Tesla Model S module Conversion.

[Mart Hale]

Hi all,

I have been running a Tesla Model S module with solar with my Outback controllers for over 5 months. I love lithium!!!!!!

Now I have seen the need to get a BMS for this setup, and adding more logic as I have found that the batteries will DIE if they are charged at 32 degrees!

I have found that someone has hacked the BMS board for this unit and I hope to be able to either use this hack, or, use my my BMS for this setup.


1) Is there an off the shelf BMS that I can use with the solar for this unit?

2) Has anyone here hacked one of these and can point me in the right direction of how not to blow myself up? I do have a couple ardruino's and I am looking at the open sourced material here -> hackaday.io/project/10098-model-s-bms-hacking


Specs of the battery ->

"They are Panasonic 18650 cells, but only Tesla has the data sheet for these. The individual cells have been verified by multiple individuals as being 3400mah. Maximum voltage is 4.2v per cell. Pack in configured 74p6s. That's 6 groups of 74 cells. Max voltage is 25.2 at 100% SOC, but it's best to only charge to 24.0v unless you monitor the individual groups, as no group should go above 4.2v."


Tesla Model S Battery Module, 24V, 250Ah, 444 Panasonic 18650 3400mAh OEM


Cheers!

 

[spinningmagnets]

Hello, Mart Hale. I am new to DIY powerwalls, but...I have been into electric bikes for many years. Whatever is not included in the published data sheets about the Tesla/Panasonic cells...I hold them in a very high regard. I am certain that once you pay the extra amount needed to acquire such a valuable resource, you will want it to last as long as possible. For that, I would plan on charging to 4.0V per cell, rather than 4.1V per cell, or especially 4.2V...

25.2V at max SOC is definitely 6S X 4.2V, so...I would take the recommendation of 24.0V seriously (24/6 = 4.0V), and this matches everything I have found out about lithium 18650 cells.

As to BMS's, it seems as if the Batrium BMS is well regarded, but it is quite expensive. In the ebike world, I have heard that the Bestech BMS's are adequate, but I have not used one yet, so I have no hands-on experience. Although, the new trend seems to be the Bluetooth connectivity, which allows you to wirelessly data-log your packs daily health with a simple laptop. Best of luck, and thanks for posting this...

 

[Mart Hale]

Thanks for your reply.


I have been looking into this hack of the battery ->

https://hackaday.io/project/10098-model-s-bms-hacking

But the requirements of my BMS would be able to deal with the flux charging of solar.

Yes you are wise in saying 4.1 as the top of the batteries should go, that will give me many years more of service. I have been running without a BMS for about 5 months, but now I want to reform

My line of thinking has been to research the Tesla S very very well, then to buy a battery pack from a wrecked car. Then make the conversion.

I am very happy that I have found someone who has figured out how to charge the unit without taking it apart.

Cheers.

 

[The-J-Man]

I was under the impression that the BMS is always active, monitoring the pack.
Its only that you cant connect multiple up and have them talking to each other. ??
Seems like people are just connecting pos and neg and running them...

If you discharge it, will it discharge to 0V?? or does it cut at like 16V or in that area?

you can Also charge the batteries at up to 45C without problem.
Great thing about the packs is they have the water cooling loop in it. Pretty sure you can modify a PC waterooling system to run with this pack if the external temp is too high.

 

[Chiptosser]

You might want to check out Jack Rickard's video's . He may have the info. that you are looking for.

 

[Mart Hale]

I was under the impression that the BMS is always active, monitoring the pack.
Its only that you cant connect multiple up and have them talking to each other. ??
Seems like people are just connecting pos and neg and running them...

If you discharge it, will it discharge to 0V?? or does it cut at like 16V or in that area?

------------------------------------

I have no idea if The BMS is active or how it works.... At present I have been connecting positive and negative together, however my last check on the pack shows me I have a problem ->

3.6
3.6
4.2
4.2
4.3
4.2

---

I have not, nor do I wish to drop the pack down to the min level, I have destroyed one pack by taking it down that low, not wanting to risk that. I have been using this just by powering the 24 V terminals on the battery. So now I need either to find a BMS for this, or to hack the one onboard.

