Heltec BMS' (up to 350A) with Active Balancing & Independent Active Balancers

[fafrd]

Thanks. I’m in contact with Heltec and will see what they have to say to me.

My biggest current issue is that I cannot find any spec for maximum charge current in their products - is it assumed to be the same as maximum discharge current (for single-port offerings)?

 

[fafrd]

They do offer an 8S LiFePO4 BMS with Active Balancing: https://www.aliexpress.com/item/1005001465861214.html?spm=2114.12010615.8148356.2.22591691BDwTNn

I’m really liking the way this is looking with solar holes replaced with a copper busbar and lug and may pull the trigger on the 240A or 300A model once I confirm maximum charge spec (are charge and discharge limits understood to be the same for single-port designs?).

 

[fafrd]

‘Solder holes’, not ‘Solar holes’...

 

[Dzl]

Looks to me the specs do explicitly show charge current (and it is the same as the discharge current limits) I think you are still on the right track confirming with the manufacturer though. Bear in mind, some of the Heltec BMSes are configurable/'smart' some are preset, and some are passive and some are active balancing, I haven't looked hard at this one, but if I were in your shoes I would verify these things with the seller just to be sure (it does look like this is an active balancing model from the title and the datasheet, but on Alibaba it always pays to double check)

 

[ArthurEld]

I didn't realize Heltec has another BMS with active balancing. But like Dzl said, that one isn't a smart BMS so no bluetooth and not user configurable.
You can order it with all of the options permanently set to your requests. That is probably ok for some people.
I bought the smart Heltec 200A BMS with 2A active balance. I like it so far but I'm only working with a 12A 48V power supply now and I don't plan to put over 100A through the BMS.

 

[fafrd]

Looks to me the specs do explicitly show charge current (and it is the same as the discharge current limits) I think you are still on the right track confirming with the manufacturer though. Bear in mind, some of the Heltec BMSes are configurable/'smart' some are preset, and some are passive and some are active balancing, I haven't looked hard at this one, but if I were in your shoes I would verify these things with the seller just to be sure (it does look like this is an active balancing model from the title and the datasheet, but on Alibaba it always pays to double check)

View attachment 28048

Thanks. Yeah, I finally found a way to read that tiny font on my phone and see that charge limits and discharge limits are identical (is that true for all single-port BMS?).

Since they have a Black Friday sale coming up in a week, I’ll wait until then to place my order.

I think Active Balancing is worth some premium, but want to understand how much delta it is costing and convince myself it is worth it (one more thing to break, among other costs).

Also, the products they have sent me have 1.5A of Active Balance current but it sounds as though others have a variant supporting 2A of Active Balance current.

And then there is the ‘Smart’ aspect - are these budget BMSs user-configurable in any way or fixed?

But after covering the basics, vendor reliability and product quality trump all these bells & whistles (at least for me).

The main reason I came to the thread was to seek feedback on Heltec as a supplier, and the more I read the more comfortable I’m getting...

 

[Dzl]

I think Active Balancing is worth some premium, but want to understand how much delta it is costing and convince myself it is worth it (one more thing to break, among other costs).

I think that is very dependent on your cells. With a well matched and balanced pack, active balancing would not be necessary. But most of the cells people buy here from the grey market are at best loosely matched. The grey market 280 EVE cells in particular seem like good candidates for active balancing since they are large capacity and not closely matched. Of course this is just a generalization and once you get and test your cells you can cycle them a few times, and judge for yourself how closely matched your particular cells are to each other.

Also, the products they have sent me have 1.5A of Active Balance current but it sounds as though others have a variant supporting 2A of Active Balance current.

I think they have active balancing models from 0.6A to 3A

And then there is the ‘Smart’ aspect - are these budget BMSs user-configurable in any way or fixed?

Some are, some aren't. Not sure about this one, From my earlier brief glance at the specs, I didn't get the feeling that it was but I only took a cursory peak.
I believe the one others (like Upnorthandpersonal) purchased with the black case is a smart BMS (blueooth and user configurable) some of the others are not.

