I got a 2006 GXP with only about 31000 miles on it. Automatic, no upgrades. Bone stock. When you are going downhill, accelerating from a stop, at about 20 mph i let off the gas and the trans clunks right into 3rd gear it seems then immediately goes back down to second gear. But only at that speed and it seems only downhill. I took it to the dealer and they took it for a ride and they didn't have anything go wrong with it. Any ideas? We bought it with 7500 miles a few years ago. What gives?
camaro_iroc87
Tranny Clunk??????
So is this normal for all Solstices basically? Can anyone tell me if it's been recalled? If so that would help out alot. Or where can I find a recall list for my 07 GXP?
There is a lengthy thread about transmission clunk. If you are getting very hard downshifts when slowing in 2 nd or 3 rd, it may be due to a couple of different sources. One source is due to one or both cam position servos failing. There are several instances of hard or clunking shifts that went away following the replacement of a cam positioned. The incipient or intermittent failure may be difficult for the dealer to diagnose.
The second source appears to be a result of the transmission going into protection mode. The transmission is adaptive, that is as the wear surfaces in the tranny change dimension, the transmission modifies it's operating parameters to retain it's factory new performance. But to do this, the transmission control module computer communicates with the engine control module. The ECM sends predictive info on power input to the TCM while the TCM measures actual input power. The TCM compares it's measured input with the predicted power from the ECM . When they match, all is well. If there is a small miss match the TCM uses the data to predictvhow it should operate or adapt. If there is a mismatch that exceeds pre set parameters, the TCM goes into safe or protected mode and you get hard shifts, generally the kind that "put me through the windshield...".
The dealers will generally not duplicate the problem or say it is normal.
Also, it appears that it is possible to get the TCM into protective mode by driving the car very hard in undulating or very twisty terrain
To figure out if you have an actual failure driven condition or just a sofrware anomaly, try rebooting the car by pulling the negative terminal for 30 minutes then drive it normally. If the clunk or hard shift is gone and stays gone, then it is likely a software data mismatch that accumulated. And sadly it is normal.
If the clunk is not gone, or returns soon, then start looking fro engine problems. in some cases a MAF sensor cleaning fixed the problem. In several others, a sticking cam positioned needed to be replaced. In these instances, the problems source was the ECM saw an undiagnosed fault, put out bad predictive data to the TCM which reacted as designed. Until the engine problem was addressed, the clunking remained.
Good luck
The second source appears to be a result of the transmission going into protection mode. The transmission is adaptive, that is as the wear surfaces in the tranny change dimension, the transmission modifies it's operating parameters to retain it's factory new performance. But to do this, the transmission control module computer communicates with the engine control module. The ECM sends predictive info on power input to the TCM while the TCM measures actual input power. The TCM compares it's measured input with the predicted power from the ECM . When they match, all is well. If there is a small miss match the TCM uses the data to predictvhow it should operate or adapt. If there is a mismatch that exceeds pre set parameters, the TCM goes into safe or protected mode and you get hard shifts, generally the kind that "put me through the windshield...".
The dealers will generally not duplicate the problem or say it is normal.
Also, it appears that it is possible to get the TCM into protective mode by driving the car very hard in undulating or very twisty terrain
To figure out if you have an actual failure driven condition or just a sofrware anomaly, try rebooting the car by pulling the negative terminal for 30 minutes then drive it normally. If the clunk or hard shift is gone and stays gone, then it is likely a software data mismatch that accumulated. And sadly it is normal.
If the clunk is not gone, or returns soon, then start looking fro engine problems. in some cases a MAF sensor cleaning fixed the problem. In several others, a sticking cam positioned needed to be replaced. In these instances, the problems source was the ECM saw an undiagnosed fault, put out bad predictive data to the TCM which reacted as designed. Until the engine problem was addressed, the clunking remained.
Good luck
Thanks for the input rob. When I get home in December I'll try reseting the computer. I'm in afghan right now. Anyway thanks again.
We have 50 cars in our club and all the autos have experienced this problem at one time or another. So far, none have been due to diagnosed failures and they have over time relearned out of safe mode. But when we get them back into the twisties and they get on it hard, they go back into safe mode and have the hard downshifting.