1) Does the BMS use resistors to balance the packs?
2) Does the Car's main CPU control this BMS?

---

You might want to check out Jack Rickard's video's . He may have the info. that you are looking for.

Yes, he already has done it, I am unsure if he it going to sell his product or make it open source.

 

[The-J-Man]

I have no idea if The BMS is active or how it works.... At present I have been connecting positive and negative together, however my last check on the pack shows me I have a problem ->

3.6
3.6
4.2
4.2
4.3
4.2

---

I have not, nor do I wish to drop the pack down to the min level, I have destroyed one pack by taking it down that low, not wanting to risk that. I have been using this just by powering the 24 V terminals on the battery. So now I need either to find a BMS for this, or to hack the one onboard.

1) Does the BMS use resistors to balance the packs?
2) Does the Car's main CPU control this BMS?

---

---

---

That really doesnt look too good. One thing to consider is that nearly all BMS's balance on the top end, maybe keep it plugged in and charging for a longer time at max overal voltage. The BMS should, if active, balance them. Take not though that at 250Ah this can take a while, multiple days.
Actually the fact that two of the blocks are at 3.6V could be beneficial. If you discharge the pack those two will reach min level first. The BMS should then, if active, cut power delivery.
If you have access to terminals where you can measure the voltage then you also have access to terminals where you could push those packs back up individually. (where can you measure it anyway, isnt it aclosed unit, with that plastic top and bottom)Maybe manually balance all the blocks and start using it again? If the see the same behaviour you might have an issue.
Im confused as to why they would drop faster, those cells should all be identical in IR and capacity and age. Have you checked all the cell fuses? maybe a few have come lose or popped.Was the pack from a crashed vehicle?


1) no clue but its more likely it shifts power.
2) the main CPU definitely does not control the battery BMS's (it certainly talks to it though)as when in a crash the connection between car and battery is capped. The BMS's need to work alone, and should work individually as when one pack gets punctured in a crash the others still need to be in control of their pack. Otherwise you don't have a safe system. Another reason as to why im sure the BMS acts all the time alone.

Oh btw, if your module is from a 85kWh pack then its 3200mAh cells, the 90 and 100kWh packs use the 3350/3400mAh cells.

 

[Mart Hale]

Actually the fact that two of the blocks are at 3.6V could be beneficial. If you discharge the pack those two will reach min level first. The BMS should then, if active, cut power delivery.
If you have access to terminals where you can measure the voltage then you also have access to terminals where you could push those packs back up individually. (where can you measure it anyway, isnt it aclosed unit, with that plastic top and bottom)Maybe manually balance all the blocks and start using it again? If the see the same behaviour you might have an issue.
Im confused as to why they would drop faster, those cells should all be identical in IR and capacity and age. Have you checked all the cell fuses? maybe a few have come lose or popped.Was the pack from a crashed vehicle?

--------------------------

It does make sense that the bms would balance on the top end. I have been trying to figure out how in the world to charge just these 2 cells banks at such a high capacity.

But as I thought about it, I starting thinking these two cell banks are in series, if I charge them both together with a 12V 4 amp battery charger I could zero in on the cells that are a problem and get them charged evenly with the rest of the bank.


I will look for any blown cells in the bank Good suggestions.

Mart

It is starting to look like what I want is to create my own custom BMS that I can set the top voltage at 4.1 or 4.0 for the longevitity of this pack.

---

Ok, I think I know what is going on with these packs.....


I have set the top limit for my Outback charge control to be 4.1 X 6. = 24.6


However.... In order to reach the top limit of all 6 of these backs for the BMS to top level it would 4.2 * 6 = 25.2

What I need to do is change the top voltage of the charge controller up to 25.2 which will all all the cells to balance. Once the cells balance, I can then again set the voltage back down to 24.6 or hack the onboard BMS to set the upper limit to 4.1 ( which is what I would love to see )

 

[The-J-Man]

--------------------------

It does make sense that the bms would balance on the top end. I have been trying to figure out how in the world to charge just these 2 cells banks at such a high capacity.

