QNBBM balancers

That is the same as the 1s balancers you can find. Its just 7 of them on same board... They only do balancing and nothing else. No protection or reporting back. If you look further down you see images of the 1s.

Expensive system if you are after a BMS. Good system if you actually need active balancing due to uneven capacities in your packs that you cannot fix.
 
I use the 4s version of these balancers on a 4s lifepo4 110ah battery. They work pretty good, they balance all the time they are connected they transfer any high cell voltage to the low cells. Comes in handy when you are charging at higher amps. On my system I just connected a 4s balance connector to these balancers and plug them ino a 4s connector on the battery thats all you need. There is no on/off switch, no programming. Works on lifepo4 or li-ion.

On my 4s battery I had to use a balance charger to get a full charge becauseone cell always reads higher than the other 3, this always causes the bms to shutoff charging early. With these balancers, I can nowget a full charge everytime using any 12 volt charger.

image_ajqgwf.jpg
 
daromer said:
That is the same as the 1s balancers you can find. Its just 7 of them on same board... They only do balancing and nothing else. No protection or reporting back. If you look further down you see images of the 1s.

How does this systemwork?

They share a common (DC, AC?)balancing bus and every module is trying to bring its own battery to same voltage as the balancing bus, ie if balancing bus is higher voltage than cell connected to modulecharges the cell from balancing bus and vice versa? Just a guess, I am close?
 
Correct. Common bus. take of give kind of so nothing special. Bi-directional DC.
 
Thank you for the info! Ive been looking at these all week. They seem so simple yet soooo expensive when you're working on a 13s-17s battery pack like I am! lol. Im definitely wanting to stick to an active balancing set up for my project. Has anyone found a less expensive, active balancing system?
 
Bubba said:

Looks like that circuit would only work with a 12V system, the MP2307 chip has a max of 23V....

The QNBBM balancers are probably best suited to large battery packs where the balancing current from a passive BMS can't keep up.

Like Daromer pointed out they only balance, but they do it well.... you need other gear to monitor & control.
 
Redpacket said:
Bubba said:

Looks like that circuit would only work with a 12V system, the MP2307 chip has a max of 23V....

The QNBBM balancers are probably best suited to large battery packs where the balancing current from a passive BMS can't keep up.

Like Daromer pointed out they only balance, but they do it well.... you need other gear to monitor & control.
Easy balancer doesn't care about the overall system voltage. They get their power from the adjacent couple of cells not from the total system.
Similar to the QNBBM Balancers. They may actually be very similar.

So the 23V doesn't matter.
 
Bubba said:
Redpacket said:
Looks like that circuit would only work with a 12V system, the MP2307 chip has a max of 23V....

The QNBBM balancers are probably best suited to large battery packs where the balancing current from a passive BMS can't keep up.

Like Daromer pointed out they only balance, but they do it well.... you need other gear to monitor & control.
Easy balancer doesn't care about the overall system voltage. They get their power from the adjacent couple of cells not from the total system.
Similar to the QNBBM Balancers. They may actually be very similar.

So the 23V doesn't matter.
Unfortunately the balancer here https://www.afdhalatifftan.com/2017/06/l...ancer.html seems to have a few issues:
- it can't do cells in a series pack, it can only do cells in parallel because of the non-isolated "Vcom"
- the chip is a 5V input "buck" step down so it's not bi-directional - it can only push power from a full cell back to the Vcom rail
- where does the energy pushed into Vcom go? There's no mechanism to get in into other cells?
- the chip is being used out of spec ie even a full 4.2V cell is well under the chips data sheet min 4.75V operating voltage.

AFAIK the QNBBM balancers
- are full bi-directional
- have two halves: the low voltage side powered by the one cell it's connected to and the "common rail" side connected between all the balance modules. The couple of adjacent cells don't matter other than they are also on the "common rail".
- the "common rail" is isolated from the cells.
- they can definitely be connected to series connected cells in packs because the common rail side is isolated.
 
Has anyone taken one apart and reverse engineered it? Would be interesting to try and make a DIY version or at least understand the mechanism to use in other projects.
 
