Pack Performance and Launch Mode Limits

[hostman]

I agree. Incomplete info and fear and uncertainty and doubt are the usual causes of pitchforks and jihad. That is the human condition -- for many anyway. Others prefer to wait for data and evidence and specific real information. The energy and enthusiasm for this topic shld be directed to acquiring more info, not making conclusions that are simply not supported by the evidence to date.

Until we know what the counter limits are, and what the remedy is (eg replacing battery pack to get less fatigued wire bonds?) all the upset generated on this thread is based on ignorance.

I'm simply not offended by the creative design choice to control the aging and fatigue of key components. It would preserve the longevity , including well past the warranty period. I suspect that if the design were implemented through Hardware rather than software people would be a lot less offended. For instance the total battery capacity already the grades based on the function of use. More 100% charges and 0% discharges the grave over time based on the amount of use, the battery capacity. It just so happens that that is implemented in the hardware rather than in the software.

suppose the battery pack output also degraded not just as a function of full recharging Cycles but as a function of high output discharge Cycles. Would everybody be waging holy war against that Hardware design choice?



Seriously think that out. If someone really wants a 10 second sedan or SUV EV, there are no other options. If someone wants 300 mile range EV there are no other options.

If I was buying a race car to take to the strip every day, id guess i wld pause and realize that if certain (still unknown!) events occur if have to replace parts that degraded or bypass them and face the consequences.

I also take comfort in the fact that like the AP functionality, these software choices can be continually refined. Nothing is set in stone. But let's still see exactly what the current facts are.

Hold on just a moment. The battery losing capacity is an inevitable consequence of physics, for all battery chemistries identified to-date. That is completely different than a software limiter to compensate for a hardware error, which is what this smacks of. And I have proven this issue is real, on my own vehicle.

 

[hostman]

You again completely miss the point. This is just a engineering design choice. Did they disclose all their design choices to you? E.g., How often the batteries degrade relative to Full Charge and Zero Charge events?

It's quite simple.

I will concede however, that it would have been better for them to not even bother with Launch Mode since it doesn't provide any benefit, and apparently stresses the powertrain more than Foot Pedal Mode. I do feel bad for people like hostman who used Launch Mode indiscriminately, without knowledge of the stresses and the resulting engineering choice to mitigate that stress over the lifetime of the battery (or other powertrain component).

I also suspect that they are not ready to disclose exactly what those engineering choices are because since they are implemented through software, they can change them, just as they do AP design choices. Let's see what they eventually settle on (for the time being) and what those choices are. But need more info to determine what they are.

To be fair, I am an automotive engineer, and I have extensive experience in electric powertrains. Unless there is a design error there is no reason for this limitation. You'll note the non-P models are sufficiently-designed to not need this.

 

[drewsny]

To be fair, I am an automotive engineer, and I have extensive experience in electric powertrains. Unless there is a design error there is no reason for this limitation. You'll note the non-P models are sufficiently-designed to not need this.

Exactly. Is it not fair to assume that if tesla would have known of this before hand they would have designed for it to not happen? I mean there's a reason why tesla has larger motors for the rear of P cars instead of just pushing the power through the smaller motors and limiting the power later on when it starts to wear more aggressively.

 

[hostman]

To put things in perspective, Ludicrous mode on my vehicle was installed end of March and v8.0 came in end of September and disabled it. So that's $1000/mo I paid for the function. Give me a break. No other performance car in the world does such things.

The only time I can recall this was even *attempted* was a software adjustment to the launch mode of the GT-R. And Nissan ended up replacing people's $20k transmissions for free due to the backlash.

 

[hostman]

Exactly. Is it not fair to assume that if tesla would have known of this before hand they would have designed for it to not happen? I mean there's a reason why tesla has larger motors for the rear of P cars instead of just pushing the power through the smaller motors and limiting the power later on when it starts to wear more aggressively.

Correct. I think the issue has been nailed down to the bonding wires inside the pack. However, I don't really see how limiting power is going to help them, unless there is a significant thermal stress cycling issue. Most solutions to this kind of problem involve thermal limiting, which was already present in the pack controller.

 

[drewsny]

Correct. I think the issue has been nailed down to the bonding wires inside the pack. However, I don't really see how limiting power is going to help them, unless there is a significant thermal stress cycling issue. Most solutions to this kind of problem involve thermal limiting, which was already present in the pack controller.

Easy fix then. Either refund the purchase of ludicrous mode for everyone or offer an "at cost" upgrade to a 100kwh battery pack. At cost meaning tesla also buys back your current pack at a fair price.

 

[AnonNJ]

I just reread the notes I typed into Keep while speaking with the SM last Saturday. I copied them nearly verbatim here but left out what appears to be one critical word. They are protecting bond wire "welds".

Anyone else get their digits yet?

 

[malcolm]

They are protecting bond wire "welds".

Thanks for noting this.

Well at least it's not a cell chemistry issue- which would be a longer-term fix (I know, no comfort to present owners)

If true, it's still a gigantic headache.

 

[AWDtsla]

Thanks for noting this.

