2010 Evil DH, 8000watts, LR Big Block

[TysonScott]

This is my first build, I have been hanging around here for about a year. My goal is to achieve a very secure fast build. I like to hit jumps, drops, dh tracks and tow my friend, as well as having motorcycle (~50mph) speeds to use a source of transportation.

Parts
Motor: I ordered a Lightning Rods Big Block motor kit for the 83mm bb 750$ after shipping
Battery: I got some ebike battery's from Battery Hookup they are 4 x 10s 4p 36v 12.8ah, cells are MH1. I hope to put these 4 packs into 2s 2p for 72v 25.6ah. 250$ for cells + 100$ for 72v 100a bms
Controller: I'm not sure about this one yet (any suggestions?), I'm looking into the shul cl350.
Bike: 2010 Evil Revolt DH Bike 775
Drivetrain: I'm just gonna upgrade as I break what is currently there
Throttle: Going to buy an IP67 domino as soon as I can find one that's not 150$

Bike
It has a really nice flat area to mount the battery to and the downtube goes up linearly right after the BB so I don't have to worry about my motor hanging down.

Some BEEFY forks, the damper on it actually worked super well for a 12 year old bike from a guy who did not seem like the regularly scheduled maintenance type of guy.

For the battery I taped them together to get the shape of them, to see if it'll fit, it does. And I tested the fork compression and it does not rub.

The guy who owned the bike before me was a KOOK. Hardly ever cleaned it and stored it outside, but the damage wasn't too bad. I took it and stripped every thing down to clean and regrease. Unfortunately he thought that the suspension settings were loose ( since the knobs moved) so he had tightened all of them down as much as possible I was able to undo them though with a little rubber clamp thing. Other than that the bike was in pretty good condition.


Problems:
Connecting the battery modules. i know i cant just connect the main positive and negatives of each module together into 2s 2p and the bms still work, but im not sure how to connect the modules to achieve the 20s 8p battery and connect the bms. If anyone can give me some advice that would be greatly appreciated.

 

[handlworks]

That's a lotta battery! Pumped to see where this build goes.

 

[amberwolf]

Battery: I got some ebike battery's from Battery Hookup they are 4 x 10s 4p 36v 12.8ah, cells are MH1. I hope to put these 4 packs into 2s 2p for 72v 25.6ah. 250$ for cells + 100$ for 72v 100a bms
<snip>

Problems:
Connecting the battery modules. i know i cant just connect the main positive and negatives of each module together into 2s 2p and the bms still work, but im not sure how to connect the modules to achieve the 20s 8p battery and connect the bms.

A link to the specific modules' sale page may provide helpful info, if there is more than what you've listed above, and/or there are other images showing the internal module construction..

If the packs are 10s4p each, then seriesing 4 of them in pairs would provide 20s 8p.

To use a single 20s BMS to monitor them all you would then need to connect the positives of all 8 of the most negative cells together (by making a connection from one of the modules' cells across to the other one "beside" it in the complete setup), and that to the most negative of the positive input of the BMS sense wires.

Then the positives of the next most positive 8 cells would be connected together, and that then to the next most positive input of the BMS sense wires.

Repeat for every set of 8 cells to create 20 parallel groups of 8, each wired to the BMS in order per the BMS's specific wiring directions (which could be different than the above, but probably are not).

Before making any of the connections you will need to test the voltages of each 4p group, to make sure they are all as close to the same as possible as the equivalent 4p group in the module you will be paralleling to it, and that all groups in all modules are approximately equal. If any of them are significantly different, you will probably have trouble with those over time as they are not the same properties as the rest of the groups and will not charge or discharge the same, and will change more with time and usage.


Side note: If these are the LG MH1, a google search says they are "10A" cells. If you only have 8p, that's only 80A; so you may wish to program the BMS to limit at that rather than the 100A which will strain the cells, causing excessive voltage sag, internal heating, and premature aging. I'd also recommend tuning the controller not to pull more current than that (less if possible, to make it less stressful for the cells).

It also means that the system won't be 8kW, since the battery will only be able to safely provide 80A x 72v = 5760w, and since the system is not 100% efficient (probably around 80 or less) it's likely something closer to half the 8kW power level you are after.


Also keep in mind that as these are used cells, they probably don't have their full original capacity of 3200mAh each (and may sag more at full current draw than if they were new). How much capacity they do have (and how much current they can really output) you'll have to determine experimentally once the system is built and can be tested under real-world loading.

It also depends on the LVC for your controller and the BMS cell-level LVC, as well as the BMS HVC and your charger voltage. The full capacity of the cells is only available by going all the way down to the absolute minimum cell votlage, and up to the absolute max. Most systems are setup to avoid the former, taking at least a few percent off right there; if you would like the cells to last longer lifespan then increasing LVC and decreasing HVC (if the BMS is programmable) may help with that, but will decrease available capacity.

