains high. In the OP's example, the CPU restarts at the entry point of the code. That is why his idle is now corrected to normal. That is a Big Difference and I would say (just my opinion) that these two failures have nothing in common even though the end result is a high idle in both.
Your problem is and would be so much easier to correct. You can actually go into the assembly language of the software and step through it, find where it goes wrong (when it refers to a stored RPM value) and correct it. With the OP's problem you would have to take a snap shot (memory dump) of the software code at the time of failure. Then POSSIBLY you could step backward in the code to see how you arrived to the point of crash. But with the OP's, all the memory is cleared on re-boot so a memory dump must be done on the fly and at the time of the crash..
Lets hope what the OP is experiencing is just a 1 out of 100,000 fluke and can be corrected with an ECM, throttle body assembly or twist grip pots. I am also glad to see that all HD models did not go to throttle by wire in 1 calender year. That could have been the kiss of death. At least if it does turn out to more than a 1 time fluke it will be containable.
Grillfish
Junior Member
Keep us posted on progress. Mine too is TBW, but I only get a few DTCs thrown once in a while (correlation error..). I couldn't imagine the engine reving up to 4000+ on it's own. Sounds so dangerous. As stated before, I'd start a note book of who you contacted, what they did (nothing!), when this happened, dates, times, etc... I'd also send a registered letter to HD corp informing them of the situation requesting immediate response.
scsharp2260
Member
The worst part about this is because it is my wife's bike they automatically discount what she is saying. They are happy about women buying bikes but they still don't think they know anything. This last time I watched the one writer actually roll his eyes as my wife told them what was happening. I think I am going to just drop in there, I am afraid they are just giving her lip service and not looking at the bike at all. And yes I intend to get Harley involved in this and any one else I think could help me.
ironmark
Junior Member
Is the bike showing any codes & are you turning the bike off when the rpm's reach 4000 ? If the bike is showing codes you may have to erase the codes for the bike to run properly ! Even if you do the 4 on 4 off ( I use the 5 on 5 off myself ) method to cured the problem a exsiting code may make the problem happen again . Shutting the bike down @ 4000 will put the rpm speed into the ecm memory and the bike will resume the previous idel speed @ start up . If you use the kill switch to shut the bike down your hand may be moving the throttle some and that will effect the idel speed , let the bike idel down completly & then turn the bike off with the ignition switch only and that may eliminate your issues . On my 08 RG I had the dealer reflash the ecm and cancel the ide memory .
Hoople
Account Removed
I believe it is a much deeper problem and here's why.
Keep in mind I have not read any white pages associated with the HD ECM monitor strategy but what really worries me about this entire topic of throttle by wire is the apparent ECM architecture that was chosen by HD.
It is easier to tell of one proven Throttle by Wire strategy that has worked in the past and then compare it to what we have. 2005 was the 1st year for throttle by wire for the Ford Mustang GT. Since many were being sold, the last thing Ford wanted was lots of run away 300HP engines. For safety, a monitor Strategy (hardware & software) was used within the ECM. The monitor system in the ECM was distributed across TWO processors. The main powertrain control processor (nicknamed Black-Oak) and a second monitoring processor called an Enhanced-Quizzer processor (nicknamed E-Quizzer).
Independent Plausibility Check software (IPC) was embedded in the main Processor (Black Oak). It was responsible for comparing drivers foot peddle position to the torque actually generated by the engine (distance traveled vs time). Since main control of the engine and IPC software shared the same CPU, potential common failures could exist. That was the reason a separate E Quizzer chip was added. It was the watchdog of the entire system. If it saw an impaired BlackOak decision, it would take failure mode action. In most cases that would be fuel injection shut down.
Mechanical foot peddle control consisted of 3 pots and the DC throttle motor feedback consisted of 2 pots. All 3 peddle pots must match & send the same readings for the software to obey the command. The IPC software then looked at peddle commands and compared them to actual delivered engine torque. Even if the 3 peddle pots matched, if the expected delivered torque did not match, the E-Quizzer stepped in.
Hence if the foot peddle was pushed down only a 1/4" and the delivered torque was 180 ft/lbs, that would be considered an impossibility and the Quizzer would step in and over-ride the powertrain CPU and shut down the fuel injectors. The E-Quizzer performed watchdog over BlackOak and IPC software and looked for non logical decisions of either. (Can you see any of this helping our Imported friend in their time of trouble?)