But as I thought about it, I starting thinking these two cell banks are in series, if I charge them both together with a 12V 4 amp battery charger I could zero in on the cells that are a problem and get them charged evenly with the rest of the bank.


I will look for any blown cells in the bank Good suggestions.

Mart

It is starting to look like what I want is to create my own custom BMS that I can set the top voltage at 4.1 or 4.0 for the longevitity of this pack.

---

Ok, I think I know what is going on with these packs.....


I have set the top limit for my Outback charge control to be 4.1 X 6. = 24.6


However.... In order to reach the top limit of all 6 of these backs for the BMS to top level it would 4.2 * 6 = 25.2

What I need to do is change the top voltage of the charge controller up to 25.2 which will all all the cells to balance. Once the cells balance, I can then again set the voltage back down to 24.6 or hack the onboard BMS to set the upper limit to 4.1 ( which is what I would love to see )

---

Yeah either you connect them singely or you do both at the same time.
You need something like a iMax B6 to charge them. Dont use that 12V power supply, it will over charge them to 12V, not smart (BMS cant stop it because your behind it)You need a Li-ion Charger that can do 8.4V. The higher amperage the better. Any amperage will work, might just take a while. Im not sure a charger exists with wich you could charge those packs to quickly with, as that would be in the area of 200A so just blast away
You dont need your own BMS to be honest, just charge your cells fully every month and the BMS in the pack should take care of it. Note that the Teslas charge to full frequently and battery degradation on the cars is incredibly little. If you have a look at the data collected it looks like the cells will reach 3000-4000 cycles. A full charge every months to balance out your cells if they become out of balance wont harm them a single bit.
But I still think something isnt right with the pack if it gets that unbalanced, most likely less capacity in those blocks due to popped or broken fuses, check all of that first!!!
Yes there you go, set the charge cut off at higher. Still recommend pushing up those two low blocks before hand as it could take like a monthfor the BMS to balance charge those. It probably doesnt balance with more then 2A.

Hacking the BMS and editing the top voltage is a bit extreme : Probably easier to get a shell less Model S, popping all parts on the wall and using the MCU slider to change the battery charge limit

 

[Mart Hale]

Yeah either you connect them singely or you do both at the same time.
You need something like a iMax B6 to charge them. Dont use that 12V power supply, it will over charge them to 12V, not smart (BMS cant stop it because your behind it)You need a Li-ion Charger that can do 8.4V. The higher amperage the better. Any amperage will work, might just take a while. Im not sure a charger exists with wich you could charge those packs to quickly with, as that would be in the area of 200A so just blast away
You dont need your own BMS to be honest, just charge your cells fully every month and the BMS in the pack should take care of it. Note that the Teslas charge to full frequently and battery degradation on the cars is incredibly little. If you have a look at the data collected it looks like the cells will reach 3000-4000 cycles. A full charge every months to balance out your cells if they become out of balance wont harm them a single bit.
But I still think something isnt right with the pack if it gets that unbalanced, most likely less capacity in those blocks due to popped or broken fuses, check all of that first!!!
Yes there you go, set the charge cut off at higher. Still recommend pushing up those two low blocks before hand as it could take like a monthfor the BMS to balance charge those. It probably doesnt balance with more then 2A.

Hacking the BMS and editing the top voltage is a bit extreme : Probably easier to get a shell less Model S, popping all parts on the wall and using the MCU slider to change the battery charge limit

Yep I have been looking for a power supply that will match 4 v, 72 amps..... I do want to push both of the banks up, but not sure how to get it there.

I have considered putting a load on the other cells to get the resistance to match so the other cells will charge.

