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

Tesla Roadster Battery Care

This site may earn commission on affiliate links.
I decided to try to set my Roadster so that it starts charging late at night rather than as soon as I get home, with the thought that that way it won't be sitting around at a full standard-mode charge all night. The idea wasn't to try to get it to a partial charge, just to have it start later.

I've never used timed charging mode before, so maybe I did something wrong. I got home, parked the car and set it to start charging at 4AM. I plugged it in and it started charging immediately. I got back into the car and again set it to charge at 4AM. It stopped charging and put up a note saying "scheduled charging at 4AM" or something like that.

This morning, I got to the car and it hadn't charged. It had a screen that said "plug in cable" or something, even though it was plugged in. I switched it to "charge on plug-in" and it started charging, but it only added 2 ideal miles in the time it took me to take the garbage to the curb.

Luckily, I have enough charge to get me through the day, so it's no big deal. Still, it's annoying.

Has anyone else had problems like this? Any suggestions?

I had this same issue.
Set the time to charge in the car.

Then after that has been set, open the charging port and plug in the car.

If the car is already plugged in when you are messing around with the time, it is not reliable for actually starting at that time.

This is a bug in the firmware. If the car is already plugged in when you are messing with that timing setting, it doesn't always notice that you have set it to start charging at a certain time.

Opening the charging port door (and plugging in) after this has been set seems to be the key to the car noticing your settings.
 
This reminds me a lot of how I charge my NiCd-car. The Norwegian electric car enthusiasts have done a lot of thinking and research into how we can improve the lifetime of our batteries.

Flooded NiCd batteries are balanced by overcharging. As soon as each cell has reached full SOC it will start to burn off excess charge by electrolysing the water in the electrolyte, which slowly disappears as hydrogen and oxygen. Those cells that reach full SOC early will consume more electrolyte. So the cells will be balanced electrically, but will in the process get imbalanced electrolyte levels - and they must absolutely not run dry.

The cells also self-discharge at slightly different rates, and self-discharge is greatest at high SOC, so leaving the battery at a full SOC causes more imbalance.

We want to keep the battery balanced while also maximizing the time between electrolyte top-ups. Minimizing the number of balancing charges and time spent at high SOC is good. This seems, through an entirely different mechanism, to lead to exactly the same treatment the Roadster wants.

I avoid charging if I can. If I have almost enough charge, I will just charge the needed amount using a timer. When it's down to about 20% SOC, I will let it charge completely to balance the pack (driving at a very low SOC is bad for NiCds). I also use a timer to start the charging as late as possible. An added bonus is that the batteries perform a little better right after charging.

This will easily cut the time spent at high SOC by 90% or more. I gather the Roadster has a built-in timer, which makes it easier.

But this really ought to be implemented in firmware. The necessary input parameters are the range target for the charging session and when the car must be ready. Use of profiles linked to geographical position would make the setting of the parameters very convenient - in most cases all that would be required is a single OK to confirm the profile that was used the last time.
 
Fantastic thread! Thanks James for the post you started it with.

These kinds of charging strategies and SOC considerations are going to be common knowledge 10 years from now, just as most people currently know about things like octane numbers, engine knock, choke, blue exhaust, etc. For now, it's just us on the bleeding edge who get to watch this knowledge base accrue :)
 
Ok. Here's an idea:

The car could offer the following recharge profiles:

Range/Performance
Standard
Custom 1
Custom 2
Custom 3
Custom 4
Storage

All profiles except Range/Performance and Storage carry out cell balancing after the recharge has finished.

As mentioned earlier, Storage isn't really a recharge mode, you just set it and the car gets on with the job of keeping the battery happy somehow. Cell balancing is redundant(?) for Range/Performance - I'm guessing you can't shift charge between "full" cells.

For a given mode the system displays recharging start and end times; owners choose which one they prefer to set.

The Range/Performance and Standard settings recharge the battery to a non-adjustable charge level. Owners can only set the recharge current and (by default) the "finish charging/cell balancing by XX:XX am/pm". This way, if the car happens to be at a different location, owners don't have to work out a new start time. Also if the time needed for charging/ cell balancing changes as the battery ages(?) this will not be noticed.

Hopefully they can sort out the problem of the car not recognising settings if the charge port is open etc.

Owners can decide if they need the recharge (or recharge/balance) period to end immediately prior to departure (recommended for Range/Performance) or if the end should come just before higher daytime electricity prices start or to coincide with some on-site renewable energy source etc.

The system would need to alert owners if the chosen recharge current is too low to achieve the necessary charge and cell balance within the set time. Owners can then make necessary adjustments.

