How BMS Works (Gotway pack teardown)

[alcatraz]

The balancing time depends on the imbalance. If you notice that your pack voltage after charging isn't what it used to be it's probably because it hasn't finished balancing, or is unable to balance.

When a pack is new the balancing circuit does very little and the cells arrive at 4.2v at nearly the same time.

[alcatraz]

Most bms can only cope with small imbalances.

Unfortunately most work only by discharging the high groups. Imagine if 1 or 2 groups are low by 0.2v. 18 groups (tons of charge) need to be dissipated through the tiny resistors which takes forever for the high capacity packs. Lets say it discharges them by a mere 0.03v then it charges back up to 4.2v (sometimes 4.23-4.25, erroneously, poor voltage sensoring) .

That cycle needs to be repeated like 10-20 times. Each time the high groups could be going over 4.20v by a small amount which isn't good.

So most stock bms' work kind of poorly when it comes to balancing. First they aren't really needed, then they work for a few cycles, until they can't efficienty cope with the imbalance and it just starts to further damage the pack with all the balance cycling.

So charging your wheel when it's new consists of a time of 99% charging and 1% balancing. When it's crap it's 10% charging and 90%. If the balancer would work better, by charging the low groups, then you could come back to 98% - 2% even with older packs. You can continue to ride packs that would have been retired 10.000km ago safely this way.

[Chriull]

18 hours ago, RagingGrandpa said:

@NuRxG - it won't help

If a pack is imbalanced, the imbalance will remain as you discharge it.

Imho one of the most important points - healthy cells do not get imbalanced by discharging! With a following charge they'll reach the same voltages again as before - no (?real?) matter which capacitance, resistance, etc they have!

Quote

The only way to correct the imbalance with our BMS is to achieve the top-balancing cell voltage threshold of 4.20V, which causes the bleed resistors to be activated for the high cells. During discharge, all cells will quickly become below 4.20V, and so the bleed resistors will do nothing.

And as @NuRxG mentioned the balancing effect is "doubled" - after the charge the bleeding resistor discharges the cell downto the ~4.19V again. 

So beside the bleeding resistors charging and discharging is a "highly symmetric" process.

Just once a cell is pushed too much in borderline conditions it starts to drift off ...

 

 

Like mentioned in TI's pdf (1) (imho a nice "little" overview regarding (im)balance)

"To make the matters worse, the affects of cell degradation caused by imbalance is autoaccelerating, once a cell has a lower capacity, it is exposed to increasingly higher voltage during charge which makes it degrade faster so its capacity becomes even less, which closes the runaway circle."

Somewhere i also read, that li ion cells increase impedance at their low S(tate)O(F)C(harge) % /voltages - could not find it in this paper (so was somewhere else or i overlooked it?). So this with the non linearity between voltage and SOC is imho the second reason for misbalances to happen.

So, from what i gathered till now, your case D (one (or sometimes two ) cell with too high voltage - the weakest, most beaten one) gets weakened but a bit balanced during the final stage of charging. And the same once discharged - it will become by it's lower capacity more discharged as the others and again stressed by low voltage condition...

Most likely once a misbalance is noticed it's already too late and one should not think about how to rebalance the pack but replace it or the weak cells!

Edit: Maybe by not discharging the pack too much one could extend the usable period after some "forced/intesified balancing"

(1) https://www.ti.com/download/trng/docs/seminar/Topic 2 - Battery Cell Balancing - What to Balance and How.pdf

5 hours ago, NuRxG said:

This is pretty hypothetical tho since batteries only have a few hundred cycles and were talking hundreds of hours of balancing time lol!

Cycles count as full cycle from 0% to 100% and downto 0% again. If one would just discharge some % and then charge again for balancing this would just be some hundreth of a cycle.

But as written above stressing and degrading exactly the weakest link of the pack again...

Edited March 3, 2021 by Chriull

[RagingGrandpa]

Supposedly, Gotway is reducing its BMS charging current limit settings.

My 800wh 84V modules from 2019 would accept about 10A each (unsafe for charging; tested only briefly).
This provides short-circuit protection at the charging port.

Below claims the new 900wh 100V modules will now permit only 3A each.

It would suck to not be able to use our popular 900w chargers on 1800wh wheels like Hero and RS
There were occasional reports of overheated and charred BMS charge-stop FETs in some RS and MSP though... perhaps this is the response.
 

Quote

EUC.SALE - Important information regarding new BEGODE PRODUCTS.
Due last safety updates to the all new BEGODE wheels like RS, EXN, TESLA, HERO, COMMANDER and etc.
NEW protection board was added to the every Battery pack.

Please note: 900wh, 100V battery pack have max charging current limit 3A.

Example
BEGODE RS have 2x900wh, total 1800wh. Max charging curent 6A!!!!
EXN - 9A
Monster PRO, Commander 12A.

Charging with more then rated, can damage Battery pack and will be not allow by battery protection system.

https://www.facebook.com/groups/ElectricUnicycle/posts/4440719026026026/

Edited October 20, 2021 by RagingGrandpa

[Planemo]

What a joke. Limiting input current to such a woeful amount when the cells/P construction itself could quite happily and safely take far more says nothing to me other than Begode can't build bms' for shit.

