Custom Battery Packs & BMS in a Monster Pro

[-XX-]

I bought a Monster Pro from somebody and don’t realize it had a aftermarket BMS and a custom battery pack setup. He didn’t say anything and I didn’t even think to open up the machine and inspect it. The set up is (4) 6s8p packs connected in series with a single 24s BMS. The batteries are model 3-NCR21700, 4800mAh, best I can come up with is that they’re what’s used in a Tesla model 3, possibly made by Panasonic, possibly just a Chinese knock-off. 
A while back I had a cut out, didn’t realize how low the battery was and couldn’t hear the beeper at 50mph so I was still pushing it. I charged the wheel and it sat for a few months before got the parts to rebuild it, but now there’s a dead string of batteries in one of the packs. 
Here’s my questions:

1. What is the advantage of (4) 25.2v packs in series as opposed to the (4) 100.8v packs in parallel.

 2. Is only just the one bms enough? I would think each pack should have its own board plus one for the entire system. 

3. Can I charge just the string of dead cells, can I replace just the dead cells, or should I scrap the whole thing and buy new packs.

4. Is this a decent BMS or should I replace it?

[Planemo]

Wow, someone surely must know something about this build, it's pretty unique. The BMS actually looks OK, but I know you have to be careful using generic BMS's on EUC's, something to do with not wanting any current limitation on the output I think.

Re Q1, I don't think theres anything inherently wrong with the battery config, many EUC's run serial packs these days.

Re Q2, providing the BMS is advanced enough, it may even be better than having one per pack like we are used to seeing lol

Re Q3, deffo worth trying to charge the dead cells and running a few cycles to see how they are performing. Presume they are finished though, especially if the voltage has gone below 2.5v. no idea where you would pick up matching cells if you wanted to replace that string though..

Someone will be along shortly who knows a lot more

 

[RagingGrandpa]

1 hour ago, -XX- said:

I had a cut out, didn’t realize how low the battery was and couldn’t hear the beeper at 50mph so I was still pushing it.

now there’s a dead string of batteries in one of the packs. 

Can I charge just the string of dead cells?

You shouldn't.

If any cells are resting below ~1.5 volts, they are permanently damaged and are hazardous to re-use. 
 

1 hour ago, -XX- said:

can I replace just the dead cells

You could, but this is not a very practical choice, because you'll have to find good cells of the same type, and weld them in place in a robust way using special tools and supplies. 
 

1 hour ago, -XX- said:

The batteries are model 3-NCR21700

Those 2170's were popular in the aftermarket during 2019, because there were a number of Tesla Motors EV packs that were rejected and salvaged. An EV (automobile) pack contains thousands of cells, and typically the packs are rejected for build reasons not related to the cells themselves... but it's a risky and inconsistent thing, to rip cells out of a pack assembly and then resell them separately. The chemical guts of the cell may have been new; or maybe not (depending on how the EV pack was tested). Maybe the cells got wet. Maybe cells were stored at full charge in a hot location for a few months. Maybe 5 of the cells came from a gently-used pack; and other 3 came from a pack that experienced a short-circuit. You can't know... so better to avoid "salvage" cells. 
 

1 hour ago, -XX- said:

should I scrap the whole thing and buy new packs

It's the safer bet. 
Compare the price of replacement packs, to the cost of restoring the interior of a house after a very-smoky fire. 

[alcatraz]

How dead are the cells?

The good news about the 25v x 4 configuration is that replacing one pack is enough to get it back to operation. The bad news is that it's always hard to find well matched cells. Might be worth a try if you want to try and save it.

This whole pack setup smells local chinese diy. I've seen them and I actually have one just like it as an addon pack. Same cells NCR21700 - 4800 and a 达理 bms.

These BMS' are very cheap and take forever to balance, BUT if it works why not use it? The beauty is that you have the wires there and can confirm cell voltages manually. It's not all sealed away in a mystery cube.

The reason for the cutout is possibly because the pack output is routed through the bms. A no-no on eucs. Ensure the output is connected straight to the pack + and -. Then you're good to go.

And remember that you don't have overvoltage protection when braking on a full charge and temperature monitoring. You can install a 1S buzzer to get around it. It will buzz if you go over 4.20v or cross a certain temperature.

