Pack Performance and Launch Mode Limits

[jerjozwik]

i have a v2 pack. max logged power was 483kw.

 

[hostman]

i attempted to explain the situation to my petrol head co-worker. when i mentioned the whole "no one knows what driving style results in failed packs..." he responded with "well yah, we have 100 years of gasoline tech knowledge to pull from." i said "exactly, pretty much everyone knows not to drive around pegging the redline all day long."

Actually, new engines are often validated at the redline for 1M miles equivalent.

 

[St Charles]

Very good points and questions. We need this info to understand the scope of all this. Could it be we are all riding a warranty time bomb? I am guessing it is only the 1600 amp group. But as unethical as things have progressed? Who knows.

This is a good point and question. Is it that 1600A is just too much draw from these batteries? If such is the case, then wouldn't 1750A be too much on the 100kWh batteries as well?

I did some digging. I'm not claiming to know the math well enough but there are some interesting thoughts here. Tesla uses NCA batteries in their packs. These NCA batteries are well suited for capacity and cycle life. Something they are apparently not known for is high amp draw. Based on the following sites (SITE 1 - SITE 2) The C rating for these type of batteries hovers right around 1C. As far as batteries go, that's not very good. So, how much stress are we putting on out 90kWh packs at 1600A? If I did my math right it should be just north of 7C. That's 7x the rated amp draw. If you want to kill a Li-Ion battery, the fastest way to do so, short of physical damage, is to draw too much power all at once.

If this is true then it's not cooling or wire fuses. It's the Cells themselves giving up the ghost. This also means the problem has no real fix beyond staying away from 1600A.

thoughts?

 

[trayloader]

Maybe it's all much simpler in the end.
Only a new Battery and a fresh drivetrain can bear these loads.
And a fresh fuse with
fresh capable metal alloy without being overtaxed.
Then come the extremes.
Repeatedly.
Combine that with the hundreds of thousands of times partially discharging and charging by the magic "regen".
Maybe we are witnessing the diesel scandal equivalent in batteries.
It's maybe much more difficult to project "longevity" and "durability" here.
While our bright leader constantly teases us with faster and faster acceleration times. And "boring" fogbombs.
Fatigue.

 

[Andyw2100]

Isn't this the classic bait and switch?

Actually it is not.

The classic bait and switch involves "baiting" a customer to come in for an advertised item at a low price, and then when the customer arrives "switching" that customer to a higher priced item, saying the advertised item is not available.

I obviously do not condone Tesla's actions here, but it is not the "classic bait and switch."

 

[JRP3]

These NCA batteries are well suited for capacity and cycle life. Something they are apparently not known for is high amp draw. Based on the following sites (SITE 1 - SITE 2) The C rating for these type of batteries hovers right around 1C.

You can't necessarily extrapolate general specs about a chemistry to a specific version of that chemistry. Tesla/Panasonic are not using a generic NCA chemistry. Also momentary high C rates are not the same as continuous.

I would point out that if you expect any car to last well beyond the warranty period you don't drive it balls to the walls all day every day, be it ICE or EV.

 

[St Charles]

You can't necessarily extrapolate general specs about a chemistry to a specific version of that chemistry. Tesla/Panasonic are not using a generic NCA chemistry. Also momentary high C rates are not the same as continuous.

I agree with you but there are C values for momentary and continuous, published or not. Additionally, while there is no doubt that Panasonic/Tesla batteries are not the standard NCA chemistry, it is very unlikely that they are able to bump the safe discharge rate enough to account for what we are able to get from our cars now.

In the LiPo world, leaving batteries at extreme SoC will reduce capacity while abusing charge/discharge causes battery expansion and possible explosion/fire. The Li-Ion world is not soo different. We already understand and accept the rules for SoC. What has never been communicated is the effect of high amp draw from any Tesla battery pack. It wasn't until the P85D came out that this was even a consideration. Now that it is, we are seeing failing packs from owners who fancy a spirited drive more than sparingly.

I would point out that if you expect any car to last well beyond the warranty period you don't drive it balls to the walls all day every day, be it ICE or EV.

People are asking is what that means in a Tesla? ICE cars are well understood, EV cars are not. I can put my foot to the floor in an ICE car and not generally cause damage until I get to high RPM. That is not the case in EV as I can get to 1600A almost immediately. If this is what is killing the battery then Tesla has two real options: Make a pack that can tolerate the discharge rate or reduce the draw from the motors. They tried the second option already.

Replacing battery packs is not a long term solution.

 

[n2mb_racing]

Is the Power limitation outside of LM really only impacting the P90DL v3 owners that were doing 1600A?
or if you prefer are we seeing lower than 1500A when pedal mashing on all other non-1600A batteries (P90DL v1, v2, P85DL)?

