-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Strange regen braking problem - shaking
- Thread starter teslick2
- Start date
I have a 2013 model S and have been having this exact same problem for about 2 years. It only happens in the winter ( I live in Northern California), only in the morning, only in the first few miles, only when the car is just off the charger and has been charged to 80% or more. I live on a hill so the first few miles engage the regen quite a bit. Shaking gets quite violent if I just let it go. I usually stop it by touch the accelerator lightly. Touching the brakes lightly doesn’t help. I have captured log files for the service center but they don’t see anything. I asked them to look into their internal database for any engineering notes or technical service bulletins but they found nothing.
Recently, my HV battery has failed. I think there is one module with one shorted cell but they don’t repair the battery, only replace it with a reconditioned battery. Trying to decide if I should invest in the new battery or junk the car. Considering that the Bluebook on the car is only $20k at this point, before I invest $16kin a battery, I would kind of like to know that the vibration isn’t a major defect that is going to get worse after I replace the battery. On the other hand, if the problem goes away with the new battery, that would probably be an interesting data point.
Anyone have any further data or insights on this problem since the original posting?
Recently, my HV battery has failed. I think there is one module with one shorted cell but they don’t repair the battery, only replace it with a reconditioned battery. Trying to decide if I should invest in the new battery or junk the car. Considering that the Bluebook on the car is only $20k at this point, before I invest $16kin a battery, I would kind of like to know that the vibration isn’t a major defect that is going to get worse after I replace the battery. On the other hand, if the problem goes away with the new battery, that would probably be an interesting data point.
Anyone have any further data or insights on this problem since the original posting?
I’m fairly confident that the shaking is being caused by whatever is limiting the regen braking. I experienced the issue again last week when I charged my car to 98% and regen was reduced. I also experience it when the car is left in the cold and the battery is cold and regen is reduced.
Any engineers here that can guess or explain why?
VegarHenriksen
Member
You might want to consider reaching out to someone that can repair the HV battery instead of doing a full replacement.
VegarHenriksen
Member
I think this has to do with the motor being cold and the system trying to find out how much or little power can be provided to or from the battery pack. This only happens when regen is limited such as cold mornings or high state of charge. It may be logic that isn't operating perfectly smooth, but it's such an edge-case and probably only affects the old Model S drive units that there is no point in Tesla to investigate.
I would love to hear from Model 3 owners if this happens to them, but because the 3 uses different motors it might not.
Don't quote me on this. I'm just trying a logical explanation based on my experiences as I do experience this very often in the mornings living in the hills.
What's super interesting is that I can mimmick this behavior when accelerating as well. If the motor is cold and the system is trying to heat up the battery while providing power to the motor it might give off a pulsating effect. This happens when the throttle is kept fairly steady while accelerating and not accelerating to the power limit. The pulsating or shaking effect can be stopped by pressing the pedal more or letting go of it a little bit.
I think this is just the way the component heating system works in the old Model S. There is a lot of factors to take into account as far as where it is sending power and how much.
I'd love to hear from someone at Tesla, but being based in hotter climates I doubt we'll ever hear from an engineer that has actually experienced this, thought about it and has the technical knowledge of why it happens.
I would love to hear from Model 3 owners if this happens to them, but because the 3 uses different motors it might not.
Don't quote me on this. I'm just trying a logical explanation based on my experiences as I do experience this very often in the mornings living in the hills.
What's super interesting is that I can mimmick this behavior when accelerating as well. If the motor is cold and the system is trying to heat up the battery while providing power to the motor it might give off a pulsating effect. This happens when the throttle is kept fairly steady while accelerating and not accelerating to the power limit. The pulsating or shaking effect can be stopped by pressing the pedal more or letting go of it a little bit.
I think this is just the way the component heating system works in the old Model S. There is a lot of factors to take into account as far as where it is sending power and how much.
I'd love to hear from someone at Tesla, but being based in hotter climates I doubt we'll ever hear from an engineer that has actually experienced this, thought about it and has the technical knowledge of why it happens.
Do you know or have a reason to think this is not drive shaft coupling related?
I forget what they call the couplings. I think of them as CV joints. If the effect is definitely a judder rather than surging (slower changes in acceleration / deceleration) then I'd look up some drive shaft coupling threads.
