How BMS Works (Gotway pack teardown)

[RagingGrandpa]

1 hour ago, Cody said:

better is to attach Small schottky Diodes to prevents cell reversal polarity, during deep discharge

It's an interesting point.

It would increase the cost of the EUC by $30 or more, to backup something the BMS normally prevents. Respectfully- If I were choosing where to spend the money in an EUC, I'd put the $30 into thicker copper for the battery output harness, or better o-ring seals for water resistance, before spending it on something that offers no benefit during normal use.

p.s. props for the current-limiting incandescent bulb, those do come in handy!

Edited June 10, 2020 by RagingGrandpa

[WI_Hedgehog]

2 hours ago, RagingGrandpa said:

It would increase the cost of the EUC by $30 or more, to backup something the BMS normally prevents.

Do you mean the cost of a Smart BMS?
I thought Gotway, and most other EUC manufacturers, don't monitor cells during discharge, only charge. If that's the case @Cody makes a good point.

Aren't schottkey diodes $0.28 each out of China, less in bulk?

[Cody]

 

schottkey is an really Cheap upgrade, and really userfull when you lose an cell (ageing batteries pack, one CID open, so the whole pack still deliver power safely by limiting undervoltage to less than 0,5v, witch cause "NO" damage for hours of play with (but if you go over -2.1*v damage occur ...)
Nasa revieuve uploading here

...

Chineese great test :

voila voila finaly found the Better studdies EVER on lithium, by chinisian coronated GREAT brain ;3
Ill love thoos scientits that GO MUCH Below standards tests !
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4957210/pdf/srep30248.pdf

 

  

Edited June 12, 2020 by Cody
opost

[RagingGrandpa]

On 6/10/2020 at 12:50 PM, WI_Hedgehog said:

Aren't schottkey diodes $0.28 each out of China, less in bulk?

Right, so 1 diode per parallel group per pack, 2 or 3 packs per EUC, 24S, so $21 of diodes plus NRE...

(Please I'd like my $30 spent on copper instead)

Edited June 15, 2020 by RagingGrandpa

[Cody]

USA:     0,135 €
China:     0.0055 €

SS34, less than 2.43$ per battery pack.

and this stuff is absolutely need as an basic equipment in a pack, because
cell safety Current Interrupt Device can't hold more than 3 cell, and here we are talking about an 18cell series packs... only 18 rectifier to make is much more safe.

[Chronic]

The PCM protection circuit used in the second half of 2019 has a reverse current diode

But it's still a low-cost circuit. This GW design style and method was quite shocking to me lol

Edited June 23, 2020 by Chronic

[xorbe]

Holy sh*t you couldn't have paid me to do that wire short test!

Edited June 23, 2020 by xorbe

[RagingGrandpa]

On 6/23/2020 at 11:47 AM, xorbe said:

Holy sh*t you couldn't have paid me to do that wire short test!

I finally worked up the courage to do this test... not with a boxcutter, but with a 12A 250V fuse and a toggle switch. My 2019-built 84V MSX is protected! (The fuse did not blow, and the wheel continued balancing.)

Also FYI, Veteran has a different approach to BMS:

• The 24S pack is split into two 12S modules, with each managing its own top-balancing.
• The two modules share a charge-stop trigger signal, which is ultimately sent back up to the ESC.
• Charging interruption is performed at the ESC. The ESC sends charging current directly to the pack's high-current output.
• Pack overcurrent is protected with an ANL-style melting fuse. (Note: these are not rated for 100VDC, and will likely cause excessive arcing when exposed to a bolted short.)

Edited January 14 by RagingGrandpa
(img url)

[Planemo]

23 minutes ago, RagingGrandpa said:

 

• The two modules share a charge-stop trigger signal, which is ultimately sent back up to the ESC.

 

...but then back down to the main batt output? So the pack will still be receiving charge current when full, just not via the BMS and it's individual strings?

[NuRxG]

A lot of really good info, thanks!

[HamishTHaggis]

On 6/9/2020 at 11:54 AM, RagingGrandpa said:

Yes, you can easily probe them all per below.

The spacing is tight- take care not to short them! Especially when cutting the shrink!
If soldering, tapemask one at a time... risky business with cells connected.
Soldering will be difficult- most boards have a conformal lacquer painted over this area, for corrosion resistance.

The only part that's missing is B-, which is easy to probe manually at the pack output, but challenging to find a solder spot that's small and safe... if it's this exact board, I suggest the little unused pad near 'R005'.

 

I don't see a need for inspecting per-cell voltages frequently. Maybe once a year, or whenever you have the shell open, or whenever you notice an abnormality during charging... just bringing all 25 pins to an internal connector would be awesome.

(As before, take care with insulation protection for this new pigtail. It must be well-secured inside the wheel in a way that cannot chafe or wear. Smaller wire is better- they'll melt at a lower current if they short out... but you'll likely burn the balancing traces off the BMS too, destroying it.)