[RagingGrandpa]

More BMS changes from Gotway in 2021: we suspect they've added pack-output shutoff transistors.

(Those are good to prevent excessive arcing during a controller or wire harness failure; but risky because if they shut off at the wrong time, it results in a cut-out crash while riding.)

Hope we get to see inside these new packs soon!

  

11 hours ago, meepmeepmayer said:

EcoDrift shows the latest Monster Pro

[div]

On 10/20/2021 at 9:32 PM, RagingGrandpa said:

Below claims the new 900wh 100V modules will now permit only 3A each.

It seems most of the LG 21700 cells have a standard charge rate of about 0.3C, which gives about 6A charge speed for a 1800Wh battery :/ (so even before BG stated this it was the case for the cell specs)

[RagingGrandpa]

59 minutes ago, div said:

LG 21700 cells have a standard charge rate of about 0.3C

Sure, but that's not the maximum rated charging current.
0.3C is not a "must not exceed" number. Think carefully about what's happening when you do an emergency stop while riding the EUC... (>20A charging!)

And, what if GW/BG uses nicer cells, like the 50E's showing up in their packs today? Will we be stuck with 3A per pack?
Those cells specify a maximum charging current of (4.9A * 2P = 9.8A per pack). A shame to not let us use the cells we're paying for.

 

Edited November 23, 2021 by RagingGrandpa

[div]

6 hours ago, RagingGrandpa said:

0.3C is not a "must not exceed" number. Think carefully about what's happening when you do an emergency stop while riding the EUC... (>20A charging!)

Those are completely different situations: one case is recommended « healthy » charge current where you choose your charger, the other (braking hard) is not a choice in general, and it is wear on the cells one have to accept.

Same as “do not discharge bellow” doesn’t mean “discharge to”.

I’m all for rapid charging, hence I did that charge rate calculation a good while ago, and found the LG 21700 disappointing.

edit: the 50E looks nice, with 0.5C standard charge. The new packs ought to be specced / rated accordingly.. (I also like the cycle life of 500->80% where my Sherman’s Panasonic only do 300-> 70%.

Edited November 24, 2021 by div

[alcatraz]

LG M50T and Samsung 50E are very similar. They both can't handle sustained 15A without overheating. They are 10A cells.

Samsung 50G are 15A sustained cells.

[RagingGrandpa]

Also learned that some Kingsong S18 BMS did include status LEDs that reveal activity of the balancing resistors, which would be handy for troubleshooting a disassembled pack...

https://forum.electricunicycle.org/topic/17867-king-song-s18-discussion/?do=findComment&comment=406023

19 hours ago, enaon said:

https://streamable.com/t8e53z

you can see them here blinking, it is a video from a Russian user.

And one more thing I noticed just now, while upoading the video for you. His bms has an mcu on it, mine has not, very interesting  

[RagingGrandpa]

GW/BG added more safety features to the new 134V Master

Notably:

• Cell temperature sensors (2 per BMS)
• Audible beeper in each BMS
  • Alarm for overtemperature
  • Alarm for disconnected balance wires
  • (and likely additional conditions)

Neat! 

 

[widdershins]

On 10/20/2021 at 12:32 PM, RagingGrandpa said:

Supposedly, Gotway is reducing its BMS charging current limit settings.

My 800wh 84V modules from 2019 would accept about 10A each (unsafe for charging; tested only briefly).
This provides short-circuit protection at the charging port.

Below claims the new 900wh 100V modules will now permit only 3A each.

It would suck to not be able to use our popular 900w chargers on 1800wh wheels like Hero and RS
There were occasional reports of overheated and charred BMS charge-stop FETs in some RS and MSP though... perhaps this is the response.
 

damn. do we know what this looks like on the EX20S packs?

[RagingGrandpa]

14 hours ago, widdershins said:

[charging current limits]

damn. do we know what this looks like on the EX20S packs?

100V EX20S is the same pack layout as MP24, so I presume the limit is the same (12A)

Quote

EUC.SALE - Important information regarding new BEGODE PRODUCTS.
Due last safety updates to the all new BEGODE wheels like RS, EXN, TESLA, HERO, COMMANDER and etc.
NEW protection board was added to the every Battery pack.

Please note: 900wh, 100V battery pack have max charging current limit 3A.

Example
BEGODE RS have 2x900wh, total 1800wh. Max charging curent 6A!!!!
EXN - 9A
Monster PRO, Commander 12A.

Charging with more then rated, can damage Battery pack and will be not allow by battery protection system.

https://www.facebook.com/groups/ElectricUnicycle/posts/4440719026026026/

[thendless]

@RagingGrandpa Good info to add about current 100V Gotway BMS: https://www.facebook.com/groups/666662156765084?multi_permalinks=5514473368650581&hoisted_section_header_type=recently_seen

Edited October 11, 2022 by RagingGrandpa
(img url)

[RagingGrandpa]

Nice. They've brought in the updates that the new 134V packs use.