Since the car remembers recharge settings for a given location it could also have the ability to record the mileage between key locations including those which don't offer recharging. The car could learn your typical mileage and preferred driving style. It knows where you live, it could know where you work and how you get there :)

The owner may choose to allocate settings for particular regular journeys or just daily or weekend mileage to one of the four Custom profiles.

As before, these allow the owner to set the start or finish time and recharge current. In addition, the charge level can also be set anywhere from 80% to 20% in 5% increments by the owner and (if the profile's auto-optimise setting is on) this could be gradually adjusted by the car using its knowledge of the route, your driving and maybe things like seasons of the year (which would affect average battery temperatures).

Obviously all four of these profiles could be allocated with very similar or very different settings between one or maybe more regular drivers.

It would be useful if the default names could be changed (USB keyboard?) from Custom 1, Custom 2 etc to things like Home2Work, Recharge@Work, FridayNight etc for easy selection.
 
This is a bug in the firmware. If the car is already plugged in when you are messing with that timing setting, it doesn't always notice that you have set it to start charging at a certain time.

Opening the charging port door (and plugging in) after this has been set seems to be the key to the car noticing your settings.

I tried this and it worked fine last night. I think from now on I'll leave it that way, there's no need to let it sit at full standard mode charge overnight when I'm sleeping.
 
The greatest benefit of an updated firmware with more advanced charging options would be that the car would be able to balance its battery once the range target is reached without bringing it to a high state of charge. Convenience is nice, but as I understand it, balancing cannot currently be done except by completing a full standard mode charge.

I think users should not have to worry about SOC. By telling the software what range you need and when, the car would be able to select the optimal SOC target within those parameters.

Assume for instance that battery deterioration at 20 deg C is lowest at 40% SOC, while at 30 deg C, 30% SOC is better. If the user specifies a needed range of 35 miles, the car is free to select the optimal SOC target. Input the needed range, let Tesla choose SOC target.

The idea of several customizable profiles is good, but there should also be a choice for setting one-off charging parameters, so you don't have to modify one of your profiles to do that.

By linking to geographical position, I was thinking that the car might present a choice between those charging profiles that have been used at that location previously. It probably also knows the time and date, so it might even suggest the correct profile for weekends when plugged in at home on a friday evening.
 
Last edited:
I sent some questions to Tesla Motors asking about strategies for extending the life of my battery pack. I was looking for more tips beyond just plugging in and recharging to standard mode each day.

I received a very detailed response from Dan Myggen and he gave me permission to post it here.
I was in a showroom last week and the current ESTIMATE for a battery replacement is somewhere between $30,000-$40,000. Is this correct? Also, if the battery is reduced in capacity by 60-65%, at least it is still functioning which is not what I was led to believe is the case after 100,000 miles.
 
I was in a showroom last week and the current ESTIMATE for a battery replacement is somewhere between $30,000-$40,000.

You could lock your battery replacement cost at $12,000 by purchasing the option from Tesla. In reality, the cost will likely drop within 5 years and the real cost will be lower. Or the battery you get in 5 years may actually provide 400-500 miles of range. The technology and the pricing is improving every year.

Also, if the battery is reduced in capacity by 60-65%, at least it is still functioning which is not what I was led to believe is the case after 100,000 miles.

Most people would be fine with 120 ideal miles of range in standard mode. I know that I am fine with that most days. Within 5 years we will have more recharging options out in the wild.
 
You could lock your battery replacement cost at $12,000 by purchasing the option from Tesla. In reality, the cost will likely drop within 5 years and the real cost will be lower. Or the battery you get in 5 years may actually provide 400-500 miles of range. The technology and the pricing is improving every year.



Most people would be fine with 120 ideal miles of range in standard mode. I know that I am fine with that most days. Within 5 years we will have more recharging options out in the wild.
If I understood the battery warranty correctly, it does not guaranty a replacement, only a warranty against being defective from year three on to 100,000 miles or 7 years. And I was mistaken in that the battery life is reduced to 60-65% not by as pointed out by tennis trs.
 
If I understood the battery warranty correctly, it does not guaranty a replacement, only a warranty against being defective from year three on to 100,000 miles or 7 years. And I was mistaken in that the battery life is reduced to 60-65% not by as pointed out by tennis trs.

I think you are confusing a $5,000 ESA (Extended Service Agreement) for 3 additional years (years 4-6 if your original Warranty is 3 yrs, or years 5-7 if original warranty is 4 yrs) vis-a-vis the Battery Replacement Agreement ("BRA" :tongue:).

The BRA actually does guarantee a replacement battery between years 3 (or 4) and 10. If after 7 yrs it's the pre-paid $12k; if earlier it's an additional $2k/yr; if later it earns a $1k refund/yr.

