Top Balancing "How to"

[ArthurEld]

Hi, surely I was a bit hesitant after well over a week of charing, I had the charge voltage little too high and I was measuring the charging voltage, not the internal voltage of the battery. so the batteries were 90% full. So absorbed slowly the up to 3.600V, and followed the one battery that was behind every now and then. The battery did catch up with the others.

So after charging them up to 3.000V they are now settled all 3.573V equal over the 8 cells. I'm happy, and can now start building my 560Ah 12V battery with REC-BMS and Victron chargers.

The reason I asked about the absorb was because your cells settled so far the first time. But you probably also had a bad connection on one cell.
Hopefully you cleaned the connections on that cell.
When there is a bad connection it takes forever to charge because the energy can't get through.

 

[Gauranga108]

Hi everyone I have been attempting to top balance my 32 eve 280 cells. My process due to the size has been to use my solar system and bms to charge to higher soc then hook up in parallel and charge the rest of the way with 10amp dc power supply. That being said in the process of charging via the bms naturally I had some runners that tripped the bms over cell high voltage I will include bms screen shot. From there I have linked the first 16 cells of 32 in parallel and they are almost at 3.65. My question is if I have a few runners or unmatched cells once they are top balance then I run them, as I discharge once they need charged again won't those lower capacity cells or runners start shooting up fast again in the stage of charging, there for defeating thr purpose?

 

[Gauranga108]

Also on a side note my system is set up 16s2p. I have two separate banks of 16 each with 200amp daly bms then paralleled. I thought it might be a coincidence but that number 6 cell i replaced out with one from my other bank of 16 cells that was seemed normal and then continued the bulk of the charging prior to switch to parallel, but so happens that same number 6 position with the new cell from other bank shot up...? any thoughts?

 

[snoobler]

Hi everyone I have been attempting to top balance my 32 eve 280 cells. My process due to the size has been to use my solar system and bms to charge to higher soc then hook up in parallel and charge the rest of the way with 10amp dc power supply. That being said in the process of charging via the bms naturally I had some runners that tripped the bms over cell high voltage I will include bms screen shot. From there I have linked the first 16 cells of 32 in parallel and they are almost at 3.65. My question is if I have a few runners or unmatched cells once they are top balance then I run them, as I discharge once they need charged again won't those lower capacity cells or runners start shooting up fast again in the stage of charging, there for defeating thr purpose?

The point of top balancing is to get all cells at true 100% SoC.

As you discharge them, they all fall at different speeds from 100% based on each cell's capacity. When the first one is at 20%, and you terminate the discharge, the others might be at 21-23%. When you charge them, they all climb at different rates and return to 100% at the same time.

It is very important that you configure balancing correctly. The only time balancing should be enabled is 1) during charge, 2) when cells are above 3.4V, and 3) when cells deviate more than 30mv (.03V). Balancing at all times can actually undo a top balance.

Also on a side note my system is set up 16s2p. I have two separate banks of 16 each with 200amp daly bms then paralleled. I thought it might be a coincidence but that number 6 cell i replaced out with one from my other bank of 16 cells that was seemed normal and then continued the bulk of the charging prior to switch to parallel, but so happens that same number 6 position with the new cell from other bank shot up...? any thoughts?

1) always confirm BMS reported readings with a separate meter.
2) if you see a runner, swap positions with another cell. If the runner moves, it's the cell. If the runner doesn't move, it's the BMS.

 

[Gauranga108]

Thanks that was helpful ?

 

[RogerD]

For current best practices on top balancing cells, please watch the video and download the PDF at the following page:

Top Balancing Cells Using a Low Cost Benchtop Power Supply

For top balancing. Does it matter if you use a Longwei 3010DF VS K3010D power supply? They are both 30v / 10a. The price difference is only $11, so pretty much irrelevant. I'm leaning toward the K3010D because it's smaller. But I have no problem buying the larger 3010DF if there is some benefit.

3010DF

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

www.amazon.com

K3010D

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

www.amazon.com

 

[Substrate]

Take note of what Will says at around 4:23 into the video.

That you may run into problems trying to charge these up from a low state of charge with such little current compared to capacity.

If your charger is less than .2C, (which is the typical spec for a "standard charge" on spec sheets for LFP) - then there is a chance that either of two things happen if the cells aren't already at a higher state of charge:

1) Your cells seem to stall and never reach 3.65v
2) It might take DAYS to do so.

Which is bad.

That standard charge spec is there for a reason. It has to do with being too inefficient to do the job properly.

On another chemistry like lead-acid, you might have seen some trying to charge their 200ah agm bank with a weedy little 800ma "battery tender" from the auto-parts store, and spending a week doing so. It is such common abuse that most modern tenders include cautions against doing this in the manuals - which many don't read.

Just saying - if your cells come pretty well charged to begin with, no problems. But how do you know, unless you consult a resting voltage chart, AND have a trusted multimeter?

The temptation is to pay good $$ for your huge LFP bank, and then cheap out on the support infrastructure. I recommend not doing that. Will's video is great, but in our excitement to get going, it might look like a 10A power is suitable as a universal solution, when it is not, as evidenced by the mention at 4:23 into the vid - which is great overall!

My suggestion with a large number of cells in parallel is to pair up cell capacity that don't exceed 0.2C of your power supply, and do it multiple times if need be on successive smaller pairings, instead of laying a huge string across the bench and trying to do it at all at once.

With only a 10A power supply, I would pair up cells no larger than 50ah in capacity overall, do the top balance, and repeat as necessary to the others. Do that once, and you'll spend the money for a bigger charger next time.

Your bms balancers should take care of any smaller irregularities when you put them all together as a system.