I talked with Dave at DDM and he imparted the information about the TCM and ECM comparing data.
Others have contributed the info about repairing the cam position servo failures.
Its all out there in other threads.
The bad news is that unless you have a mechanical failure - something like the MAF is failed or dirty or one of the cam servos is bad, it is "normal" in that it is how the code works. I suspect its an unintended consequence of reusing existing code.
I talked with Dave at DDM and he imparted the information about the TCM and ECM comparing data.
Others have contributed the info about repairing the cam position servo failures.
Its all out there in other threads.
The bad news is that unless you have a mechanical failure - something like the MAF is failed or dirty or one of the cam servos is bad, it is "normal" in that it is how the code works. I suspect its an unintended consequence of reusing existing code.
So do the dealerships know about this problem? or is there some kind of new tune that's out that no longer allows that?
go read the thread on Transmission Suckage
GM does not recognize this as a problem.
The logic goes like this as far as I can intuit
If the transmission was not adaptive, then over time there would be noticable degredation in performance as the internal parts wore.
People would be unhappy with the noticable degredation and be causing problems for GM demanding that it get fixed, when there is really nothing wrong with it, its just natural wear and it still functions, just not quite as crisply as it did before the wear occurred.
So given that adaptive is required, you have to be concerned about over or under adapting. Stated another way, the transmission controller will do its thing to make the car go while it assesses ongoing, cumulative wear and adapts itself to compensate for that wear. But the transmission is only looking internally at its own data for operation and wear assessment. It has no outside calibration point.
So the design goes outside the TCM and compares its data with that provided by the ECM. The ECM is controlling the engine room and can be treated as "ground truth" with respect to input torque and HP. So this input from the ECM can be used to help the TCM keep track of where it is at with respect to its internal assessment of the need to adapt, and act as a sanity check if you will.
But the design has to take into account the condition where the ECM and TCM data are not matching or there are other unanticipated problems with the transmission. Rather than leave you walking because the tranny can not figure out its actual adaptive state, they built in an "if all else fails, make it work" safe mode so if the worst happens you can still drive the car. Unfortunately that safe mode really hammers you with hard downshifts and is not user friendly.
So if you have a condition where the TCM compares its data with that coming from the ECM and the delta exceeds a set value(s), it MUST go into safe mode.
The TCM and ECM data can go out of whack because there is a problem with the TCM, or there is a problem with the ECM (or the engine), or there is evidence that very aggressive driving can also drive the TCM into safe mode.
The ECM data can be a problem because there is a real failure or because the ECM can not figure out its inputs and throws its virtual hands in dispair.
Hard failures that affect the ECM's ability to figure out where its at based on inputs means that any of the input sensors can fail or give bad data from being dirty.
Or the ECM can THINK something is wrong when it adjusts paramaters based on what the sensors are telling it and does NOT get the resultant changes in performance that were demanded.
So if you have bad sensors or just dirty ones, the ECM can have a problem. Cleaning them should return the ECM to a happy state.
If you have a bad "active" device, like the spark on each cylendar, or the cam position sensor or actuators . .. basically anything that changes engine paramaters can fail and you have good data on out of acceptable range conditions.
So . . .having said all that . . .
GM says its normal because
Going into safe mode is normal and necessary. You may not like it but that is normal too.
A failure that is sensed by or that affects the ECM output is also "normal" because its due to a failed component, and the TCM / ECM are reacting to the failure "as designed" even though we dont like how it does it.
If you cause the TCM to go into safe mode by driving very aggressively, GM can make the claim that is normal too . .. and to a certain extent they are correct. Basically I dont think the TCM can tell the difference between some failure modes that drive the ECM out of whack and "non failure" driven safe mode entry. Stated another way, the TCM can see some types of hard, driving in undulating terrain as a potential failure and has NO CHOICE but to go into safe mode. And I think they address this condition by allowing the TCM to exit safe mode over a number of key cycles as it "unlearns" the safe mode by operating with good data for ~ 6 keycycles.
Or you can reboot the TCM and it will reboot in normal mode and dissremember that it was in safe mode as that is apparantly its default start up mode.
