Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Balancing the battery

This site may earn commission on affiliate links.

David99

Well-Known Member
Supporting Member
Jan 31, 2014
5,506
9,330
Nomad (mostly US)
I keep reading a lot about balancing the battery and many people here seem to know exactly where it is happening and how. Now I'm pretty much aware that when many cells are in a battery pack they they need to be as similar as possible or else the weakest cell will effectively determine the entire pack capacity. What I don't understand is how many people here are so confident in knowing what and when the Model S is balancing the batteries. I can't find any reference in the manual about it and I'm not aware of any official statement from Tesla about it. In several discussions here I see people commenting on range loss caused by imbalance followed by a more or less detailed recommendation on how to balance it again. I can't help but wonder where that confidence is coming from. Is there any official info from Tesla about it, or are people just making things up?

All I could find is the battery tear down where the person found resistors that could possibly be used to "bleed off" charge from individual segments. These resistors are pretty small and can dissipate only a very small amount of power, though. Not enough to explain extended charge sessions at 2 or more kW that people comment on saying it would be balancing.

There is no doubt that imbalance with the pack is an issue, but wouldn't it be much better to prevent it from getting out of balance instead of constantly trying to fix it? Tesla recommends to keep the car plugged in all the time, even when not charging. I do that and I noticed the Model S will often turn on the coolant pump and run it for a short time. It is using grid power but very little so it's not heating nor cooling the battery (there is also no fan running). What it does IMHO is ensuring the temperature among all cells is the same at all times. I think that is key to keep the battery in balance. If two cells are identical, there is no reason they will drift apart in capacity if used the exact same way. But we all know too well that temperature has a significant impact on cell efficiency. If two identical cells are charged or discharged the same way but have different temperatures, they will eventually drift apart in performance/capacity. By keeping them at the same temperature at any given time, even when just resting, they will stay the same.
 
Most of the 'confidence' from my perspective appears to be anecdotal experience combined with tid bits from the inter-webs and perhaps the off-hand comment from a Tesla tech. I've seen my car 'charging' at ~100% SOC for >20 minutes and instead of 'time remaining' it says 'calculating' while still drawing >1 amp from the grid... I can only assume that it's somehow topping off the battery... The other odd thing I've experienced after an extended 'Max Range' session is on at least two occasions I've drive >10 miles and consumed >3kWh without my rated range dropping by even 1 mile...

I do think there are sufficient differences in the cells just due to manufacturing tolerances that balancing is a major issue even if all cells are cycled evenly; Solar cells are a good example if this. A battery is significantly more complicated than a solar cell and there are differences between solar cells of ~5%. For any given solar module there are usually ~6 different power levels. This is a consequence of lumping matching cells together. Lower solar cells form a 240w module while higher cells form a 265w module. They were all manufactured the same way... those are just the tolerances. Solar cells are just a little easier to measure. You generally need to cycle a battery to determine its actual capacity.
 
Tesla has been completely silent on the balancing topic, David--so you're correct in that it's conjecture.

People see low power draw at the end of a range charge and assume it's balancing...but that's not the only possibility. Since we know the charging power must taper as the battery "fills", this could simply be Tesla slowly giving sips of energy to the pack as it gets topped off.
 
I think some of the speculation comes from knowledge of the Roadster, where we have access to the service diagnostics screens and can watch the balancing process in action. The assumption is that Tesla carried over the same approach to the Model S, which is not a bad guess since the pack design is very similar. Agreed the details for the S are completely unknown.
 
I do think there are sufficient differences in the cells just due to manufacturing tolerances that balancing is a major issue even if all cells are cycled evenly ...

I actually doubt that. Not that there are no differences in manufacturing, there certainly are, but cell matching for battery packs has been done for a long time for example in battery packs for RC cars/helicopters. Each cell is tested and then packs are combined based on best match. This way the differences between cells are mostly eliminated. I assume Tesla does match cells they put together in a battery pack.

I also noticed the sometimes odd behavior of the remaining rated range when driving. I had situations where it would not change during normal driving, and other times where it would just drop a few miles for no apparent reason. Again, I think it's speculation this would be an effect of balancing. For all we know (based on an email from Tesla) the rated range is based on a mathematical model that is adjusted based on realtime info from the battery (voltage at a certain temperature and load) so the unexpected changes in remaining range could be be just that, putting the calculated range in sync with the actual data from the battery.
 
There has been no alternative hypothesis suggested for the up-to-an-hour time where "1 minute remaining" and "charging" messages appear. If it's not balancing, and the temperature is moderate (no battery heating or cooling), then what other reason could there be? Besides the Roadster experience, the Model S battery pack has been disassembled and circuitry found that appears to be connected with balancing. So I believe we have a bit more than anecdotal evidence.
 
> charging power must taper as the battery "fills" [Todd Burch]

Sure, but the balancing act is stopping for indefinite period, then charging again for another indefinite period, etc for an hour or more. Definitely a thought process going on here.