 

[CarelHassink]

Hello Mart,

I'm located in Europe, Netherlands.
I have done quite a few second life battery packs from ev's / phev's
including Tesla packs,

perhaps check out Collin Kidder,
he's the man behind Jack Ricard, who hacked the Tesla can protocol.
Jack is the marketing man selling this.

https://hackaday.io/project/10098-model-s-bms-hacking

Be aware everything is experimental etc, think before doing etc.
But my advice, run the Tesla pod with a bms. Tesla's or another

You couldgo with a Batrium, take out the Tesla bms board, use the leads to the 7 serial battery blocks
find the white connector as in foto, hook it up to longmons or this special version batrium module.
https://www.batrium.com/search?q=rc7s
and your good to go.





I have small video, showing a hacked Tesla bms board, done by a customer from Austria, 8 pods canbus connected, (using this set-up you could hook-up max 64 pods )
but cannot attach it to this post, or just don't know how to do it.
sent me an e-mail and i sent it to you.

best Carel Hassink

---

 

[The-J-Man]

Yep I have been looking for a power supply that will match 4 v, 72 amps..... I do want to push both of the banks up, but not sure how to get it there.

I have considered putting a load on the other cells to get the resistance to match so the other cells will charge.

That will be a very expensive power supply. Honestly just get a hobby charger like the one I have for like 40bucks. Attach the leads and charge with 5A or whatever. It will take like 2 full days but by the time youve found a 72A charger that doesnt cost like 1 grand you will be done.

---

Hello Mart,

I'm located in Europe, Netherlands.
I have done quite a few second life battery packs from ev's / phev's
including Tesla packs,

perhaps check out Collin Kidder,
he's the man behind Jack Ricard, who hacked the Tesla can protocol.
Jack is the marketing man selling this.

https://hackaday.io/project/10098-model-s-bms-hacking

Be aware everything is experimental etc, think before doing etc.
But my advice, run the Tesla pod with a bms. Tesla's or another

You couldgo with a Batrium, take out the Tesla bms board, use the leads to the 7 serial battery blocks
find the white connector as in foto, hook it up to longmons or this special version batrium module.
https://www.batrium.com/search?q=rc7s
and your good to go.


I have small video, showing a hacked Tesla bms board, done by a customer from Austria, 8 pods canbus connected, (using this set-up you could hook-up max 64 pods )
but cannot attach it to this post, or just don't know how to do it.
sent me an e-mail and i sent it to you.

best Carel Hassink

---

---

Very interesting. Do you know if the board is active on the modules if you use them singley?

 

[dan.fador]

Hello Mert,

do you solve the problem whit BMS? I have to S modules and for now are balanced but I want to use them whit a BMS. I order an Arduino Due and some another parts like sugest in hackaday project.

I'm curios how do you solve the problem.

Thanks!

 

[Dougrowe1]

Yep I have been looking for a power supply that will match 4 v, 72 amps..... I do want to push both of the banks up, but not sure how to get it there.

I have considered putting a load on the other cells to get the resistance to match so the other cells will charge.

That will be a very expensive power supply. Honestly just get a hobby charger like the one I have for like 40bucks. Attach the leads and charge with 5A or whatever. It will take like 2 full days but by the time youve found a 72A charger that doesnt cost like 1 grand you will be done.

---

Hello Mart,

I'm located in Europe, Netherlands.
I have done quite a few second life battery packs from ev's / phev's
including Tesla packs,

perhaps check out Collin Kidder,
he's the man behind Jack Ricard, who hacked the Tesla can protocol.
Jack is the marketing man selling this.

https://hackaday.io/project/10098-model-s-bms-hacking

Be aware everything is experimental etc, think before doing etc.
But my advice, run the Tesla pod with a bms. Tesla's or another

You couldgo with a Batrium, take out the Tesla bms board, use the leads to the 7 serial battery blocks
find the white connector as in foto, hook it up to longmons or this special version batrium module.
https://www.batrium.com/search?q=rc7s
and your good to go.


I have small video, showing a hacked Tesla bms board, done by a customer from Austria, 8 pods canbus connected, (using this set-up you could hook-up max 64 pods )
but cannot attach it to this post, or just don't know how to do it.
sent me an e-mail and i sent it to you.

best Carel Hassink

---

---

Very interesting. Do you know if the board is active on the modules if you use them singley?