Well at least it's not a cell chemistry issue- which would be a longer-term fix (I know, no comfort to present owners)

If true, it's still a gigantic headache.

The labor to replace the cell bonds means you might as well change the whole pack, so for the existing cars cell chemistry doesn't matter. However for P100D moving forward, if they still haven't nailed the cell chemistry, then there could certainly be yet another limiter to protect the packs. Now if they couldn't figure out a wire and a weld, you think the cell chemistry is going to be an easier engineering exercise?

 

[vgrinshpun]

The labor to replace the cell bonds means you might as well change the whole pack, so for the existing cars cell chemistry doesn't matter. However for P100D moving forward, if they still haven't nailed the cell chemistry, then there could certainly be yet another limiter to protect the packs. Now if they couldn't figure out a wire and a weld, you think the cell chemistry is going to be an easier engineering exercise?

You should stop dishing out on poor "wire bonds" engineering. Even if they are the problem (although from technical point of view it is highly unlikely - this is probably case of SC technician opining about subject he/she does not fully understand - knowing enough to be dangerous) it (problem) is NOT in poor engineering - it is inherent.

It is counterintuitive for non-electrical engineers, but problem with the counters and extracting maximum amount of power for short duration of time from the powertrain that is rated around 69kW continuous duty is not in poor engineering, but in engineering that was excessively sophisticated...

 

[smac]

At cost meaning tesla also buys back your current pack at a fair price.

Heck Tesla could use them in software limited mode for either 90D's or 75's...

 

[malcolm]

Now if they couldn't figure out a wire and a weld, you think the cell chemistry is going to be an easier engineering exercise?

I think the weld issue is an emergent problem which has only come to light relatively recently.

IIRC the 2170 cells have both electrodes at the same end so the connector design will have been revisited.

Plus I think the cell chemistry has been modified too.

But again, no comfort to current owners.

 

[lolachampcar]

I would encourage everyone to read bhzmark's posts carefully.
The poster is either trying to wind people up or she/he actually believes that what has been done "in software" is an acceptable practice. If that belief is real, my plan is to understand that point of view as best I possibly can as it likely mirrors the same type of thinking going on within the Tesla brain trust.

vgrinshpun,
The problem is most likely thermal cyclic fatigue of the per cell interconnect wires. Note the P100's change to a flex circuit. Also note that the BMS is doing the counting and the BMS broadcast max permissible current is what is being reduced.

What is counterintuitive for engineers (and likely anyone else) is trying to make sense of your last sentence.

 

[vgrinshpun]

I think the weld issue is an emergent problem which has only come to light relatively recently.

IIRC the 2170 cells have both electrodes at the same end so the connector design will have been revisited.

Plus I think the cell chemistry has been modified too.

But again, no comfort to current owners.

Interconnection design was most likely already addressed in P100D - by using flexible printed circuits.

Once again, I would not put high trust in conclusion that ''poor wire bonds" is the component that requires cutting back power output based on counters

 

[malcolm]

At cost meaning tesla also buys back your current pack at a fair price.

It would make an excellent Powerwall 1.

 

[AWDtsla]

Heck Tesla could use them in software limited mode for either 90D's or 75's...

Legally it must be sold as a used pack, so it can't just be sold in a new chassis as a new 75.

 

[bhzmark]

or she/he actually believes that what has been done "in software" is an acceptable practice. If that belief is real, my plan is to understand that point of view as best I possibly can as it likely mirrors the same type of thinking going on within the Tesla brain trust.

Thank you for trying to understand. Very admirable and also a breath of fresh air. And yes I am trying to think about this from a Tesla perspective and how they make their engineering design choices -- and not hung up on the distinction between hardware and software design. The functionality and the performance and longevity attributes are what they are, however they got there.

But I'm also primarily just saying that the data and other information is so thin so far, it just doesn't justify the outrage.

 

[smac]

Legally it must be sold as a used pack, so it can't just be sold in a new chassis as a new 75.

True, but it depends how it is dressed up. Would you like a brand new 75 pack, or a used 90 pack in a "75" for "85" money but still with an 8 year warranty.... who knows.

Let's be frank here, my car has had a replacement drive train. Sure as day follows night, it was a reconditioned unit. As long as it's warranted I really don't care.

It's time for Tesla to start thinking outside the box on how they head this off.

 

[malcolm]

Interconnection design was most likely already addressed in P100D - by using flexible printed circuits.

Once again, I would not put high trust in conclusion that ''poor wire bonds" is the component that requires cutting back power output based on counters

Did I read correctly that P100Ds also have counters?

Presumably if true, we don't know if these are present simply to record and report back the pack status as Tesla continues to monitor the new design in the field or if they are needed to modify pack output.

*sigh*

 

[stopcrazypp]

Did I read correctly that P100Ds also have counters?

Presumably if true, we don't know if these are present simply to record and report back the pack status as Tesla continues to monitor the new design in the field or if they are needed to modify pack output.

*sigh*

I'm going to guess that pretty much all the packs have a similar counter and that this counter existed before this limiter was implemented. It was just used for this purpose after engineering found out some component was wearing faster when doing Ludicrous launches.