[-XX-]

20 minutes ago, alcatraz said:

The reason for the cutout is possibly because the pack output is routed through the bms. A no-no on eucs. Ensure the output is connected straight to the pack + and -. Then you're good to go.

The positive output goes directly to the main board, the neg is routed thru the bms. Are you saying to bypass the bms entirely with the neg, or to add a jumper to bypass it but also leaving it connected, or what? However I’m pretty sure the cut out was me trying to push 50+mph when the battery was almost dead. At that speed I couldn’t hear the beeper but now I got a Speed Guardian, it’s hella ugly on the side of my helmet but it works.

[alcatraz]

Bypass entirely. If you want some protection add a 100A fuse. That way a broken board wont short the packs. Not all board failures short the packs, but it could happen.

The negative lead goes through the bms so it can cut off the output. That is not safe. Also, it might not even be compatible with braking currents. These bms were designed to be used with scooters and such.

[alcatraz]

To be clear. You still run a negative cable from the pack to the bms, so you get proper charging monitoring. (B-)

Don't use the P- terminal.

Edited August 30, 2023 by alcatraz

[alcatraz]

What's the voltage on that "dead" cell group?

[Mike Sacristan]

It sounds like a Loomo wheel. They released a 2700Wh Nik as well. And a Monster. I did not know of any Monster Pro.

So one of the riders in our group had trouble with her Loomo Nik a few months ago. It didn't charge to full anymore so I knew it was a dead cell group or a problem with the BMS. We opened the wheel and measured the cell groups and one group was dead. We got the pack repaired by a battery guy. He replaced the bad cells with Samsung 50G since they are very close in specs to the Tesla cells. I stress tested the wheel with a donkey ride and it worked perfectly. Voltage sag and recovery was as expected. Now a few months later the wheel still works perfectly and charges as it should.

There is only room for three packs in the Nik and they were different max voltages... but still connected in series to make 100.8. The problem with this is if one pack fails entirely then the wheel becomes a 75v wheel or even less which would mean instant tiltback/cutoff while riding. While if my Fat Nik with three packs in parallel had a pack failure I would still be riding a 4s6p 1800Wh Nikola Plus. 

[alcatraz]

Maybe the dead group is a loosened balance wire. It would show up as two dead groups though.

[-XX-]

On 8/29/2023 at 7:03 PM, alcatraz said:

What's the voltage on that "dead" cell group?

The dead string was less than a volt. I connected it to a good 18650 and charged it to 2.5v, then connected it to my nitecore charger and it took a full charge. I then charged a cpl other groups in that pack which were a little low so now all 24 groups are equal and I'm about ready to rewrap this pack. 

On 8/29/2023 at 7:03 PM, alcatraz said:

What's the voltage on that "dead" cell group

 

[-XX-]

On 8/29/2023 at 7:00 PM, alcatraz said:

To be clear. You still run a negative cable from the pack to the bms, so you get proper charging monitoring. (B-)

Don't use the P- terminal.

Can I simply add a wire connecting P- & B- to get the same result

[alcatraz]

Well, the output P- isn't used at all. You can leave it unconnected. The plus wires are all connected together. Ensure that the board has thick wires all the way from the pack. The bms doesn't need thick wires dispite the terminals looking beefy. It's because you don't use its shutoff management. You basically just use it for charging (slow charging up to 5A)

Charging a 1v lithium cell back to normal voltage is a risky move.

People have done it before when building powerwalls with fuse wires (zero risk of one cell shorting out the cell group = explosion). However on an euc you better be careful. That cell group is compromised meaning, don't store the wheel indoors and wear your gear when riding even slow.

 That bms is unlikely going to be able to keep a pack that had a group at 1v, running reliably. The group will have accelerated self discharge. (discharge faster than the other groups). And the bms balancer will not be able to drop the voltage on the remaining groups fast enough to match. You'll see it by the pack not charging to 100%.

If that's the case you need to replace the dead cells. Any engineer would recommend you to do it now already.

Powerwall use with fuse wires is one thing. Vehicle use another.

Edited September 9, 2023 by alcatraz

[RagingGrandpa]

On 9/9/2023 at 1:54 AM, -XX- said:

The dead string was less than a volt.