Now compare this to what we have in the OP's situation. Assuming you believe the OP when he says the twist grip is at Zero and the RPM is at 4000 RPM (which I do), then this is what we have. Double redundancy throttle pots with both pots sitting on zero. Most likely 2 throttle motor feedback pots sitting on a value that is OTHER than Zero. A tach input signal to the ECM that says the engine IS going 4000 RPM. A ground speed reading of zero mph or near zero. Possibly a front or rear brake switch on. And the CPU sees nothing wrong with any of the above. It does not even generate a check engine code or historical code. Yes a bad set of twist grip pots could cause the above but keep in mind the average mean failure rate for control pots may be 1 in 10,000. With double redundancy we would have at least a 10K X 10K simultaneous mean average failure rate.
Yes possible but highly unlikely.
It is not the fact that there is a mechanical malfunction with the above. The problem will eventually be repaired by replacing some part. The real concern I have is that the system does not recognize that a problem even exists. There seems to be No E-Quizzer or form of E-Quizzer. There seems to be no IPC software or form of IPC software. There seems to no watchdog or form of watchdog at all. That's what worries me.
Keep in mind I have not read any white pages associated with the HD ECM monitor strategy but what really worries me about this entire topic of throttle by wire is the apparent ECM architecture that was chosen by HD.
It is easier to tell of one proven Throttle by Wire strategy that has worked in the past and then compare it to what we have. 2005 was the 1st year for throttle by wire for the Ford Mustang GT. Since many were being sold, the last thing Ford wanted was lots of run away 300HP engines. For safety, a monitor Strategy (hardware & software) was used within the ECM. The monitor system in the ECM was distributed across TWO processors. The main powertrain control processor (nicknamed Black-Oak) and a second monitoring processor called an Enhanced-Quizzer processor (nicknamed E-Quizzer).
Independent Plausibility Check software (IPC) was embedded in the main Processor (Black Oak). It was responsible for comparing drivers foot peddle position to the torque actually generated by the engine (distance traveled vs time). Since main control of the engine and IPC software shared the same CPU, potential common failures could exist. That was the reason a separate E Quizzer chip was added. It was the watchdog of the entire system. If it saw an impaired BlackOak decision, it would take failure mode action. In most cases that would be fuel injection shut down.
Mechanical foot peddle control consisted of 3 pots and the DC throttle motor feedback consisted of 2 pots. All 3 peddle pots must match & send the same readings for the software to obey the command. The IPC software then looked at peddle commands and compared them to actual delivered engine torque. Even if the 3 peddle pots matched, if the expected delivered torque did not match, the E-Quizzer stepped in.
Hence if the foot peddle was pushed down only a 1/4" and the delivered torque was 180 ft/lbs, that would be considered an impossibility and the Quizzer would step in and over-ride the powertrain CPU and shut down the fuel injectors. The E-Quizzer performed watchdog over BlackOak and IPC software and looked for non logical decisions of either. (Can you see any of this helping our Imported friend in their time of trouble?)
Now compare this to what we have in the OP's situation. Assuming you believe the OP when he says the twist grip is at Zero and the RPM is at 4000 RPM (which I do), then this is what we have. Double redundancy throttle pots with both pots sitting on zero. Most likely 2 throttle motor feedback pots sitting on a value that is OTHER than Zero. A tach input signal to the ECM that says the engine IS going 4000 RPM. A ground speed reading of zero mph or near zero. Possibly a front or rear brake switch on. And the CPU sees nothing wrong with any of the above. It does not even generate a check engine code or historical code. Yes a bad set of twist grip pots could cause the above but keep in mind the average mean failure rate for control pots may be 1 in 10,000. With double redundancy we would have at least a 10K X 10K simultaneous mean average failure rate.
Yes possible but highly unlikely.
It is not the fact that there is a mechanical malfunction with the above. The problem will eventually be repaired by replacing some part. The real concern I have is that the system does not recognize that a problem even exists. There seems to be No E-Quizzer or form of E-Quizzer. There seems to be no IPC software or form of IPC software. There seems to no watchdog or form of watchdog at all. That's what worries me.
scsharp2260
Member
Well the dealer just called and told us to pick up the bike, when we asked was it fixed they would only say they replaced the throttle position sensor. When my wife asked was that all they said that's all it needed. I hope so; but I don't think it's going to be that easy.
J
Jack Klarich
Guest
I hope your problem is fixed, If not back to the dealer, this could be just the beginings of serious problems with throttle by wire, I dont know why there are so many problems with this , It seems to have infected our bikes as well as our cages, i see them every day in our shop and Im wondering who will be next, Good Luck and keep us all posted on your results, jack