Why especially in the cold? May be old grease is thicker or temp changes tolerances in a worn coupling. Does steadily driving around a corner to change the weight balance have any effect?
Braking is applied 'after' the drive shafts so depending on the amount of brake application and the steepness off gradient braking unlikely to change load on the couplings.
Just an idea to keep in mind. Might be completely wrong of course.
I forget what they call the couplings. I think of them as CV joints. If the effect is definitely a judder rather than surging (slower changes in acceleration / deceleration) then I'd look up some drive shaft coupling threads.
Why especially in the cold? May be old grease is thicker or temp changes tolerances in a worn coupling. Does steadily driving around a corner to change the weight balance have any effect?
Braking is applied 'after' the drive shafts so depending on the amount of brake application and the steepness off gradient braking unlikely to change load on the couplings.
Just an idea to keep in mind. Might be completely wrong of course.
2013 MS85 74k miles RevQ LDU, RevB original battery pack in colder hilly Pacific North West. Always parked in garage and charged to 70% SOC.
Have this issue for a couple of years now (not sure exactly when started). 2 steep downhills near home. 1 short (1/2 block) and 1 long (1/4 mile). Whenever regen is severely limited from cold weather and first use after long park, first hill regen will be fine, 2nd long hill will start out okay but as soon as regen limit starts to taper off by software, shaking starts. Obviously hard to reproduce with Tesla SC if they don't have long downhills nearby and after cold soaking the car overnight.
I think @VegarHenriksen analysis in post #25 is correct. Here is my analysis
- All cars mentioned in this thread seems to be LDUs (large drive units, all RWD, all P*Ds). Not sure what battery pack everyone is on and the battery internal conditions. Also unknown what everyone's LDU coolant leakage collateral damage situation is (coolant seal leaks often, coolant gets aerosolized and rust rotor/stator and travel to the inverter eventually damaging that) I rebuild my LDU recently. No visible inverter damage (although coolant had just started to pool there) and slight rust on stator and rotor surfaces (cleaned up slightly) but should not affect magnetic field.
- Generic regen literature mentions a power dumping resistor to protect the battery when regen needs to be limited. Our Tesla's don't have this. I'd imagine anything that can dump 30kW+ of regen power is fairly large and require liquid cooling. Here is a thread, read the size of locomotive's 60kW regen dump capability in post #3 Consistent regen performance with load dump resistor - idea | Tesla Motors Club
- This might be due to firmware update at some point a couple of years back OR aging of some system. Age of vehicles mentioned in this thread seems older. Perhaps most on older battery packs. LDUs perhaps a mix of ages given the frequent failure/replacements.
My Guess How these MS regen works
I'm not EE (CS background) so here is my best layman EE explanation. Pure EEs can do much better I'm sure.
- Car accelerates when battery energy is applied to generate electromagnetic field (EMF) in the stator. Its the EMF phase difference (north/south pole along the rotational circumference) from rotor that cause the rotor to rotate.
- During regen, the mechanical force from tire/drive gears generate EMF but in the opposite way (phase?). And inverter forward the generated energy to the battery.
- To limit region, can do the following 2 things.
A. First just to apply energy to advance the stator phase. When it matches the rotor's acceleration from going downhill. There is no regen. The gap between them will be how much regen is allowed. This of course will cost some energy but not much since just rotating the rotor and not driving the car.
B. Apply frictional brakes. This slows down the motor rotor driven by the tires/drive gears and once again, closing the phase gap between the rotors rotation against the stator EMF in the windings. Tesla has updated firmware (I believe for 3 and Y only at this time?) to do this. Tesla updates car software to start applying regular brakes when regenerative is limited | Electrek
So Tesla's software is deciding to adjust stator EMF phase to limit regen and seems to have some kind unstable adjustment behavior. We can manually override this by applying the accelerator ourselves just slightly ahead of the software controlled regen limit and the shaking will go away (Not a good work around of course to accelerate downhill without friction brakes) So I vote its Tesla's firmware updates. However, only stats we have in this thread are older cars so could be some electronic hardware wear somewhere.