The main reason I came to the thread was to seek feedback on Heltec as a supplier, and the more I read the more comfortable I’m getting...

I think its still early days with in our relationship/familiarity with this manufacturer. I haven't heard anything particularly negative, but there are only a handful of people with experience so far. I think @upnorthandpersonal may have the longest experience with it.

 

[juanmijm]

I didn't realize Heltec has another BMS with active balancing. But like Dzl said, that one isn't a smart BMS so no bluetooth and not user configurable.
You can order it with all of the options permanently set to your requests. That is probably ok for some people.
I bought the smart Heltec 200A BMS with 2A active balance. I like it so far but I'm only working with a 12A 48V power supply now and I don't plan to put over 100A through the BMS.

What exactly is that bms? Can you give me a link? Because I mess with the products on the heltec website, they are not well defined

 

[PeterBC]

I too am interested in their active balancing, especially for a potential 8S x 3P x 280AH arrangement ... I'm concerned that 80ma might be light on for balancing 3 x 280AH cells in Parallel .. But I'm leaning to 3P x 8S arrangement with 3 BMS's to share load.
The Heltec website sucks.. the way they arrange their products and limited info online worries me. The 300A unit only supports 50A charging ?
and likewise down for lower current units. Perhaps its better to just add their active balancers in parallel with a BMS with up to 5A(?) support for 8S in a simple unit ? I must say my head is spinning from reading about all these and some negative comments about DALY and other cheap Chinese units. Often its not the cheap that is the issue , but communication, documentation and lack of consistency that worry most.

For ref : https://heltec-bms.com/project/8s-b...-300a-350a-3-2v-lifepo4-same-port-split-port/
says a 300a unit has 50A max charging current. The 350A spec has nothing stated.

But the AliExpress site is a bit better but I couldnt find this one from the entry level of that site !

62.59US $ |3s 4s 7s 8s 12v 24v Bms 1.5a Active Equalizer Balancer Continuous 150a 240a 300a 360a Lifepo4/ Lithium Battery Protection Board - Battery Accessories & Charger Accessories - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

And of course states the charging current is same as discharge limits. Which means 3 x 150A units would be great for my application.

 

[ArthurEld]

What exactly is that bms? Can you give me a link? Because I mess with the products on the heltec website, they are not well defined

I was talking about the one fafrd posted. see below

They do offer an 8S LiFePO4 BMS with Active Balancing: https://www.aliexpress.com/item/1005001465861214.html?spm=2114.12010615.8148356.2.22591691BDwTNn

I’m really liking the way this is looking with solar holes replaced with a copper busbar and lug and may pull the trigger on the 240A or 300A model once I confirm maximum charge spec (are charge and discharge limits understood to be the same for single-port designs?).

 

[ArthurEld]

Another big problem with some of the non smart BMSs is that you can't monitor anything with it. They might do what they are supposed to do but you can't see what is happening unless there is a way to connect a display device.

 

[upnorthandpersonal]

Just for reference, this is the one I have:

95.99US $ |Smart Bms 1a/2a Active Balance Battery Protection Board 14s ~ 24s 100a 150a 200a 300a Phone App Lifepo4 Li-ion Lto 16s 20s - Battery Accessories & Charger Accessories - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

It does what it is supposed to do, can actually handle the rated current (I'm using two of the the 100A ones for now, I'll put a bigger one sometime next year), and it's configurable the way I want it with proper low temp cut-off, low/high cell voltage, etc.

 

[fafrd]

I didn't realize Heltec has another BMS with active balancing. But like Dzl said, that one isn't a smart BMS so no bluetooth and not user configurable.
You can order it with all of the options permanently set to your requests. That is probably ok for some people.
I bought the smart Heltec 200A BMS with 2A active balance. I like it so far but I'm only working with a 12A 48V power supply now and I don't plan to put over 100A through the BMS.

My current thinking is that while Bluetooth monitoring would be nice, it’s just one more chip that can fail, and I can wire in a parallel monitoring port for a BattGO monitor for lower incremental cost (and better fault-tolerance).