So . . . . unfortunately the safe mode entry without a failure IS normal as far as GM is concerned currently.
I suspect it will take an after market tuner to eventually address this challenge, and it is a challenge. If you cause the TCM to NOT enter safe mode then you could be left standing on the side of the road due to wear that would have still functioned in safe mode. Or you may get the transmission into a mode that actually increases wear significantly or even potentially will damage it in some manner.:willy::thumbs::willy:
GM does not recognize this as a problem.
The logic goes like this as far as I can intuit
If the transmission was not adaptive, then over time there would be noticable degredation in performance as the internal parts wore.
People would be unhappy with the noticable degredation and be causing problems for GM demanding that it get fixed, when there is really nothing wrong with it, its just natural wear and it still functions, just not quite as crisply as it did before the wear occurred.
So given that adaptive is required, you have to be concerned about over or under adapting. Stated another way, the transmission controller will do its thing to make the car go while it assesses ongoing, cumulative wear and adapts itself to compensate for that wear. But the transmission is only looking internally at its own data for operation and wear assessment. It has no outside calibration point.
So the design goes outside the TCM and compares its data with that provided by the ECM. The ECM is controlling the engine room and can be treated as "ground truth" with respect to input torque and HP. So this input from the ECM can be used to help the TCM keep track of where it is at with respect to its internal assessment of the need to adapt, and act as a sanity check if you will.
But the design has to take into account the condition where the ECM and TCM data are not matching or there are other unanticipated problems with the transmission. Rather than leave you walking because the tranny can not figure out its actual adaptive state, they built in an "if all else fails, make it work" safe mode so if the worst happens you can still drive the car. Unfortunately that safe mode really hammers you with hard downshifts and is not user friendly.
So if you have a condition where the TCM compares its data with that coming from the ECM and the delta exceeds a set value(s), it MUST go into safe mode.
The TCM and ECM data can go out of whack because there is a problem with the TCM, or there is a problem with the ECM (or the engine), or there is evidence that very aggressive driving can also drive the TCM into safe mode.
The ECM data can be a problem because there is a real failure or because the ECM can not figure out its inputs and throws its virtual hands in dispair.
Hard failures that affect the ECM's ability to figure out where its at based on inputs means that any of the input sensors can fail or give bad data from being dirty.
Or the ECM can THINK something is wrong when it adjusts paramaters based on what the sensors are telling it and does NOT get the resultant changes in performance that were demanded.
So if you have bad sensors or just dirty ones, the ECM can have a problem. Cleaning them should return the ECM to a happy state.
If you have a bad "active" device, like the spark on each cylendar, or the cam position sensor or actuators . .. basically anything that changes engine paramaters can fail and you have good data on out of acceptable range conditions.
So . . .having said all that . . .
GM says its normal because
Going into safe mode is normal and necessary. You may not like it but that is normal too.
A failure that is sensed by or that affects the ECM output is also "normal" because its due to a failed component, and the TCM / ECM are reacting to the failure "as designed" even though we dont like how it does it.
If you cause the TCM to go into safe mode by driving very aggressively, GM can make the claim that is normal too . .. and to a certain extent they are correct. Basically I dont think the TCM can tell the difference between some failure modes that drive the ECM out of whack and "non failure" driven safe mode entry. Stated another way, the TCM can see some types of hard, driving in undulating terrain as a potential failure and has NO CHOICE but to go into safe mode. And I think they address this condition by allowing the TCM to exit safe mode over a number of key cycles as it "unlearns" the safe mode by operating with good data for ~ 6 keycycles.
Or you can reboot the TCM and it will reboot in normal mode and dissremember that it was in safe mode as that is apparantly its default start up mode.
So . . . . unfortunately the safe mode entry without a failure IS normal as far as GM is concerned currently.
I suspect it will take an after market tuner to eventually address this challenge, and it is a challenge. If you cause the TCM to NOT enter safe mode then you could be left standing on the side of the road due to wear that would have still functioned in safe mode. Or you may get the transmission into a mode that actually increases wear significantly or even potentially will damage it in some manner.:willy::thumbs::willy:
Okay. I disconnected the battery and waited till the next morning to reconnect it and the trans is still clunking around at low speeds. what else could give???
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