If it's still limited to 3A recharging, that's sad tho.

[mike_bike_kite]

Quick question: Does anyone know what the type of connectors are called on the 900Wh 21700 LT M80T packs. There are 3 leads - one to charge the packs, one to connect to the main board and a balance lead that goes between the packs. I'm primarily after the charge connectors name but any would be nice. The newer packs have the MT60 3 pin power connector but the older ones, like mine, don't. 

[RagingGrandpa]

1 hour ago, mike_bike_kite said:

Does anyone know what the type of connectors are called on the 900Wh 21700 packs.
The newer packs have the MT60 3 pin power connector but the older ones, like mine, don't. 

Main output: XT60 
Recharging input: XT30 
Charge-stop trigger: JST SM SMP04

And like you mentioned, the newest packs combine the first two connectors, into a 3-pin MT60, where two pins are the main output, and the 3rd pin is the recharging negative.
(B+ is shared in the EUC harness, so there is no need for a 4th pin, since output positive and recharging positive are the same circuit.)

On 4/26/2022 at 3:18 AM, EMA said:

• old gen pack:
  • xt60 for discharge
  • xt30 for charge
  • black 4 wires connector for BMS talking ( 2 black 2 red )
• new gen pack
  • mt60 connector, discharge plus the white 3rd cable
  • black 4 wires connector for BMS talking with only 1 black cable

You can retrofit "new" 900wh packs into "old" EUC's, because the EUC controller doesn't know or care about how the battery is connected. 
To retrofit, you do need the correct small wire harness to join the multiple packs. This pack harness will be XT60 for old-style, and MT60 for new-style.
 

Edited December 19, 2022 by RagingGrandpa

[mike_bike_kite]

@RagingGrandpaThanks for the really quick response!

[Mark Frey]

The begode RS and Nicola  that had the faulty packs have the tri connectors connecting the battery’s togeather and seems one smaller wire is in one of the tri and is a black and a small red also just is tapped into other large red  tri and both end up in a small double yellow terminal one going to motherboard it seems and another double small terminal is forked out off same black and red I assume must connect to battery pack (faulty battery pack 2100 cells were removed) since I bought a nic and rs with no batterys from alien rides. Anyway so what I want to do is be able to swap out my batteries from a commander into the nic and rs. And also be able to swap them back in. I could even make the Nicola a three battery version or a two it seems. 
  I know this sounds confusing but if one saw it they would know. I can figure out how to do a pic soon. Anyway wonder if would be easier to do a step down from the tri connectors or perhaps just cut and solder on new double to match commanders battery 

[Mark Frey]

There is also this other black four pin connector that must of been connected between the two bad battery backs .  It has two small black and two red it is seen taped there 

[Mark Frey]

This is commander style and black four pin has only one small red wire that connects both packs and also leads to motherboard or perhaps other side batteries. Small yellow double terminals connect both and also lead towards motherboard/other side batteries. Anyway was hoping to figure out best way to interchange the commanders packs with nic/rs style bogus faulty packs terminals. One thing odd is black terminal does not have black wire as is seen in rs/nic. It only has a single small red in that four pin terminal. Could someone five easy rundown on what each does again? I did read this thread and sort of have it down. The black terminal with the small red wire must be a trigger or cutoff. I need to study it all again. 

[Mark Frey]

Anyway yea ..what do you think would be the most simple way to integrate these two styles?  Why does nic/rs  style  black terminal have two red and two black  and commander has only a single red in the four pin terminal?  Am not sure how that black four pin connected in the rs since seems could of only been connected between the two batteries if pack from other side had long enough for wires to pass through over top. You know there is no other way to imagine it. I can’t know because I don’t have the faulty batteries to see what was like
 

[Mark Frey]

Not sure what year they began using the tri connectors but think it could be really new since judging from some videos only 7 months ago it seems the weel had old 2 plug connectors. It seems the black is a communication wire. Will keep studying it. It seems they did some upgrades but I still don’t see the beeper the way the wheels upgraded packs have. Not sure about exact date of the wheel since have never had a battery in it to check the manufacture date

[RagingGrandpa]

On 12/24/2022 at 2:10 PM, Mark Frey said:

rundown on what each does again?

This was succinct:

On 4/26/2022 at 3:18 AM, EMA said:

• old gen pack:
  • xt60 for discharge
  • xt30 for charge
  • black 4-pin connector for BMS talking ( 2 black 2 red )
• new gen pack
  • mt60 connector, discharge plus the white 3rd cable (charging negative)
  • black 4-pin connector for BMS talking ( with only 1 RED wire )

 

On 12/24/2022 at 2:10 PM, Mark Frey said:

One thing odd is black terminal does not have black wire as is seen in rs/nic. It only has a single small red in that four pin terminal.

That SMP04 is the charge-stop trigger link.
Because the packs share a reference voltage at their main output, one wire is sufficient to send this signal (a constant voltage). 
If you have a 4-pin SMP04 harness (like your MSuper) and install it between new-style packs, then 3 of the pins are not used. That's ok.

 

Edited January 4, 2023 by RagingGrandpa