User-configurability is more interesting but sounds like a headache and I’m not convinced it’s worth the trouble (for me).

Of all the parameters, it is probably the cell balance detection threshold of 2.9V that has me most concerned.

I’m only interested in top-balancing and don’t see much benefit is boosting weak cells near the bottom of the discharge curve. That’ll just cause the weak cell(s) to top out more quickly when charging and won’t gain much of anything in terms of extended capacity.

My current passive BMS only balances above 3.55V so it is not active during a discharge cycle and only kicks-in well above the ‘knee’ when charging.

At a balance cut-off of 2.9V, any weak cells will be getting boosted through most of their discharge cycle and then that boosted charge will need to be pulled out and redistributed during the charge cycle.

So I may ask the manufacturer if I can get the BMS with balance Turn-On set to a higher voltage like 3.55V or 3.5V.

From my point of view (and as others have already stated), LiFePO4 battery capacity will fundamentally be limited by the capacity of the weakest cell.

The first 70Ah LFP battery I built has a weak cell that is the first to discharge, but what I’ve seen is that over time/cycles, that cell strays from being the first to recharge to lagging the stronger cells ;which means the battery is losing available capacity).

So my main goal with Active Balancing is to boost any weak cells to be balanced with the stronger cells at the end of a charge cycle.

Is there any specification for how much of a voltage difference is required between the lowest cell and the highest cell to engage Active Balancing?

 

[fafrd]

I too am interested in their active balancing, especially for a potential 8S x 3P x 280AH arrangement ... I'm concerned that 80ma might be light on for balancing 3 x 280AH cells in Parallel .. But I'm leaning to 3P x 8S arrangement with 3 BMS's to share load.
The Heltec website sucks.. the way they arrange their products and limited info online worries me. The 300A unit only supports 50A charging ?
and likewise down for lower current units. Perhaps its better to just add their active balancers in parallel with a BMS with up to 5A(?) support for 8S in a simple unit ? I must say my head is spinning from reading about all these and some negative comments about DALY and other cheap Chinese units. Often its not the cheap that is the issue , but communication, documentation and lack of consistency that worry most.

For ref : https://heltec-bms.com/project/8s-b...-300a-350a-3-2v-lifepo4-same-port-split-port/
says a 300a unit has 50A max charging current. The 350A spec has nothing stated.

But the AliExpress site is a bit better but I couldnt find this one from the entry level of that site !

62.59US $ |3s 4s 7s 8s 12v 24v Bms 1.5a Active Equalizer Balancer Continuous 150a 240a 300a 360a Lifepo4/ Lithium Battery Protection Board - Battery Accessories & Charger Accessories - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

And of course states the charging current is same as discharge limits. Which means 3 x 150A units would be great for my application.

This is the same as the model I am considering.

Reading carefully through the extended description, it seems to imply this:

50mA of Passive Balance kicking-in at 3.65V.

Up to 1.5A of Active Balance current kicking-in when cells above 2.9V have greater than 100mV difference from any ‘adjacent cell’.

So the balance only appears to work between the highest-charged cell and one of its neighbors when the voltage difference exceeds 100mV.

In my case with an 8S charger at 28.6V, perfect balance equates to all cells being at 3.575V and Active Balance will only kick-in if a weak cell is above 6.675 when the other cells are

 

[fafrd]

Darn, that’s never happened to me before.

...other cells are below 3.575v.

When that happens, 1.5A of current will be shifted from the high cell to it’s neighbors (presumably stopping as soon as the delta drops below 100mV).

You really need a delta of well-over 100mV above the knee for this Active Balance to do much of anything (and you also need to alternate strongest and weakest cells)...

 

[fafrd]

If I’ve understood correctly that Heltec’s Active Balancer BMS technology only kicks-in when a cell has more than 100mV delta versus one of it’s neighbors, that means it’s only going to be effective we’ll up the knee.

Looking at the generic LiFePO4 charge curves, 3.32V is ~90% SOC and 3.22V is ~20% SOC, so even cells with a 50% difference in capacity will not exceed 100mV of difference until the weak cell is below 20% SOC.