I'm currently rebuilding a 100-volt MSX, and while I've built a number of battery packs, I've not built one on these BMSs before.  Some of the wiring is missing so I'm recreating it. As such, would it be safe to connect the charger port directly to both BMS's in parallel, so each has its own direct power supply? Obviously would want to maintain the charge stop linkage as a backup, but would there be any issue with direct links for charging to each board?

[Arbolest]

4 hours ago, HamishTHaggis said:

I'm currently rebuilding a 100-volt MSX, and while I've built a number of battery packs, I've not built one on these BMSs before.  Some of the wiring is missing so I'm recreating it. As such, would it be safe to connect the charger port directly to both BMS's in parallel, so each has its own direct power supply? Obviously would want to maintain the charge stop linkage as a backup, but would there be any issue with direct links for charging to each board?

Absolutely! That's how all of Gotway's wheels are connected anyway! All of the battery packs are independent and wired in parallel.

[HamishTHaggis]

8 hours ago, Arbolest said:

Absolutely! That's how all of Gotway's wheels are connected anyway! All of the battery packs are independent and wired in parallel.

Let me clarify slightly - I realize the packs are paralleled going to the board, but on the basis of @RagingGrandpa explanation (and the wiring I have in the wheel) I think the charge port only connects directly to one of the packs, the other charging through the parallel main connection.  I'm asking if there's any reason I wouldn't connect the charge port to both packs directly, since I have to rebuild that wiring anyway.

[Planemo]

Don't know if it helps, but my MSX has two pairs of wires coming off the charge port (so 4 pins are used on the charge plug). Each pair goes to one pack on each side of the wheel. This is how it came from the factory.

So no, one of my packs doesn't charge via the parallel main connection.

[HamishTHaggis]

2 minutes ago, Planemo said:

Don't know if it helps, but my MSX has two pairs of wires coming off the charge port (so 4 pins are used on the charge plug). Each pair goes to one pack on each side of the wheel. This is how it came from the factory.

So no, one of my packs doesn't charge via the parallel main connection.

Really?  This one and the one i opened to compare had 4 pins connected, but they all go to one XT30 (both + twisted together, both - as well) which is connected to the main pack.  So even if you connect two chargers/high amperage charger on 4 pins it would charge one of the backs back through the main power connection.

There's so much variation between batches on these wheels I shouldn't be surprised I guess.

[Planemo]

Yep, I deffo had 2x XT30 plugs, one for each pack. It seems odd that you don't I must admit. Each of mine sit at the top of the shell, in the cavity above the packs.

[RagingGrandpa]

42 minutes ago, HamishTHaggis said:

on the basis of @RagingGrandpa explanation (and the wiring I have in the wheel) I think the charge port only connects directly to one of the packs, the other charging through the parallel main connection

Good distinction.

My original post here was regarding a rather old Gotway from 2015, which had 3 packs (each with its own BMS PCB), and only one of them connected to the charging input.

Since then (all Gotway's after 2018), Gotway changed to connecting each pack to the charging input.

In both the old and current Gotway's, it's important to note that all of the packs share the 'charge stop trigger link' wires (the small 4-pin connector in the pack harness), so that cell UV/OV in any one pack will interrupt the charging input on all of them.

@HamishTHaggis as long as you retain the trigger link wires between your two 100V MSX packs, I think it is an equivalent choice. You could connect one pack, or both, to the recharging port.

Please don't recharge your 1200wh wheel at a rate higher than 5A. And, from personal experience, I think a "5A charge port connected to only one pack" is reliable.

[Arbolest]

7 hours ago, HamishTHaggis said:

Let me clarify slightly - I realize the packs are paralleled going to the board, but on the basis of @RagingGrandpa explanation (and the wiring I have in the wheel) I think the charge port only connects directly to one of the packs, the other charging through the parallel main connection.  I'm asking if there's any reason I wouldn't connect the charge port to both packs directly, since I have to rebuild that wiring anyway.

None at all. Each battery pack is made to operate independently of any other packs, so each one has viable output, charging, and "charge stop" leads that can all be wired in parallel with the same leads from any number of other packs.

While it may be possible to only connect a single pack to the charging input wires and allow the other packs to charge through the output wires, my OCD demands that I have every pack wired up the same, sharing the same discreet cabling for Charging, Output, and "charge stop".

[HamishTHaggis]

12 hours ago, RagingGrandpa said:

Good distinction.

My original post here was regarding a rather old Gotway from 2015, which had 3 packs (each with its own BMS PCB), and only one of them connected to the charging input.

Since then (all Gotway's after 2018), Gotway changed to connecting each pack to the charging input.

In both the old and current Gotway's, it's important to note that all of the packs share the 'charge stop trigger link' wires (the small 4-pin connector in the pack harness), so that cell UV/OV in any one pack will interrupt the charging input on all of them.