No, it has limited my P90DL with the v2 battery. I used to get 490 kW, now I only get 465 kW when mashing the pedal. I can get about 485 kW, but only when using launch mode. It is limited to 465 kW at all other times. That means we will only get to use the maximum power very infrequently. I can't find many instances where I have time to set launch mode and rip off a perfect launch, but I have many opportunities to floor it while already rolling.

 

[St Charles]

No, it has limited my P90DL with the v2 battery. I used to get 490 kW, now I only get 465 kW when mashing the pedal. I can get about 485 kW, but only when using launch mode. It is limited to 465 kW at all other times. That means we will only get to use the maximum power very infrequently. I can't find many instances where I have time to set launch mode and rip off a perfect launch, but I have many opportunities to floor it while already rolling.

Is that 485kW with 100% SoC, Max Battery Ready, and Launch Mode?
How is it without MB?
And then how is just Ludicrous?

I know you have answered these in other posts, i'm just hoping to consolidate answers for posterity.

 

[davidc18]

for reference, on FW 7.1 at 90-95% soc I can get 489 (max battery enabled but not ready), all the time via pedal stomp.

 

[St Charles]

for reference, on FW 7.1 at 90-95% soc I can get 489 (max battery enabled but not ready), all the time via pedal stomp.

The same for me but I am a bit worried about a sudden battery failure leaving me on the side of the road while on a trip. Eventually I would like to upgrade.

 

[AnxietyRanger]

Actually it is not.

The classic bait and switch involves "baiting" a customer to come in for an advertised item at a low price, and then when the customer arrives "switching" that customer to a higher priced item, saying the advertised item is not available.

I obviously do not condone Tesla's actions here, but it is not the "classic bait and switch."

I appreciate the history lesson and obviously am on the same side of the debate as Anydw2100, but just one point about this. I think the language has evolved enough on this matter that this one still could be "classic bait and switch" in spirit (as in, of the same obvious nature), if not "classic bait and switch" literally, which I agree it is not per the above definition.

Certainly the words bait and switch nowadays are used to refer to marketing one thing and later it turning out to be something else. In that sense this is, if this remains the case, a "classic" example of that - you being lured in with one promise and then that promise morphs into something else to suit the seller.

What is unclear here is intent, of course, and timing. I doubt Tesla originally intended to deceive, which would be part of bait and switch, but it seems plausible they may have chosen to later on (after the sale). With the various software changes, certainly it can be argued they have bait and switched down the line...

I guess some might argue that is not bait and switching, let alone classic, if it happens after the fact and without original intent. The debate, no doubt, will rage on. Will the issue itself last or will Tesla really make it right?

 

[n2mb_racing]

Is that 485kW with 100% SoC, Max Battery Ready, and Launch Mode?
How is it without MB?
And then how is just Ludicrous?

Actually, looking again:

459 kW max with MB at 97% SOC and no launch mode (may not have been fully warmed up?)
480 kW max on average (one peak to 483 kW) with MBR and Launch Mode at 93-95% SOC.
465 kW max with MBR and no launch mode at 93-95% SOC

I'm not sure if I did any without MB turned on. I'm not sure what you mean by just ludicrous. Wouldn't that be the same as without MB? Or do you mean cold battery?

Previously, I would get 485 - 490 kW with MBR, with or without launch mode. I did back to back runs with and without launch mode and they were the same, +/- 1 kW.

I don't have data higher since it is hard to get to a place to test it without losing ~5%.

I'll do another run once my CAN cable gets here...

Edit: Graph of power loss below vs. software version.

 

[n2mb_racing]

The same for me but I am a bit worried about a sudden battery failure leaving me on the side of the road while on a trip. Eventually I would like to upgrade.

If bond wires are the problem, then I don't think it will be a sudden failure. My guess is exponential failure, where it gets worse and worse quickly, but not instantly. I think there will be dash warnings before it fails.

 

[jerjozwik]

The same for me but I am a bit worried about a sudden battery failure leaving me on the side of the road while on a trip. Eventually I would like to upgrade.

fret not, there is nothing sudden about it. i got the notification but the car was completely derivable. went about my 20 mile drive to the service center without any real change in behavior other then turning off ludicrous mode and setting regen to low.

 

[yak-55]

I beg to differ, check my signature ...but my buy timing might have been somewhat unique.

Since "Countergate Part 2/LM only" started, I don't remember reading about other P85DL owners attempt to see if Power is limited below 1500A without LM. I can't try much here in the current weather, by the time the battery reaches Max Heat, SOC will be well below 90%!
So I won't be able to contribute for another couple of months which only makes me even more curious!

Is the Power limitation outside of LM really only impacting the P90DL v3 owners that were doing 1600A?
or if you prefer are we seeing lower than 1500A when pedal mashing on all other non-1600A batteries (P90DL v1, v2, P85DL)?

Wondering if they just isolated the P90DL v3 owners a little more than before with this latest change...
This could also imply that owning a PxxDL outside warranty might be ok if it never was able to reach 1600A?

I would very much like to know this for the P85D(L) also. Please do collect the data !
Thanks.

 

[Tech_Guy]

If bond wires are the problem, then I don't think it will be a sudden failure. My guess is exponential failure, where it gets worse and worse quickly, but not instantly. I think there will be dash warnings before it fails.

I don't think the dash warnings were added until the recent update to "restore power". For example I never had a dash warning though my car had lost 40 then 100 hp the power was decreasing but I had no dashboard warning until the recent update...

 

[St Charles]

Actually, looking again:

459 kW max with MB at 97% SOC and no launch mode (may not have been fully warmed up?)
480 kW max on average (one peak to 483 kW) with MBR and Launch Mode at 93-95% SOC.
465 kW max with MBR and no launch mode at 93-95% SOC

I'm not sure if I did any without MB turned on. I'm not sure what you mean by just ludicrous. Wouldn't that be the same as without MB? Or do you mean cold battery?

Previously, I would get 485 - 490 kW with MBR, with or without launch mode. I did back to back runs with and without launch mode and they were the same, +/- 1 kW.

I don't have data higher since it is hard to get to a place to test it without losing ~5%.

I'll do another run once my CAN cable gets here...

Edit: Graph of power loss below vs. software version.View attachment 218100

In 7.1:
Ludi was 465
MB meant an immediate 20kW power increase
MBR added an additional 7-10kW
Launch Mode did nothing additional

 

[n2mb_racing]

In 7.1:
Ludi was 465
MB meant an immediate 20kW power increase
MBR added an additional 7-10kW
Launch Mode did nothing additional

Ok, yeah. 8.0 was the same for me until the latest update. Now it seems MB and MBR do nothing unless you use launch mode.

 

[stopcrazypp]

This is a good point and question. Is it that 1600A is just too much draw from these batteries? If such is the case, then wouldn't 1750A be too much on the 100kWh batteries as well?

I did some digging. I'm not claiming to know the math well enough but there are some interesting thoughts here. Tesla uses NCA batteries in their packs. These NCA batteries are well suited for capacity and cycle life. Something they are apparently not known for is high amp draw. Based on the following sites (SITE 1 - SITE 2) The C rating for these type of batteries hovers right around 1C. As far as batteries go, that's not very good. So, how much stress are we putting on out 90kWh packs at 1600A? If I did my math right it should be just north of 7C. That's 7x the rated amp draw. If you want to kill a Li-Ion battery, the fastest way to do so, short of physical damage, is to draw too much power all at once.

If this is true then it's not cooling or wire fuses. It's the Cells themselves giving up the ghost. This also means the problem has no real fix beyond staying away from 1600A.

thoughts?

Let's do some math first to see how much current each cell pulls.
85kWh/90kWh: 7104 cells, 74 cells parallel
P85D (Insane): 1300A/74 = 17.6A
85kWh/90kWh (Ludicrous/Limiter): 1500A/74 = 20.3A
90kWh v2/v3: 7104 cells, 74 cells parallel, 1600A/74 = 21.6A
100kWh: 8256 cells, 86 cells parallel, 1750A/86 = 20.3A

So actually we can take ~22A as the max current (100kWh is actually less).

The NCR18650B is your typical Panasonic NCA high capacity (low power) cell that hovers around 1C typical discharge, with 2C (6.8A) max continuous. Burst discharge (5-6 sec) is rated at 12A (3.53C) by third parties.
Panasonic 18650 battery NCR18650B 3400mah Li-ion rechargeable 3.7V

However people have pulled up to 30A even from these cells at a burst with no problems:
Panasonic NCR18650B Lithium-Ion [Archive] - FPVLAB - FPV Without The Interference

However, there are other NCA cells optimized for high drain, like the NCR18650PD or NCR18650PF. They are rated for 10A continuous (3.45C). 5-6 sec Burst at 18A (6.2C)
18650 NCR18650PF Panasonic IMR 2900 mAh Li-ion Battery button top | Orbtronic

There's a new cheaper version of the NCR18650B, NCR18650BE, which is wrapped as a Keeppower IMR18650 3200mAh. Someone did a 20A continuous (6.25C) test on it and it was able to handle it fine (although apparent capacity expectedly dropped, it still discharged ~6 minutes at that power with a 2.8V cut-off).
KeepPower 18650 3200mah (new Panasonic NCR18650BE cell) quick review
Test/Review of Keeppower IMR18650 3200mAh (Black) 2014

Keep in mind the tests above above are with ambient air. Tesla's liquid cooling will allow the cells to be pushed further. Overall, from seeing data from other Panasonic NCA cells which can even handle 20A continuous, I don't think a 22A burst discharge is that big a deal for this type of cell, esp. with liquid cooling.