Above the ‘knee’ and close to 3.65C, 100mV translates to ~2.5% SOC, so the weakest cell can be over 3.65V @ 100% SOC while the strongest cell is below 3.55V @ 97.5% SOC.

Presumably you’d be in ‘float’ or voltage-charge mode by then so the Active Balance would begin shifting 1.5A (or 750mA if the 50% duty cycle is correct) of current from that weakest cell to an adjacent strongest cell.

After a full hour, the weakest cell will have transferred 1.5Ah (or 0.75Ah) to that stronger cell, meaning voltage will have dropped to ~3.63V (assuming 280Ah cells) and Active Balance will likely have shut off.

So if I’m understanding Heltec’s Active Balancing BMS correctly, it only does anything when charging well up the knee to 3.65V or higher.

If only charging to 80% SOC / 3.31V, you’ll never get Active Balancing to engage unless you have a truly dead cell (18.5% SOC / 3.21V).

My greater concern is at the lower discharge knee - a weak cell could drop below 3.0V (9% SOC) while stronger cells are above 3.12V (14% SOC). This would engage Active Balance in a manner that will need to be ‘undone’ at the high-end (as I’ve already stated in an earlier post).

On the other hand, that low-end Active Balance will just guarantee that the weak cell hits 3.65V first when charging (at which point Active Balance will engage again to redistribute charge to an adjacent strong cell), so it may not really be much of a problem...

I’m thinking this is probably a good technology to deal with weak/mismatched cells but still thinking it through.

 

[Dzl]

Your asking/pondering/musing about some interesting questions. I don't have much to add. But I am following along.

The balancing logic of this BMS seems unusual, but on the other hand, I think the balancing logic we are most familiar with is specific to passive balancing. Active balancing often seems to balance throughout more of the voltage range, I suppose because it can, whereas a passive balancing BMS can't do anything other than waste energy at the bottom end of the spectrum.

Theoretically, an active balancing BMS with a high enough balance current could balance a pack at the bottom and the top. But this would require a high enough balance current and a small enough imbalance between cells that it could compensate in the time it had (effectively performing a bottom balance and top balance once each full cycle). I think 15% to 30% of energy is still wasted with an active balancing BMS (as opposed to 100% w/ passive), so there are some losses, but because energy is transferred rather than dissipated as heat, you can feasibly balance at low SOC. At least this is what I'm thinking right now, this is really the first and only time I've really thought hard about the logical differences between active and passive balancing, and I'm not very confident in my understanding yet.

You raise many good points and questions worth exploring.

 

[ArthurEld]

The Heltec Smart BMS with active balance will balance any time the difference between cells exceeds the "Trigger Diff Volt" parameter setting.

From the manual -
e) Trigger Diff Volt - Triggered balance difference Voltage is the only parameter that controls balance. When the balance switch is on, when the maximum difference voltage of the battery pack exceeds this value, the balance starts, and the balance ends when the difference voltage is lower than this value. For example, the balance trigger voltage difference is set to 0.01V. When the battery pack voltage difference is greater than 0.01V, balance is started, and when the battery pack voltage difference is lower than 0.01V, the balance is ended. (It is recommended that the balance trigger difference voltage of the battery above 50AH is 0.005V, and the balance trigger difference voltage of the battery below 50AH is 0.01V)

This is from the Alibaba Heltec

 

[Gazoo]

Active balancing often seems to balance throughout more of the voltage range, I suppose because it can, whereas a passive balancing BMS can't do anything other than waste energy at the bottom end of the spectrum.

I find this interesting as well. The passive balancing on my BMS doesn't kick in until the voltage is more than 3.4 volts per cell or 27.2 volts for the pack voltage. Above that is when I start to see higher than a 10mv difference between the cells. Last night I realized the delta setting for balancing was set to 30 mv's. So I set it to 15mv's which is the recommended setting for my Overkill BMS. As I recall the balancing current is 70ma's.