Here is an explanation on Roadster regen The Magic of Tesla Roadster Regenerative Braking | Tesla
BTW, my battery has 22mV imbalance which I'm trying to figure out how to rebalance. My normal battery usage pattern avoid high SOC (I'm at 70%) and never go >=90% at supercharger (slow charging rate), and avoid constant coolant pumps running (> 90% SOC). So my car has several years of use without ever triggering rebalancing (triggers at >= 93% SOC according to early day testing by @wk057) Don't know if this is imbalance is involved in this regen shaking problem. Just noting for full info.
Have this issue for a couple of years now (not sure exactly when started). 2 steep downhills near home. 1 short (1/2 block) and 1 long (1/4 mile). Whenever regen is severely limited from cold weather and first use after long park, first hill regen will be fine, 2nd long hill will start out okay but as soon as regen limit starts to taper off by software, shaking starts. Obviously hard to reproduce with Tesla SC if they don't have long downhills nearby and after cold soaking the car overnight.
I think @VegarHenriksen analysis in post #25 is correct. Here is my analysis
- All cars mentioned in this thread seems to be LDUs (large drive units, all RWD, all P*Ds). Not sure what battery pack everyone is on and the battery internal conditions. Also unknown what everyone's LDU coolant leakage collateral damage situation is (coolant seal leaks often, coolant gets aerosolized and rust rotor/stator and travel to the inverter eventually damaging that) I rebuild my LDU recently. No visible inverter damage (although coolant had just started to pool there) and slight rust on stator and rotor surfaces (cleaned up slightly) but should not affect magnetic field.
- Generic regen literature mentions a power dumping resistor to protect the battery when regen needs to be limited. Our Tesla's don't have this. I'd imagine anything that can dump 30kW+ of regen power is fairly large and require liquid cooling. Here is a thread, read the size of locomotive's 60kW regen dump capability in post #3 Consistent regen performance with load dump resistor - idea | Tesla Motors Club
- This might be due to firmware update at some point a couple of years back OR aging of some system. Age of vehicles mentioned in this thread seems older. Perhaps most on older battery packs. LDUs perhaps a mix of ages given the frequent failure/replacements.
My Guess How these MS regen works
I'm not EE (CS background) so here is my best layman EE explanation. Pure EEs can do much better I'm sure.
- Car accelerates when battery energy is applied to generate electromagnetic field (EMF) in the stator. Its the EMF phase difference (north/south pole along the rotational circumference) from rotor that cause the rotor to rotate.
- During regen, the mechanical force from tire/drive gears generate EMF but in the opposite way (phase?). And inverter forward the generated energy to the battery.
- To limit region, can do the following 2 things.
A. First just to apply energy to advance the stator phase. When it matches the rotor's acceleration from going downhill. There is no regen. The gap between them will be how much regen is allowed. This of course will cost some energy but not much since just rotating the rotor and not driving the car.
B. Apply frictional brakes. This slows down the motor rotor driven by the tires/drive gears and once again, closing the phase gap between the rotors rotation against the stator EMF in the windings. Tesla has updated firmware (I believe for 3 and Y only at this time?) to do this. Tesla updates car software to start applying regular brakes when regenerative is limited | Electrek
So Tesla's software is deciding to adjust stator EMF phase to limit regen and seems to have some kind unstable adjustment behavior. We can manually override this by applying the accelerator ourselves just slightly ahead of the software controlled regen limit and the shaking will go away (Not a good work around of course to accelerate downhill without friction brakes) So I vote its Tesla's firmware updates. However, only stats we have in this thread are older cars so could be some electronic hardware wear somewhere.
Here is an explanation on Roadster regen The Magic of Tesla Roadster Regenerative Braking | Tesla
BTW, my battery has 22mV imbalance which I'm trying to figure out how to rebalance. My normal battery usage pattern avoid high SOC (I'm at 70%) and never go >=90% at supercharger (slow charging rate), and avoid constant coolant pumps running (> 90% SOC). So my car has several years of use without ever triggering rebalancing (triggers at >= 93% SOC according to early day testing by @wk057) Don't know if this is imbalance is involved in this regen shaking problem. Just noting for full info.
Last edited:
Similar threads