---

---

Did you ever find out if the board is active on the modules if you use them singley?

 

[Dougrowe1]

Hello Mert,

do you solve the problem whit BMS? I have to S modules and for now are balanced but I want to use them whit a BMS. I order an Arduino Due and some another parts like sugest in hackaday project.

I'm curios how do you solve the problem.

Thanks!

How is this going so far? I'm looking at purchasing some of these tesla batteries. Wondering if you followed the hackaday project to completion.

 

[Mart Hale]

Ok, back on this thread again. I have fixed set of 2 low batteries. the solution was to put a 12 V car charger at 3 amp to the cells directly. I did this at 2 hour intervals checking and verifying that the power came up on them, after 14 hours of charging in this manner the cells are now balanced and I am breathing much easier I have found a youtuber that has a nifty method of keeping these balanced with a very small BMS, link ->

So far this looks like the solution for me, I will keep checking these but if this keeps the batteries balanced I may expand to more modules. I am also trying to do the math as to buying a used Nissan leaf ( whole car ) and attempting to put an inverter on the car so that I can have a portable battery bank, verses buying more tesla battery modules.


Thanks guys for your suggestions, I love it when I find a better solution ;-)

 

[Brian Wilson]

I have two of the ModelS packs and I bought the ZeroEV replacement board and use a cheap cell checker/balancer. The original Tesla slave board is a balancer but very very slow to balance. If you watch it in a very dark room you will see a very small led flash as it balances. The reason the balancing is so slow is the leads to the cell groups is so small and can not carry much current.

 

[CarelHassink]

Right, more updated info

1) as for the in-balance in the pack, most likely contamination with cooling fluids, that somehow made a slow short on the positive side of the18650's in that cell
I've had a few modules showing the same. Try to detect where on the module, which cell, mostly seen by a stain on the surface area. Be carefull module is live!!, but I managed to clean out with 98% synthetic alcohol, and blow dry immediately with an compressed dried air.
2) charging individual cellblocks in the module, I used a Revolectrix powerlab 8, controlled charge upto 30A pinpointed on the needed end-voltage matching the other cells
3) SIMPbms see this:https://github.com/tomdebree
Tom, now living in the UK again, has made a shield + Teensy3.2 + adjusted software, based on previous open sourcework of Collin Kidder, to manage the Tesla slaveboards present on each module. ( and some more oem cmu's of other brands)

This SIMPbms will do a lot more, and will work with cmu's from GS YUASA, VW, BMW (in progress) can controlled chargers like Brusa TC Elcon Eltek VOLT
more features
-Overvoltage monitoring
-Temperature monitoring
-Undervoltage monitoring
-Balancing during charge
-Current measurement
-AH counter

Outputs
-Contactors/relays or other 12V signals
-PWM 'gas/fuel'gaugecontrol
-Serial interface for Victron inverters

Simpbms boards + Teensy 3.2 pre-loaded with firmwareare for sale 300/pcs ex VAT ex Shipping
Tom has a small team, me I'm located in the Netherlands Europe, contact me if interested

[size=small]







[/size]

 

[Mart Hale]

[size=large]This guy has a great setup, I am using the bms he recommended, of sense I have found the same BMS under different names.

 

[CarelHassink]

Yup,
Tom de Bree wrote the software, designed the board/shield, together with Daniel Boekel,
See source on Github: https://github.com/tomdebree

It just works with Tesla modules, really pleased.
Hooks up to Victron and SMA kit,
currently testing new project
SIMPbms to SMA HV StorageBoy series,
https://files.sma.de/dl/31186/SBS37-60-DAU1902-V21web.pdf

so we got hold of such an inverter, capable handling from 100Vdc till 550Vdc, plus a stack of BYD HV battery modules + BMU
Plan is to sniff the can protocol, and hopefully Tom de Bree can do his magic making the code.

If all works,we can hook-up complete EV packs to SBS series

Carel