One sentence in the BRA that worries me a little is
If applicable law requires Tesla to return a replaced battery to you, you agree to pay Tesla the additional cost of the replaced battery.
What locality requires this ? And how much is that additional expense going to run ?

Also, what happens if your Roadster is totaled and you choose not to repair/replace it ? Refund ? (If you choose to replace it (like Ian) is the BRA transferable ? If you choose to replace with a Model S .... )
 
My reading of the "additional cost" for a "replaced battery" means that when they replace your battery, they get to keep the old one. If you insist on keeping the old one and the law is on your side, then you pay them for the old battery which you are keeping. IOW, the $12,000 BRA includes the provision that the old battery belongs to them.

Just my reading. I am not a lawyer.

The long quote in the OP says that "most" of the benefit of storing the pack with less than a full charge is realized at 87%. I read this to mean that there is little benefit from storing it at a lower SoC.

As far as degradation time, I think it will take me 20 years to put 100,000 miles on my Roadster. That's 5,000 miles a year, which is typical for me. I expect this car to have more than enough range for my needs until I am too old and arthritic to be able to get in and out of it. At that time I'll trade it in for a Model S, or whatever.
 
At grid level the total number of available used EV batteries will probably be insignificant for a very long time. But in a home-based PV system that is not grid tied, where weight is not an issue, and size is much less of an issue than in a car, a used EV battery with 50% of its original capacity will definitely have value. And even for recycling, dead batteries may have some value.
 
Even without solar a used pack could make a nice house sized UPS for emergencies, and maybe even take advantage of dual rate metering for time shifting. Charge the pack at a cheaper night rate and then use that energy during the higher day rate, or allow the smart grid to tap it for load leveling for a fee, at some point. Lots of possibilities. The LiCo cells that Tesla uses have the highest recycling value at this point of any of the Li chemistries.
 
The LiCo cells that Tesla uses have the highest recycling value at this point of any of the Li chemistries.
Interesting, I wonder if this was perhaps due to the high cost of cobalt. Be it as it may, I wanted to direct your attention to Tom Moloughney's blog. Tom is one of the MINI-E pioneers and he posted a very thoughtful comment on the plugincars.com article about battery care a fellow Leaf owner wrote couple of months ago.

Tom leased his MINI-E as part of a field trial by BMW and since there was no mileage limit, he subsequently proceeded to drive the car into the ground. His battery pack is based on 5,088 laptop-format cells manufactured by Molicel in 2008. Tom reportedly charged his MINI on average twice a day through its 12 kW onboard charger. And he charged it to full whenever possible. Based on the ratio of usable to rated capacity, a full charge likely corresponded to about 88% of true pack capacity, which is comparable to standard mode in the Roadster.

Given what we know about lithium-ion batteries, Tom should see about 6-10% reduction in battery capacity after 27 months and 50K miles. Well, the surprising thing is that there is hardly any degradation at all. Tom is not able to measure battery capacity directly, but he keeps track of the ideal range predicted by the car. Based on this data, he claims to be experiencing about 2% range reduction, which is surprising to say the least.

The reason I'm bringing this up is simple. I like the err on the side of caution and the battery care guide from Dan Myggen posted earlier in this thread helped me to understand how to treat the battery conservatively. Admittedly, Tom's experience with the MINI is a bit extreme, but it gives me the confidence that the lithium-ion pack in my car has the capacity to perform well beyond my expectations.

How is this relevant to this thread? If Tesla recommends to routinely charge the car in standard mode, they will base their battery life estimates on that. They will likely be conservative, which means that there is a good chance that we will see better results in the real-world. And if you treat the pack conservatively, you could improve upon that further. The question is then, do you really need to do that or would you be satisfied with the longevity you are getting by following the standard charging protocol.
 
Last edited:
Interesting, I wonder if this was perhaps due to the high cost of cobalt.
Basically. The lithium isn't worth recovering at this point as far as I know, nor is the aluminum, which are sold as slag, but the copper and cobalt are. (To be clear the lithium and aluminum are not disposed of, they are just not extracted from the slag as distinct elements. The slag does have industrial uses.)
How is this relevant to this thread? If Tesla recommends to routinely charge the car in standard mode, they will base their battery life estimates on that. They will likely be conservative, which means that there is a good chance that we will see better results in the real-world. And if you treat the pack conservatively, you could improve upon that further. The question is then, do you really need to do that or would you be satisfied with the longevity you are getting by following the standard charging protocol.
Agree. If you just follow the standard recommendations I think you'll do better than expected anyway in most cases.
 
Last edited: