Since our car is perfectly fine as per tesla dont see why we couldn't put it back to normal. After all was the original settings we paid substantially for.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Wiki Sudden Loss Of Range With 2019.16.x Software
- Thread starter Dutchmeeuw
- Start date
I suppose since road trips are the most frequent time you'd want this, one might want to improve Supercharger performance and range for road trips, but then leave it as "normal" the rest of the time (or if not in a hurry).
I guess the other question is - what kind of pack failure would be induced by this - I assume just more rapid capacity loss? (I'm just an outside observer, don't have a 85 kWh Model S, though I know people that do).
Lol that's pretty funny. Can i just ask why you put a volt battery in there? I've heard those batteries have pretty good range retentionI mean... It's definitely doable. I've reverse engineered virtually the entirety of the Tesla BMS... to the point where I was able to "supercharge" a Tesla that was running fine on a Chevy Volt battery in the trunk (with no Tesla battery installed at all). I limited it to 80kW peak... 5C charge rate lol. So making a modification that adjusts whatever parameters I want is perfectly doable.
^ Look at that range!(Full uncapped 2013 Chevy Volt pack)
(#3 module is in the underneath section)
Oh, and the car was too light... so added 360 lbs of concrete to the frunk to balance it out a bit.... still rode too high.
Suffice it to say, I can make these cars do what I want.
Maybe it was a range and charge rate upgrade on his 85.Lol that's pretty funny. Can i just ask why you put a volt battery in there? I've heard those batteries have pretty good range retention
Don't think there's much of a market for such a device.
Lol that's pretty funny. Can i just ask why you put a volt battery in there? I've heard those batteries have pretty good range retention
This was basically a test bed car for my customer BMS replacement. I designed a replacement BMS from the ground up that the car can't tell isn't the real deal (see the lack of alerts on the IC?
). The idea is that any battery pack or power source capable of providing the required power could then run the car. The Volt packs are great because they're relatively small (~350 lbs), 16 kWh (good enough to light real world use) and can discharge at over 20C peak (~300+kW peak) without problems. This test car was actually super quick because it was, even with hundreds of lbs of concrete in the frunk, hundreds of pounds lighter than the lightest Model S, but with as much peak power as a P85.This car had an empty battery shell in the bottom of the car, and the HVJB wires for the battery routed to the trunk for testing the BMS replacement prototype hardware/software.
The end goal was to make essentially a racing pack that would be a drop in replacement for the performance dual motor vehicles. Also, would open up the possibility of using other third party non-Tesla cells to build full packs as third party replacements, potentially less expensive (but lower range) for vehicles that needed them.
Sorry, probably a bit O/T. But still, the whole point of sharing was to just kind of prove that there's not much I don't know about these vehicles, especially on the battery side of things.
I was being half serious. On long trips I found the 40 to 80% / 85% charge just too short a duration to grab food and use the facilities if you have to shift the car as soon as it's done charging. Why not save stress on the battery an avoid indigestion at the same time? I guess it wouldn't be popular at busy chargers!
Oh, just do what everyone does to get around that limitation and set to 100%. Supposedly the 80% isn't a hard limit anymore (after I raised Cain about it on Twitter when the 80% thing screwed me over) and you can just bump it back to 100%, which will easily add 30-40 minutes.
Yeah, I was recently on a trip, and got the "High usage Supercharger" 80% thing in Tremonton, UT. Really? This is The West (TM), man! I get that there aren't a lot of Superchargers out here, so many of them are heavily used, but also...there aren't a lot of Superchargers out here! And the speed limits are 80 mph. I need to charge pretty full to make it to the next one in Twin Falls! (Burley just got opened up recently, which gives a shorter option now)
islandbayy
Active Member
COUNT ME IN! My MX is a PITA for distance driving. 74kW top charge rate, and even then for a short time. So forget towing a trailer, as the need for a high SOC and the slow charging kills it.I have maybe a dumb question.
I wonder how many affected folks would void their battery warranty entirely and install a third party device (probably for a few hundred bucks) that unlocks original range and supercharge speeds?
My Model S is still on firmware 8.1 and gets 94kW max for a decently sustained time, but not having remote access is getting old. Also my MCU is starting to fail and I'd like to upgrade to MCU2 and blow the last of my powerwall credits as tesla wouldn't deliver two powerwalls they owed (Actually took the service center calling constantly to the referral team for months to even get referral to respond).
I'd definitely install or do something to get supercharging speeds back once battery is out of warranty. MIGHT do it while still under warranty.
BTW my late 2016 MX 75D 100% SOC voltage is now 4.18 +/- 0.02 with one brick of cells about 60 mV lower than the rest of the pack after the battery rests after charging.
Interesting, since the usual issue is that a brick ends up higher than the rest due to a failed fuse or weak cell(s). A _lower_ brick would likely indicate a stuck FET or similar issue causing a drain on that group either continuously or in conjunction with balancing attempts on an adjacent group.
Could also indicate a BMB sense issue, since 60mV low should be pretty uncommon.
islandbayy
Active Member
AC Charging does really slow down at the end as it does it's damdest to bleed off that brick. I think the BMS also raised the 0% SOC Voltage now. Range-O-Meter gets really wonky at low SOC. High SOC too, example, park with 160 miles rated, come back 10 minutes later to 150 rated. At low soc, say, sub-10%, while driving, range-o-meter could jump between 10 and 20 miles either direction. The closer I get to 0% SOC and 0 miles, the "Longer" each rated mile could last. like it's stretching them out. Watching voltages, 0 remaining miles and pack voltage could be above 3v across the board. Then continue to drive (Kept a generator in the trunk in case of shut down, drove loops around my neighborhood) and after a few minutes at 0, 10 rated magically pop back up, then run that down to 0 again, voltages dropping, then 5 rated might pop back up. As if the BMS can't get a remotely accurate idea of how much remaining power is left (Could be due to that one brick though).
Last week I continued to shutdown in a Meijer parking lot with a supercharger then went to about 98%. My god, i've never seen the reported range jump around so much as I did on that drive home... It finally settled at 214, which is what my 75D has been getting lately. But with Tesla's fiddleing with the Canbus data output, whatever they are doing, TM-SPY and Scan My Tesla's calculated numbers are way off, I can only go by raw data such as actual voltages of cells :-(
Side question; as I've done EV Conversions in the past (Was WI's first EV Dealer, back in the old Lead Acid days). If I were to transplant a Tesla mechanicals into another vehicle, such as, say, an RV (Small Motor Home, such as a Ford Econoline 150 based vehicle), And everything was properly connected. Would it still be able to supercharge?
For those of us who need more afirmation:
So same current through compromised module(s) as other modules during charging, but less energy absorbed and stored by them. End of charge voltage higher due to possible different causes. Cell fuse is obvious because reduction in number of cells storing energy so faster voltage increase. High ir cells have less current through them and dissipate more energy as heat. Lower ir cells..... (? Have more current flow, do they implicitly absorb more charge? What about internal partial shorting effect? More current, self discharge, less charge absorbed into electro chemical structure?)
So do these FET's tend to fail short circuit or drift towards lower resistance? Does the balancing circuit maintain some control over FET conductivity or is it a hard fail? If hard, presumably that completely messes up balancing.
Presumably if a FET goes short, there is a slight mitigating effect during discharging but does it the exact opposite of the effect during charging.
Would that likely have similar over all effect as failed / failing FET by turning the FET on at the wrong times?
Are there enough of this catagory of fault to know typical / consistent symptoms?
Is it correct for 85 packs with current software that they use low (and high?) SOC where brick voltage of weakest bricks dictate limit of charge / discharge to determine what energy the whole pack is capable of storing?
Presumably the 'stretching out' effect is a consequence of erroneous data concerning total pack energy and any of the above BMS conditions could mess up the data.
I thought that dropping of brick voltage was inevitable once allowed to settle with hv and balancing off for some time, as the charge will equalize between cells in each brick. The amount of voltage settling between immediate end of charge and some elapsed time would be indicative of the imbalance of charge absorbtion between cells in each brick.
So same current through compromised module(s) as other modules during charging, but less energy absorbed and stored by them. End of charge voltage higher due to possible different causes. Cell fuse is obvious because reduction in number of cells storing energy so faster voltage increase. High ir cells have less current through them and dissipate more energy as heat. Lower ir cells..... (? Have more current flow, do they implicitly absorb more charge? What about internal partial shorting effect? More current, self discharge, less charge absorbed into electro chemical structure?)