@HamishTHaggis as long as you retain the trigger link wires between your two 100V MSX packs, I think it is an equivalent choice. You could connect one pack, or both, to the recharging port.

Please don't recharge your 1200wh wheel at a rate higher than 5A. And, from personal experience, I think a "5A charge port connected to only one pack" is reliable.

 

5 hours ago, Arbolest said:

None at all. Each battery pack is made to operate independently of any other packs, so each one has viable output, charging, and "charge stop" leads that can all be wired in parallel with the same leads from any number of other packs.

While it may be possible to only connect a single pack to the charging input wires and allow the other packs to charge through the output wires, my OCD demands that I have every pack wired up the same, sharing the same discreet cabling for Charging, Output, and "charge stop".

Thanks to you both for the replies and the details.  This wheel belongs to a friend, and I don't know the precise age, but sounds like it may just barely predate the change in wiring as it certainly has only one connector.  That said, it's not going to be an original 1200WH pack - it's actually going to be a 3000WH pack in a 100V msx body.  24s6p of Molicel M50a.  There is still not intention of charging above 5A though, and I certainly wouldn't given the wire gauge in use.

[HamishTHaggis]

 

On 1/16/2021 at 11:57 AM, RagingGrandpa said:

Good distinction.

My original post here was regarding a rather old Gotway from 2015, which had 3 packs (each with its own BMS PCB), and only one of them connected to the charging input.

Since then (all Gotway's after 2018), Gotway changed to connecting each pack to the charging input.

In both the old and current Gotway's, it's important to note that all of the packs share the 'charge stop trigger link' wires (the small 4-pin connector in the pack harness), so that cell UV/OV in any one pack will interrupt the charging input on all of them.

@HamishTHaggis as long as you retain the trigger link wires between your two 100V MSX packs, I think it is an equivalent choice. You could connect one pack, or both, to the recharging port.

Please don't recharge your 1200wh wheel at a rate higher than 5A. And, from personal experience, I think a "5A charge port connected to only one pack" is reliable.

@RagingGrandpa  Can you clarify about the trigger links?  How many need to be connected between packs, if there's a polarity, etc?  I don't have the originals for reference and have seen 3 or 4 different configurations on different packs.  That's about the last remaining detail for finishing this rebuild, any help obviously appreciated.

[RagingGrandpa]

I thought this was thought-provoking:

On 2/18/2021 at 10:40 AM, Planemo said:

The bigger the imbalance, the harder it is to get rid of it, to the point where in some cases it can't even be done anymore by the factory charger, no matter how long you leave it plugged in.

Some examples to consider, shown below.

The point to drive home is- cases A, B ,C, D will balance successfully!
Case C will take the longest to balance, because it has the greatest difference in voltage to overcome. Assuming an 800wh 20s3p Gotway pack, because the balancing current is only 71mA, Case C could take a whopping 127 hours to balance.

And as Planemo mentioned, worse imbalance such as cases E and F will not be successful, regardless of how long the charger is connected, because they will hit the overvoltage limit and disable the charging input before the pack can reach 84V. Repeated attempts to charge will not help.

I think E can happen due to self-discharge of cells; and F can happen from 'aged' cells that have lost their original energy capacity.

p.s. You *could* recover cases E and F without opening the pack: After the overvoltage trip occurs, connect a CV-CC supply with a very low current limit like 70mA, to match the dissipation current of the balance resistors. And wait for days.
(But the large imbalance is likely a symptom of an underlying problem with the pack, e.g. cell strap weld failure. Teardown and inspection is recommended.)
 

Edited January 14 by RagingGrandpa
(img url)

[NuRxG]

On 2/18/2021 at 11:38 AM, RagingGrandpa said:

Repeated attempts to charge will not help.

What if you discharge between them?

[RagingGrandpa]

@NuRxG - it won't help

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

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.

(Of course, if the shrink is removed, someone familiar with electronics could balance a pack manually...)

[Planemo]

1 hour ago, RagingGrandpa said:

(Of course, if the shrink is removed, someone familiar with electronics could balance a pack manually...)

Yep, done this myself before, using an RC charger. It's the only way to get things back in spec sometimes as the factory chargers/BMS just cant and wont cope with an imbalance that's gone a little too high, and no amount of charging/discharging will sort it.

[NuRxG]

13 hours ago, RagingGrandpa said:

@NuRxG - it won't help

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

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.

(Of course, if the shrink is removed, someone familiar with electronics could balance a pack manually...)

Still trying to wrap my head around this. The pack never hits 84v that means there will be a larger current than the bleed resistors can drain and the cells will go to 4.25 but there will be some bleed time from 4.2 to 4.25, and every time you discharge and recharge you'll gain a few moments of balancing and get closer and closer to a good spot? This is pretty hypothetical tho since batteries only have a few hundred cycles and were talking hundreds of hours of balancing time lol! Just trying to make sure I understand this situation right.