diy solar

diy solar

Active balancer or not?

Thanks very much for the productive discussion everyone, particularly @Steve_S for posting the video I had not seen (though I watch a lot of the channel).

I have decided to include the balancers now. If anyone is interested, I'll keep you updated.
I've noted that I've unintentionally copied one of the Big Wei batteries! I wonder if my manufacturer is also theirs and whether that's why they came out pretty much the same?
 
I know this is an old thread, but the topic was valid for what I wanted to post.

I've been a fan of active balancing. I've tested charge and discharge cycles with both the QNBBM and Heltec (capacitor-based) active balancer, and seen very good results. However, after watching @Off-Grid-Garage YouTube videos and watching his opinion evolve, I think my position has changed just a bit.

It seems to me if you have a situation where the system is cycling every day and perhaps cycling fairly deeply, an active balancer that balances all the time (QNBBM and Heltec) may actually cause cells to drift out of balance as @Off-Grid-Garage saw. The cells are spending enough time closer to the bottom knee that the balancers are trying to balance there, which can undue a top balance. Then when the cells rise up closer to the upper knee the balancers are working to get them back in (top) balance again. This can go on forever, and over time I'm guessing it isn't good.

On the other hand, if you have a system that spends a bit more of its time close to fully charged or in float, a good active balancer can successfully preserve the top balance. This is the case for our cabin system, which is normally only in use when someone is visiting the cabin: 3-4 days per week during the May through November time frame. The rest of the year the cabin is not used. This means that during the days when no one is using the cabin the active balancer can get the cells back to a good top balance.

I currently have one 8S 24V system and two 4S 12V systems, all using the Heltec active balancer. The two 12V systems will probably be cycled pretty heavily for 2-3 months every year, and periodically the rest of the year.

So now I'm thinking I want to leave the active balancer on the 24V system for our cabin, for the reasons stated above. For the two 12V systems, I want to be able to use the active balancer selectively. Not just at certain points in the charge cycle as @Off-Grid-Garage argues, but rather to enable it when the cycling is either not happening or is shallow.

To this end, I'm thinking of putting in a switch or relay control to turn the Heltec active balancing ON or OFF. The solder pads that @curiouscarbon pointed out above seem to be made to order for this.
See that spot where it says "RUN"?

That's actually two separate electrical connections. You can simply remove the blob of solder there and wire in a physical switch instead. This will allow you to control the active balancer manually.
So has anyone actually used these pads to turn the balancing on and off?
 
have not yet tried wiring a switch to the Run pads/pins, still on todo list.

this latching relay is probably what i’ll solder to the run pins to allow program controlled activation


i will solder one pad of run to NO pin and other pad of run to COM pin. that way it will be off by default and when the latching relay is set/activated, the active balancer ought to, in theory, be selectively engaged
 
thank you for the heads up! latching relays are a bit different to interface with than non latching ones as you mentioned

using one of the latching ones right now to switch on/off a victron dc/dc converter; the relay breakout board is supplied with four conductor connections: ground, 3.3v, SET, UNSET.

the SET and UNSET pins actually control MOSFET to switch the 3.3V to the relay, not sourcing power from SET or UNSET pins themselves. so ground and 3.3v are needed as you say✅

the wires come from the microcontroller, which uses 3.3v as its supply voltage level, so i am supplying it from that.

i use a pololu step down dc regulator to generate the 3.3v and the regulator supports 500 milliamps and input of up to 50V
 
at idle not doing anything it looks like this
1634933480304.jpeg
when SET it flashes green
1634933498602.jpeg
when UNSET it flashes red
1634933516133.jpeg

the microcontroller must pulse the desired pin HIGH (3.3V) for at least 10 milliseconds to activate either SET or UNSET. arduino warning
 
That board seems to be intended to use with one of Adafruit's feather boards, so it probably expects a 3V power source. Doesn't mean you can't use it, but it probably isn't how it was intended.
Totally agree to double check; there are lots of other pins that confused me at first, I went into their tutorial article and found the schematic for how the board is wired and concluded that the four connections needed for it to properly function are only ground 3.3volt, set and unset. and in practice it operates as expected?it makes a pretty satisfying little “tiK” sound when it latches on ?
 