 

[TysonScott]

Battery: I got some ebike battery's from Battery Hookup they are 4 x 10s 4p 36v 12.8ah, cells are MH1. I hope to put these 4 packs into 2s 2p for 72v 25.6ah. 250$ for cells + 100$ for 72v 100a bms
<snip>

Problems:
Connecting the battery modules. i know i cant just connect the main positive and negatives of each module together into 2s 2p and the bms still work, but im not sure how to connect the modules to achieve the 20s 8p battery and connect the bms.

A link to the specific modules' sale page may provide helpful info, if there is more than what you've listed above, and/or there are other images showing the internal module construction..

If the packs are 10s4p each, then seriesing 4 of them in pairs would provide 20s 8p.

To use a single 20s BMS to monitor them all you would then need to connect the positives of all 8 of the most negative cells together (by making a connection from one of the modules' cells across to the other one "beside" it in the complete setup), and that to the most negative of the positive input of the BMS sense wires.

Then the positives of the next most positive 8 cells would be connected together, and that then to the next most positive input of the BMS sense wires.

Repeat for every set of 8 cells to create 20 parallel groups of 8, each wired to the BMS in order per the BMS's specific wiring directions (which could be different than the above, but probably are not).

Before making any of the connections you will need to test the voltages of each 4p group, to make sure they are all as close to the same as possible as the equivalent 4p group in the module you will be paralleling to it, and that all groups in all modules are approximately equal. If any of them are significantly different, you will probably have trouble with those over time as they are not the same properties as the rest of the groups and will not charge or discharge the same, and will change more with time and usage.

thank you for such a detailed answer!
the cell modules I bought were these, however i got them from battery hookup.
https://jag35.com/collections/lithium-batteries/products/40-cell-lgmh1-18650-lithium-scooter-battery-pack

 

[amberwolf]

There's a bit more info in the post now; I was adding to it when you replied, I guess. I quoted it below for convenience.

Note that the BMS brand the Jag page reommends (Daly) has some serious failures in various posts here on ES (the most recent by Methods, who has designed various types of BMSs for different applications). Whether any particular model or batch is problematic, I don't know; you'd have to test yours under realworld conditions to verify operation, and I'd check it periodically just to be sure it's still doing it's job (the kinds of failures they exhibit seem to be "silent", in that there's nothing obviously wrong except that it's not protecting the cells as it should be).

Side note: If these are the LG MH1, a google search says they are "10A" cells. If you only have 8p, that's only 80A; so you may wish to program the BMS to limit at that rather than the 100A which will strain the cells, causing excessive voltage sag, internal heating, and premature aging. I'd also recommend tuning the controller not to pull more current than that (less if possible, to make it less stressful for the cells).

It also means that the system won't be 8kW, since the battery will only be able to safely provide 80A x 72v = 5760w, and since the system is not 100% efficient (probably around 80 or less) it's likely something closer to half the 8kW power level you are after.


Also keep in mind that as these are used cells, they probably don't have their full original capacity of 3200mAh each (and may sag more at full current draw than if they were new). How much capacity they do have (and how much current they can really output) you'll have to determine experimentally once the system is built and can be tested under real-world loading.

It also depends on the LVC for your controller and the BMS cell-level LVC, as well as the BMS HVC and your charger voltage. The full capacity of the cells is only available by going all the way down to the absolute minimum cell votlage, and up to the absolute max. Most systems are setup to avoid the former, taking at least a few percent off right there; if you would like the cells to last longer lifespan then increasing LVC and decreasing HVC (if the BMS is programmable) may help with that, but will decrease available capacity.

 

[TysonScott]

There's a bit more info in the post now; I was adding to it when you replied, I guess. I quoted it below for convenience.

Note that the BMS brand the Jag page reommends (Daly) has some serious failures in various posts here on ES (the most recent by Methods, who has designed various types of BMSs for different applications). Whether any particular model or batch is problematic, I don't know; you'd have to test yours under realworld conditions to verify operation, and I'd check it periodically just to be sure it's still doing it's job (the kinds of failures they exhibit seem to be "silent", in that there's nothing obviously wrong except that it's not protecting the cells as it should be).

Side note: If these are the LG MH1, a google search says they are "10A" cells. If you only have 8p, that's only 80A; so you may wish to program the BMS to limit at that rather than the 100A which will strain the cells, causing excessive voltage sag, internal heating, and premature aging. I'd also recommend tuning the controller not to pull more current than that (less if possible, to make it less stressful for the cells).

It also means that the system won't be 8kW, since the battery will only be able to safely provide 80A x 72v = 5760w, and since the system is not 100% efficient (probably around 80 or less) it's likely something closer to half the 8kW power level you are after.