Likewise, following an 'emptying of the battery' experience (as in 'dead in the road'), I routinely set the charge limit to ~65% but discovered the next day that the battery was near full charge. Another instance where the car thought it needed to balance battery so over-rode my instructions and did what it had to do.
--
 
Most of the 'confidence' from my perspective appears to be anecdotal experience combined with tid bits from the inter-webs and perhaps the off-hand comment from a Tesla tech. I've seen my car 'charging' at ~100% SOC for >20 minutes and instead of 'time remaining' it says 'calculating' while still drawing >1 amp from the grid...

This is not battery balancing, this is the very end of the taper curve.

- - - Updated - - -

There has been no alternative hypothesis suggested for the up-to-an-hour time where "1 minute remaining" and "charging" messages appear. If it's not balancing, and the temperature is moderate (no battery heating or cooling), then what other reason could there be?

It's been said many times. The slow charging at the very end is nothing more than the end of the taper curve. It has nothing to do with balancing. As the battery fills up, the charge current is throttled so as not to damage the batteries. This continues with as little as 1 Amp draw while the cells are being filled to the max.

- - - Updated - - -

Likewise, following an 'emptying of the battery' experience (as in 'dead in the road'), I routinely set the charge limit to ~65% but discovered the next day that the battery was near full charge. Another instance where the car thought it needed to balance battery so over-rode my instructions and did what it had to do.
--

That's impossible. The car will not override your charging preferences unless you reset them somehow. There is nothing in the car's behavior that would force a full charge after an "empty battery experience". It will charge to whatever you set. If yours did not, there is another explanation.
 
I think some of the speculation comes from knowledge of the Roadster, where we have access to the service diagnostics screens and can watch the balancing process in action.

Yet another diagnostic tool that Tesla gave Roadster owners but overlooked when designing the Model S. I would be enormously interested in seeing this data on my MS - and it would settle this question of balancing for good. C'mon Tesla - make thee a priority for 7.0.
 
Yet another diagnostic tool that Tesla gave Roadster owners but overlooked when designing the Model S. I would be enormously interested in seeing this data on my MS - and it would settle this question of balancing for good. C'mon Tesla - make thee a priority for 7.0.

Well, strictly speaking Tesla didn't give Roadster owners that internal password-protected diagnostic screen, and the Model S has a similar system. Both are restricted to Tesla service engineers only. It is just that for the roadster, the passcode seems to be the worst kept secret, while for the Model S they tightened the security on the access to that system.

For Model S, see:

Please Enter Access Code
Dev Console | Forums | Tesla Motors
 
This is not battery balancing, this is the very end of the taper curve.
...
The slow charging at the very end is nothing more than the end of the taper curve. It has nothing to do with balancing. As the battery fills up, the charge current is throttled so as not to damage the batteries. This continues with as little as 1 Amp draw while the cells are being filled to the max.

Citation needed.
 
From: Tesla Battery Range Degradation - TESLARATI.com :

An email excerpt, allegedly from a Tesla Motors representative:

“As far as your question about ‘pack balancing’, it is not really necessary. Fully charging your battery to the maximum several times may increase the estimated range that is displayed on your dash, but not necessarily because your battery packs need balancing. The algorithms used for estimating range are based on determining how much energy is stored in your battery. However, the only accurate way to measure the amount of energy a battery can store is to fully charge the battery, then completely discharge the battery to zero, then measure the amount of energy that was released. While this would give an accurate measurement, it would not be good for the battery. So the Model S uses algorithms to estimate the amount of energy a battery can store.
When a battery is fully charged, the algorithm is able ‘learn’ the ‘upper limit’ for the battery. But since you don’t fully charge your battery often (which is a good thing), the algorithm for your vehicle may no longer have an accurate value for the ‘upper limit’, causing your Estimated Range calculation to be slightly off. By fully charging your battery several times, your algorithm may relearn the ‘upper limit’ of your battery, so it may start to show a different/higher number for your Estimated Range. It is important to understand, however, that fully charging your battery several times, is not going to actually help your battery. In fact, frequent full charging of your battery can actually accelerate battery degradation.”


 
  • Informative
Reactions: Nujec

“As far as your question about ‘pack balancing’, it is not really necessary. Fully charging your battery to the maximum several times may increase the estimated range that is displayed on your dash, but not necessarily because your battery packs need balancing. The algorithms used for estimating range are based on determining how much energy is stored in your battery. However, the only accurate way to measure the amount of energy a battery can store is to fully charge the battery, then completely discharge the battery to zero, then measure the amount of energy that was released. While this would give an accurate measurement, it would not be good for the battery. So the Model S uses algorithms to estimate the amount of energy a battery can store.
When a battery is fully charged, the algorithm is able ‘learn’ the ‘upper limit’ for the battery. But since you don’t fully charge your battery often (which is a good thing), the algorithm for your vehicle may no longer have an accurate value for the ‘upper limit’, causing your Estimated Range calculation to be slightly off. By fully charging your battery several times, your algorithm may relearn the ‘upper limit’ of your battery, so it may start to show a different/higher number for your Estimated Range. It is important to understand, however, that fully charging your battery several times, is not going to actually help your battery. In fact, frequent full charging of your battery can actually accelerate battery degradation"

+1

Thank you!
 