Anyhow, I set the charger to 27.2 volts and it went into CV properly and continued charging like it's supposed to. The current came down gradually. But I noticed the voltage of the pack rising above 27.2 when the charger displayed around 300ma's. So just for kicks, I started bumping up the voltage on the charger and I kept bumping it up until the highest cell voltage of 3.65 volts caused the BMS to cut off. There was just less than a 60mv difference between the cells. And the charger displayed very little mah's that went into the cells when the BMS cut off.

I know when the pack voltage reached 27.10 volts, the Ah's in was 265. SOC was 95%. I had the BMS's nominal capacity set to 280ah's and have since set it to 275ah's since the total I have been able to get from the pack is 272ah's.

My BMS can be set to balance while charging or to keep balancing on. Since it needs in excess of 200ma's going to the cells to balance while charging, I set the balancing to remain active with the charger disconnected. It took about 2 hours for the cells to come within 15mv's of each other.

My conclusion is most people stay between the knees and there is very little difference in voltage between the cells in this area. Mine vareis between 6mv's and 20 mv's between the knees. I have found the sweet spot of my pack to be between 25.35 and 27.2 volts per cell. I don't know if this would improve with a higher amperage charger than the 12 amp charger I am using, and drawing higher currents than the 17 amps I am drawing. And I don't know what effect all of this would have on the cells drifting over time. I do know for now, I don't need any balancing as long as I stay within the knees. And I am not sure how effective any balancing would be using non capacity matched cells because the voltage on my pack shoots up once 27.2 volts is reached, and drops quickly once 25.35 volts is reached.

What I see on the top is a voltage difference of 0.083mv's when the highest cell is 3.65 volts. What I see on the bottom is a .5 voltage difference when the lowest cell is 2.50 volts. I would say my parallel top balancing worked really well....lol

 

[fafrd]

Your asking/pondering/musing about some interesting questions. I don't have much to add. But I am following along.

The balancing logic of this BMS seems unusual, but on the other hand, I think the balancing logic we are most familiar with is specific to passive balancing. Active balancing often seems to balance throughout more of the voltage range, I suppose because it can, whereas a passive balancing BMS can't do anything other than waste energy at the bottom end of the spectrum.

Theoretically, an active balancing BMS with a high enough balance current could balance a pack at the bottom and the top. But this would require a high enough balance current and a small enough imbalance between cells that it could compensate in the time it had (effectively performing a bottom balance and top balance once each full cycle). I think 15% to 30% of energy is still wasted with an active balancing BMS (as opposed to 100% w/ passive), so there are some losses, but because energy is transferred rather than dissipated as heat, you can feasibly balance at low SOC. At least this is what I'm thinking right now, this is really the first and only time I've really thought hard about the logical differences between active and passive balancing, and I'm not very confident in my understanding yet.

You raise many good points and questions worth exploring.

If you discharge at a very low rate, Active balancing to the bottom of the discharge cycle could theoretically deliver the full charge capability of the battery, dispite cell mismatch.

In the real world, this will never be the case.

My discharge will be at currents as high as 160A (0.6C) and almost never much below 24A (0.1C). No way for a full 1.5A or even 3A Active Balance at the low end to keep up with that.

The primary benefit I see to Active versus Passive Balancing is that, at the top-end above 3.5V, passive balancing will drain all high cells down to the voltage of the lowest cell, while Active Balancing will boost a single low cell to get up near the higher-voltage cells.

If a lone ‘weak’ cell is high and all other string cells are low, there will not be too much difference between Active and Passive (other than less energy lost), but when there are only one or two low cells with many more high cells, the difference is significant.

In general, a weak cell which bottoms first should also be first to top (especially if balanced near the bottom), but I’ve seen that over multiple cycles, my weak cell starts to lag at the top as well (possibly due to losing charge at the top due to passive balancing?).

So while theoretically, I can see an argument that Active Balancing doesn’t buy you much of anything, if in the real world a weak cell can end up at lower voltage near the top of the charge cycle than many stronger cells, there is a clear advantage to Active Balancing compared to passive balancing.