Also keep in mind that as these are used cells, they probably don't have their full original capacity of 3200mAh each (and may sag more at full current draw than if they were new). How much capacity they do have (and how much current they can really output) you'll have to determine experimentally once the system is built and can be tested under real-world loading.

It also depends on the LVC for your controller and the BMS cell-level LVC, as well as the BMS HVC and your charger voltage. The full capacity of the cells is only available by going all the way down to the absolute minimum cell votlage, and up to the absolute max. Most systems are setup to avoid the former, taking at least a few percent off right there; if you would like the cells to last longer lifespan then increasing LVC and decreasing HVC (if the BMS is programmable) may help with that, but will decrease available capacity.

Yea I was questionable about the cells, some places say 17 (seems pretty high though for 3200mah) some say 10a but either way I'll probably get a better battery later on when I have more confidence in my ability to build one. But for right now im just gonna try and wire these up to prove to learn the basics. Also the BMS Was a battery hookup one, I had heard questionable things about the daly ones aswell. Do you know much about the battery hookup bms?

 

[amberwolf]

Yea I was questionable about the cells, some places say 17 (seems pretty high though for 3200mah) some say 10a

As a general rule, it's safer to assume places that sell batteries and cells are exagerating or even outright lying about specs to make sales, and if you really want to know what they can do, find places like lgyte.dk or other sites that actually test batteries (but don't sell them), or threads here on ES (and similar data-and-DIY-oriented forums) by people like Pajda, etc.

Meaning...confirm before finalizing a decision based on such specs, so you don't have any surprises later on.


Also, if a seller gives specs for a cell that don't match general battery info sites like lgyte.dk, you may want to then disregard everything that seller says about any of their stuff, because the likelihood is higher that they're either exaggerating, lying, or perhaps just using the absolute max instantaneous peak just before fireball ratings without telling you that to make sales, and letting you think that it's ok to run them that hard continously.

I think there are too many places operating under the "what you don't know can't hurt their bottom line" method.


In general, it's also a good idea to pick stuff (anything, not just batteries) not based on it's maximum ability to barely handle the load, but to figure in some large margin of capability. This will help in several ways--when new, the system will not be stressed, and will perform better and possibly more efficiently, vs a system being run at it's limits. Less heat, etc. Will be larger and weigh more, but will last longer and age more slowly. The exact number for any system to use for margin would need to be determined by the designer/user based on the specific use-case, but a possible rule of thumb might be 25-50% more than your use-case requires, of capability and capacity, for batteries.

For controllers, BMS, etc., you could use the same rule of thumb, regarding current handling ability and voltage handling, as long as the controller is programmable so you can set the current limit, HVC/LVC etc to match your actual usage and battery and motor limits, etc. If it's not programmable and has a higher limit for these it won't help the longevity of the other parts of the system.

If the parts themselves were designed with any margin in them, you wouldn't really need to do this, but since they generally are not, it's up to the system designer / user to pick parts that are more capable than needed so the system as whole has that margin, for reliability.

Do you know much about the battery hookup bms?

Unfortunately I haven't even seen any posts about them.

 

[99t4]

Also, if a seller gives specs for a cell that don't match general battery info sites like lgyte.dk, you may want to then disregard everything that seller says about any of their stuff, because the likelihood is higher that they're either exaggerating, lying, or perhaps just using the absolute max instantaneous peak just before fireball ratings without telling you that to make sales, and letting you think that it's ok to run them that hard continously.

I think there are too many places operating under the "what you don't know can't hurt their bottom line" method.

I perceive this almost as a Zen Master life lesson. Applies to so many aspects of life: Retail purchases, higher education marketing, vitamin supplements, investment opportunities, lawyers, politics, dating apps, etc.

 

[Bengy22]

Controller: I'm not sure about this one yet (any suggestions?), I'm looking into the Bac 2000. Im not sure where to buy one or if ill be able to program it well. 300-400$ i think

I know lightningrods is working with HV for a bac2000 kit for the BB motor, I'm not sure if you are in the HV discord but you can ask either one about it and they will get you all the information you need. Iirc it's $650 but that's for the full kit with a display. And Greg (Mr. High Voltage) offers lifetime reprogramming on the bac's for free iirc as well.

 

[TysonScott]

I was looking at shul motors controller and i found this on. Its a quite a bit cheaper and ive heard that the vesc software is easier to program and use. than the asi software. Ive also been following this guys progress on building these controllers on facebook.

https://3shulmotors.com/shop/cl350/?v=7516fd43adaa



Also i was looking into using the 1 bms on the 4 battery's and it seemed pretty complicated. would it work to put a 36v 40a bms on each module then just make a xt90 dongle to connect them in 2s2p? is this safe?
https://www.youtube.com/watch?v=lthyqyPgacs&t=293s
https://www.youtube.com/watch?v=zBUSKtCw_kA&t=345s
This guy shows how i would connect the bms.

also i could just make a rectangle box instead of having a bit of a wonky shape to accommodate for the bms and when im done using these battery's for this ebike ill be able to use them to test other things. Ill probably eventually upgrade the battery to samsung 30q or some pouch cells. I just want to get it running for now.

samsung 30q are rated at 15a cont x8 = 120a which is also the cont that the shul controller is rated for. so mabye 10kw peak at 17.5a per cell?