It is important in regards to how to treat your battery for the longest longevity and correct operation is to totally ignore all advice given by anyone unless it is from Tesla corporate directly and in writing.
Follow the charging instructions in your manual, and ignore everything else. This includes advice from people on this forum, local Tesla service center people, internet blog postings, your father in law Jack who has been a VW mechanic for the last 30 years, your neighbor who "raced a few golf carts when he worked as a caddy in the 70s", all of it.
 
It is important in regards to how to treat your battery for the longest longevity and correct operation is to totally ignore all advice given by anyone unless it is from Tesla corporate directly and in writing.

I hear you and I get what you are saying. But for those who are a little more interested in technology it's good to dig a little deeper. There are many well known and researched facts about Lithium batteries that suggest how to best charge and operate a battery to keep it healthy for many years. The higher the state of charge the faster they age, the higher the temperature, the faster they age. So keeping it not fully charged and cool is good for the battery. BUT, and here is the irony, both higher temperatures and higher state of charge give you better performance, efficiency and range. So how you treat your battery is always a balance between these opposing factors.

And here it gets even trickier: when only charging the battery partially the capacity prediction can get less accurate fooling a lot of people into believe they have lost or gained range using a specific charging method, when in reality the 'rated range' number only changed based on the mathematical model. Jumping on the conclusion the lost or gained miles are a result of imbalance and balancing is about as likely as just the algorithm spitting out different predictions based on the limited data it has. The closer you are going to a full charge/discharge cycle, the more accurate the algorithm as it gets data from the actual real 0%/100% point of the battery. And then people jump to the conclusion that charging to 100% makes balancing kick in. What really happened was the full cycle recalibrated the algorithm.
 
It is important in regards to how to treat your battery for the longest longevity and correct operation is to totally ignore all advice given by anyone unless it is from Tesla corporate directly and in writing.

Agreed.

And my 2 cents on the idea "full charge" somehow triggers balancing (or more balancing): From the packs that have been dis-assembled in salvage cars we have some idea what the battery management system looks like. Looking at the circuit boards and chips, one can rule in or out certain things. For example, it is easy to estimate the maximum balance current that would be supported by the wiring and circuit board traces. We can also see at what "granularity" the wiring connects (short answer: Each parallel grouping). And so forth and so on.

With all of those knowns, plus some background in designing Li based power systems, I'd lean about 90% toward continuous balancing while charging. Nothing special about the "last hour of full charge" as far as just balancing is concerned.

Just my (informed, calculated) opinion. As stated multiple times, operate you car with info from Tesla, not from anybody here. Including me.
 
It is important in regards to how to treat your battery for the longest longevity and correct operation is to totally ignore all advice given by anyone unless it is from Tesla corporate directly and in writing.

Hmmm.... well.... there are some undisputed facts that Tesla doesn't put out officially. Tesla DOES say that 100% SOC shortens the battery life but they aren't specific on levels 50%-90%... numerous sources cite ~50% as the ideal SOC for lithium ion batteries. There is a lot of testing information available from numerous sources I would trust like NASA, NREL and various universities. Some things are addressed by Tesla like don't charge a cold battery and don't keep your battery at 100% SOC. And some things aren't like short cycles are better than deep cycles and the closer your battery stays to ~50% SOC and the cooler it is generally the happier it is.

I wouldn't go so far as to accuse Tesla of withholding information but I do think they have a business interest in making the car appear as simple as possible. It doesn't get much simpler than charge to 90% unless you're going on a road trip.

It's difficult to say what effect leaving the car at 50-60% SOC vs 90% will have on the longevity of the battery... some tests indicate that in the long term it could be significant; but the battery is warrantied for 8 years so probably not a big concern for most people.

More than you probably ever want or need to know about Li Ion Batteries...
 
Last edited:
Well, strictly speaking Tesla didn't give Roadster owners that internal password-protected diagnostic screen, and the Model S has a similar system. Both are restricted to Tesla service engineers only. It is just that for the roadster, the passcode seems to be the worst kept secret, while for the Model S they tightened the security on the access to that system.

Ah, yes, fair point. IIRC, though, Roadster owners by default have access to a lot of data including TPMS, battery CAC, and specific error codes that do not require access to diagnostic screens. I guess I just don't see the reason for hiding much of this info.
 
Ah, yes, fair point. IIRC, though, Roadster owners by default have access to a lot of data including TPMS, battery CAC, and specific error codes that do not require access to diagnostic screens. I guess I just don't see the reason for hiding much of this info.

Again, not really. Battery CAC is hidden, and error codes are similar to model S. TPMS is better on Roadster, though.

The end-user information and customisation options on Model S are significantly better than on Roadster, with the big exception being TPMS.
 
It's difficult to say what effect leaving the car at 50-60% SOC vs 90% will have on the longevity of the battery... some tests indicate that in the long term it could be significant; but the battery is warrantied for 8 years so probably not a big concern for most people.

I don't think the battery warranty will cover degradation, as that is probably considered normal wear and tear.