BTW, on Heltec Active Balancers:

View attachment 56600
See that spot where it says "RUN"?

That's actually two separate electrical connections. You can simply remove the blob of solder there and wire in a physical switch instead. This will allow you to control the active balancer manually.

Just thought I'd mention this, to soothe any worries that it might go haywire....

As always, ensure that any balancer used is properly configured for the battery chemistry you're using and for the specific safety range you desire.
Reviving this thread and this post. I got around to trying this out, and I seem to have failed miserably.

I unplugged my 8s Heltec balancer, then removed the blob of solder between the two pads where the RUN label is. Then I wired up a switch across the two pads. I checked the continuity with the switch both on and off, and it seemed to be connected correctly.

I believe the switch was off when I reconnected the balance lead plug to the board. The little "Run" LED flashed on, but then went off again. I turned the switch ON and OFF and back again, and the Run LED seemed to flash a couple more times and then quit. Thinking something was wrong I removed the switch and replaced the blob of solder across the two pads, and tried again. The LED does not come on at all.

I am pretty certain the Run LED was on all the time previously.

My pack is pretty balanced at the moment, so I'll have to wait until my next discharge or charge cycle to verify if I broke it or not.

Has anyone actually done this mod to the Heltec board? I know it's been talked about, but I don't recall seeing anyone post that they had done it.
 
One unclear aspect to me is:

The “RUN” pads/connector might only Disallow Balancing, yet not Engage Balancing.

Like the ”RUN” switch is red light green light, but the balancer is still looking for cross traffic (voltage mismatch) before going through the intersection (engaging balancing). Just guessing here.

With RUN enabled, I would assume Cell Voltage Mismatch is still Required to Engage Balancing.

A crazy/inadvisable idea that passed through my head on this topic was to first close/activate the run switch and then somehow perturb the balancer’s view of at least one cell voltage. With the idea being to kickstart the balancing with a “fake imbalance”. No idea how to safely affect the perceived imbalance

There’s a cell voltage difference threshold for Begin and End, so perhaps the cells are all close enough in voltage that the balancer considers its job done, waiting for more millivolts imbalance before starting.

Thank you for the update!
 
Has anyone actually done this mod to the Heltec board? I know it's been talked about, but I don't recall seeing anyone post that they had done it.
If I do this with the 4S heltec capacitive active balancer board that is on hand, I’ll be sure to share a bit about how it behaves.

cheers, thanks again for the info about how your board+battery behaves. my gut senses that it might be “working properly” and that it may begin balancing if a cell goes out of line.
 
There’s a cell voltage difference threshold for Begin and End, so perhaps the cells are all close enough in voltage that the balancer considers its job done, waiting for more millivolts imbalance before starting.
I had thought of that, but I'm pretty sure the little LED is on all the time, or at least I think it was on my 8S pack. I've torn down the 4S packs I was working on, and since those are for a friend I'm not keen to wreck his balance boards.
 
I was going to add a voltage sense relay to control the active balancer..... decided it was a solution waiting for a problem.
 
I was going to add a voltage sense relay to control the active balancer..... decided it was a solution waiting for a problem.
This is what I have on my 2 packs and works fine. At 27V the relay is triggered and balancing starts. It stays on until pack voltage drops to 26.8V. Analog circuit so fiddly to set the trim pots. I do have an Arduino connected and with hindsight this would be simpler to control the relay.
 
This is what I have on my 2 packs and works fine. At 27V the relay is triggered and balancing starts. It stays on until pack voltage drops to 26.8V. Analog circuit so fiddly to set the trim pots. I do have an Arduino connected and with hindsight this would be simpler to control the relay.
I assume the LED is ON all the time that the balancing is going, right?

Can anyone verify if the LED stays off when there is no balancing to do?
 
I have two capasitive and 3 inductive Heltecs. The lights go out once balanced and will not come back on until a .1 delta is reached.
Awesome! So maybe mine is still OK. I'll put the switch back on and leave it on, and I should see some imbalance as I get to the upper curve on charging.

I could have sworn it was on all the time, but I guess not. Thanks @grizzzman for the info!
 
Back
Top