Yeah seems there is a common ground, see the diagram here:
http://www.afdhalatifftan.com/2017/06/low-cost-active-battery-balancer.html
"vcom" is referenced to the cell's ground/0V
There's no separate common bus/rail system.

rev0 said:
Has anyone taken one apart and reverse engineered it? Would be interesting to try and make a DIY version or at least understand the mechanism to use in other projects.

daromer did a video where talks about how they work & then cut/ground one open - they're potted in epoxy or similar inside there.
http://diytechandrepairs.nu/1s-modular-active-balancer-reviewed-and-trashed/

They seem to work by making a common rail with about 10x the battery voltage as a ~3.5kHz square wave. Low cells rectify this & draw energy from the common rail, high cells are making the highest voltage square wave so they are the ones getting power pulled from them. There's going to be some smarts in the circuits to tell which way to work, etc.
 
Biderctional DC/DC. If you google that you will find ALOT of information and schematics so its nothing new at all.

There is no smart function in it at all! (In terms of logic) Since its biderectional it will either take or give energy to the lowest point. Either its the bus or the cell itself. This also determines the current transfered since its based on the voltage differential between source and destination: So if the diff is only 100mV between cells you have not much transfer of current. Perhaps at most 1A and at 50mV diff you got like 200mA of diff.

The common rail was not 10x it was same as the battery I found out later on. i had the scope on wrong settings :D
 
Redpacket said:
Yeah seems there is a common ground, see the diagram here:
http://www.afdhalatifftan.com/2017/06/low-cost-active-battery-balancer.html
"vcom" is referenced to the cell's ground/0V
There's no separate common bus/rail system.

rev0 said:
Has anyone taken one apart and reverse engineered it? Would be interesting to try and make a DIY version or at least understand the mechanism to use in other projects.

daromer did a video where talks about how they work & then cut/ground one open - they're potted in epoxy or similar inside there.
http://diytechandrepairs.nu/1s-modular-active-balancer-reviewed-and-trashed/

They seem to work by making a common rail with about 10x the battery voltage as a ~3.5kHz square wave. Low cells rectify this & draw energy from the common rail, high cells are making the highest voltage square wave so they are the ones getting power pulled from them. There's going to be some smarts in the circuits to tell which way to work, etc.
GND is relative and is shown in the balance.png as -VBATT.
The 3 connections are +VBATT, VCOMM, -VBATT There is no common ground that you attach all the modules together.
The MP2307 has internal Zenor and mosfet on the ground.

Not seeing why this would not work.
 
OK so I was thinking like this: say we have 3 equal cells in series C1 = 4.0V, C2 = 4.0V, C3 = 4.0V,
They are connected (0V)__-C1+__ -C2+__-C3+__(pack+ve)
and we have 3 of afdhalatifftan's modules, with each -VBAT & +VBATT connected across each cell. I'm assuming the 3x VCOMMs are connected together.
What would vcomm be? eg what would each separate module see vcomm to be?
Say we let vcomm be 1/2 the pack voltage (11.3/2) = ie 6V from pack 0V
Then C1's module would see a big positive feedback, C2's module nearly neutral, C3 a big negative feedback???
I don't see how it can balance like that?
 
Ah, I get it now. Each balancer is connected a 2s group of cells. The buck converter sets the midpoint voltage to be exactly half of the 2s group voltage, thus each 2s group is balanced. You need N-1 balancers for an Ns pack. Here's his comment on the website which explains:

Yes, you are very close, the following is an example for a 4S pack:

Balancer1: Cell1 and Cell2,
Balancer2: Cell2 and Cell3
Balancer3: Cell3 and Cell4

Let me know if this is still confusing.

Thank you for the kind words!

I didn't see a response to how this balances if the topmost cell is low, however, since it's a buck topology I think it may not be able to, but a synchronous buck is technically bi-directional, so I may have to build this up and play around with it.
 
His explanation is how we'd like it to work for sure.
I re-read his article & I don't know he actually live tested it with a series of cells?
Looks like he was about to send it to a couple of others to test but didn't seem to go any further?

Even if with 2s, 4s or other series cells (anything other than 1s cells parallel), it doesn't seem to solve how the resistors in the circuit compare the vcomm volts to each cell. See R3 to R8, they are all 1k. So that assumes the vcomm voltage is equal (or close) to 1/2 the cell V.

vcomm would have to be higher than all the cells, eg higher than pack +ve to put power into the top cell but issue above comes first!
But his diagrams show protection diodes (see bottom left) limiting vcomm to cell 0V to cell +ve so that rules that out.

Would love to see it work but can't see how it could right now?
 
Back
Top