 

[mxlemming]

I was looking at shul motors controller and i found this on. Its a quite a bit cheaper and ive heard that the vesc software is easier to program and use. than the asi software. Ive also been following this guys progress on building these controllers on facebook.

https://3shulmotors.com/shop/cl350/?v=7516fd43adaa



Also i was looking into using the 1 bms on the 4 battery's and it seemed pretty complicated. would it work to put a 36v 40a bms on each module then just make a xt90 dongle to connect them in 2s2p? is this safe?
https://www.youtube.com/watch?v=lthyqyPgacs&t=293s
https://www.youtube.com/watch?v=zBUSKtCw_kA&t=345s
This guy shows how i would connect the bms.

also i could just make a rectangle box instead of having a bit of a wonky shape to accommodate for the bms and when im done using these battery's for this ebike ill be able to use them to test other things. Ill probably eventually upgrade the battery to samsung 30q or some pouch cells. I just want to get it running for now.

samsung 30q are rated at 15a cont x8 = 120a which is also the cont that the shul controller is rated for. so mabye 10kw peak at 17.5a per cell?

Strongly recommend against the 30q cells. Serious self discharge issues that eventually wrote off my first pack pointlessly early. Now using p42a cells which are worlds better.

 

[TysonScott]

I was looking at shul motors controller and i found this on. Its a quite a bit cheaper and ive heard that the vesc software is easier to program and use. than the asi software. Ive also been following this guys progress on building these controllers on facebook.

https://3shulmotors.com/shop/cl350/?v=7516fd43adaa



Also i was looking into using the 1 bms on the 4 battery's and it seemed pretty complicated. would it work to put a 36v 40a bms on each module then just make a xt90 dongle to connect them in 2s2p? is this safe?
https://www.youtube.com/watch?v=lthyqyPgacs&t=293s
https://www.youtube.com/watch?v=zBUSKtCw_kA&t=345s
This guy shows how i would connect the bms.

also i could just make a rectangle box instead of having a bit of a wonky shape to accommodate for the bms and when im done using these battery's for this ebike ill be able to use them to test other things. Ill probably eventually upgrade the battery to samsung 30q or some pouch cells. I just want to get it running for now.

samsung 30q are rated at 15a cont x8 = 120a which is also the cont that the shul controller is rated for. so mabye 10kw peak at 17.5a per cell?

Strongly recommend against the 30q cells. Serious self discharge issues that eventually wrote off my first pack pointlessly early. Now using p42a cells which are worlds better.

It was between the 30q and the vtc6. Are the vtc6 known to be reliable?

 

[mxlemming]

I was looking at shul motors controller and i found this on. Its a quite a bit cheaper and ive heard that the vesc software is easier to program and use. than the asi software. Ive also been following this guys progress on building these controllers on facebook.

https://3shulmotors.com/shop/cl350/?v=7516fd43adaa



Also i was looking into using the 1 bms on the 4 battery's and it seemed pretty complicated. would it work to put a 36v 40a bms on each module then just make a xt90 dongle to connect them in 2s2p? is this safe?
https://www.youtube.com/watch?v=lthyqyPgacs&t=293s
https://www.youtube.com/watch?v=zBUSKtCw_kA&t=345s
This guy shows how i would connect the bms.

also i could just make a rectangle box instead of having a bit of a wonky shape to accommodate for the bms and when im done using these battery's for this ebike ill be able to use them to test other things. Ill probably eventually upgrade the battery to samsung 30q or some pouch cells. I just want to get it running for now.

samsung 30q are rated at 15a cont x8 = 120a which is also the cont that the shul controller is rated for. so mabye 10kw peak at 17.5a per cell?

Strongly recommend against the 30q cells. Serious self discharge issues that eventually wrote off my first pack pointlessly early. Now using p42a cells which are worlds better.

It was between the 30q and the vtc6. Are the vtc6 known to be reliable?

I haven't heard anything bad about them but honestly my battery knowledge is quite limited, I just know from bitter experience that 30q is as shite as they come. The only positive I have about them is that they didn't yet catch fire.

I definitely had genuine ones, and they were welded properly and used with a fully functional BMS.
Edit:
https://endless-sphere.com/forums/viewtopic.php?f=14&t=110052&hilit=Samsung+30q+discharge

30q= nope. Just dont go there.