So do these FET's tend to fail short circuit or drift towards lower resistance? Does the balancing circuit maintain some control over FET conductivity or is it a hard fail? If hard, presumably that completely messes up balancing.
Presumably if a FET goes short, there is a slight mitigating effect during discharging but does it the exact opposite of the effect during charging.
Would that likely have similar over all effect as failed / failing FET by turning the FET on at the wrong times?
Are there enough of this catagory of fault to know typical / consistent symptoms?
Is it correct for 85 packs with current software that they use low (and high?) SOC where brick voltage of weakest bricks dictate limit of charge / discharge to determine what energy the whole pack is capable of storing?
Presumably the 'stretching out' effect is a consequence of erroneous data concerning total pack energy and any of the above BMS conditions could mess up the data.
I thought that dropping of brick voltage was inevitable once allowed to settle with hv and balancing off for some time, as the charge will equalize between cells in each brick. The amount of voltage settling between immediate end of charge and some elapsed time would be indicative of the imbalance of charge absorbtion between cells in each brick.
Last edited:
A FET can fail all kinds of ways. Partial short and full short being the most common, I believe. Failing full open is possible, but that's usually not the FET itself. If failed partially or fully closed, it would continuously drag a cell group lower. Such an imbalance would be almost impossible to correct.
A BMB sense issue in the sense noted would be if the BMB was not reading that brick's voltage correctly, which is possible.
Correct, the weakest cell group dictates voltage cut points. The BMS doesn't allow any cell group to be pushed outside of established limits.
Generally after charging the weakest cell groups will have the highest voltage at rest. After discharging, the weakest groups would have the lowest voltage. In @islandbayy's case where one cell group is showing _lower_ than the rest after charging it suggests another issue, and I'd possibly expect it to be somewhat consistently low throughout the SoC range (while at rest)... alternatively, it would indicate that every other group is weaker than that group..... which seems less likely than an issue with some hardware on the one group.
A BMB sense issue in the sense noted would be if the BMB was not reading that brick's voltage correctly, which is possible.
Correct, the weakest cell group dictates voltage cut points. The BMS doesn't allow any cell group to be pushed outside of established limits.
Generally after charging the weakest cell groups will have the highest voltage at rest. After discharging, the weakest groups would have the lowest voltage. In @islandbayy's case where one cell group is showing _lower_ than the rest after charging it suggests another issue, and I'd possibly expect it to be somewhat consistently low throughout the SoC range (while at rest)... alternatively, it would indicate that every other group is weaker than that group..... which seems less likely than an issue with some hardware on the one group.
Last edited:
islandbayy
Active Member
Yes it is correct they use the high/low voltage to determine. What I was getting at, was the car reports "0" while the voltages are still much higher than "0".
Yes, when the weakest brick hits shutdown, thats it. Same for charging. Though for charging, it can reduce charge current and keep bleeding that full brick as much as possible. However with shutdown on one low brick, I have noticed the BMS is allowing some leeway during hard accel or bursts of high power demand. I've seen my MX dip as low as 2.3-2.4v on a hard accel as long as it wasn't sustained and then voltages go back up. All though guess-o-meter drops quite a bit more quickly.
I agree. The BMS is trying to compensate and guess as much as it can taking into consideration that one low brick. And no amount of full discharges and recharging to full or near full is able to bring it back in tune.
I'll bite! Would love my old supercharging speeds back again so long as Tesla cannot detect.I have maybe a dumb question.
I wonder how many affected folks would void their battery warranty entirely and install a third party device (probably for a few hundred bucks) that unlocks original range and supercharge speeds?
Me too. 2014 S85 and my peak rate is about 75kw and it tapers down quick. I usually pull in with less than 10% left, it starts up at about 90kw, and tapers instantly down to 74. I’ll come back an hour later after eating slowly, walking slowly, doing everything slowly to kill time and it’s barely at 70%.
If anyone wants more time to eat while charging, just buy my car from me. For a couple hundred dollars I’d buy a fix instantly, it’s a whole lot cheaper than a new car.
Similar threads