 

[RogerD]

Take note of what Will says at around 4:23 into the video.

Yes, I totally follow what Will was saying in the video. I'm getting the AIMS 12/24v 70/37.5a charger as soon as they are back in stock, which is supposed to be mid-June. I don't even have my cells yet. Just getting all my ducks in a row for when I do.

I was just curious if it mattered which charger I get to do the actual top balancing.

 

[Substrate]

Yeah, no problem. I just go wall of text too often because I've destroyed banks learning.

Don't get me wrong - LFP has super charge-acceptance, but it wasn't designed to be trickle-charged. The major issue is that LFP was designed for rapid cycling, and spending the LEAST amount of time charging - as long as you stay within the specs.

The other problem is stating the absolute without taking into account context of the real world.

What Will did in the video would be considered a "stepped" parallel balance, since he was starting out with relatively high SOC cells. And then continuing on with his secondary step.

But when WE get them from off the boat, they are at most 30% charged (or supposed to be). At very low charge rates and large capacities, time is not friendly to our cells getting them parallel balanced.

You'll be cool. If given an option to accomplish a task quickly and stay within spec, choose the option that takes the least amount of time with LFP, not the cheapest.

 

[RogerD]

Don't get me wrong - LFP has super charge-acceptance, but it wasn't designed to be trickle-charged. The major issue is that LFP was designed for rapid cycling, and spending the LEAST amount of time charging - as long as you stay within the specs.

When you say to stay within the specs ... what specs are you referring to stay within?

I'm starting with (8) 280 ah cells. The plan is to wire as 24v with overkill bms. I will more than likely buy a second set of 8 shortly after.

Wired as 24v / 280 ah bank. Using a charger that is 24v / 37.5a isn't quite .2C. Is this OK?

Should I initially wire as 12v / 280ah and charge 12v / 70a which would then be over .2C? I would have to buy a 12v bms just for doing this though.

Thanks for the advice.

 

[ArthurEld]

I top balanced half of my cells to 3.65V.
The rest won't go above 3.45 unless my balancer can't keep up
.

 

[HRTKD]

For top balancing. Does it matter if you use a Longwei 3010DF VS K3010D power supply? They are both 30v / 10a. The price difference is only $11, so pretty much irrelevant. I'm leaning toward the K3010D because it's smaller. But I have no problem buying the larger 3010DF if there is some benefit.

3010DF

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

www.amazon.com

K3010D

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

www.amazon.com

As near as I can tell, the difference between the two devices is that one has a USB charge part and the other doesn't. I have the USB charge port one and it worked fine for me, though I have yet to use the USB port and probably never will.

 

[Just John]

For top balancing. Does it matter if you use a Longwei 3010DF VS K3010D power supply? They are both 30v / 10a. The price difference is only $11, so pretty much irrelevant. I'm leaning toward the K3010D because it's smaller. But I have no problem buying the larger 3010DF if there is some benefit.

3010DF

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Switching Regulated Multifunctional Power Supply Digital with USB Interface, Disply with Output Power Lab Grade: Amazon.com: Industrial & Scientific

www.amazon.com

K3010D

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

DC Power Supply Variable 30V 10A, 4-Digital LED Display, Precision Adjustable Regulated Switching Power Supply Digital with Alligator Leads US Power Cord : Amazon.com: Industrial & Scientific

www.amazon.com

Both of those supplies are very nearly identical inside, one has a 2 amp USB circuit, the other doesn't.
Both are solid choices if you don't want to spend money.

Cheap power supply for top balancing (tested and works well).

First, a link to the cheapest 30 volt, 10 amp supply I could find on Amazon. There are others on occasion that go on special for a dollar or two less, but this one is pretty consistently the cheapest I could find. https://smile.amazon.com/gp/product/B07512KQDW/ An alternative that is not as...

diysolarforum.com

The RD6018 is a step up (programmable):

133.0US $ 20% OFF|Rd6006w Rd6012w Rd6018w Rd6006pw Dc Dc Bench Power Supply Multimeter Whole Set Not Assembled Adjustable Buck Mode Converter - Multimeters - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

You can also hook them up as "solar input" to your charge controller and get a lot more amps, just FYI.

Unfortunately the Tekpower 40 amp supply I bought from Amazon for $190 is out of stock.

Wanptek is the company I think either made the Tekpower, or copied their design. They have some reasonably priced high amp supplies.
Pick the right voltage and they make a good charger as well.

146.85US $ 31% OFF|Bench Source Laboratory Dc Stabilized Power Supply Adjustable Variable Digital Regulated Power Supply 30v 20a 60v 10a Wanptek - Switching Power Supply - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

 

[Just John]

Yes, I totally follow what Will was saying in the video. I'm getting the AIMS 12/24v 70/37.5a charger as soon as they are back in stock, which is supposed to be mid-June. I don't even have my cells yet. Just getting all my ducks in a row for when I do.

I was just curious if it mattered which charger I get to do the actual top balancing.

Well, if you want to do top balancing (I recommend it), then the Aims is not going to work. It won't go down to 3.65v. The Aims is for charging packs.

 

[Horsefly]

I don't know that I would recommend someone doing it on purpose, but due to some strange fortunes I have ended up with 4 separate bench power supplies, all CV/CC adjustable output. Three can go up to 10A and one goes up to 20A. By adjusting them all to 3.65V I can connect them in parallel and have a 3 x 10 + 20 = 50A charger. I don't think I want to push the cheap 10A chargers all the way to 10A, so maybe it would be more like 45A max.

 

[Just John]

I don't know that I would recommend someone doing it on purpose, but due to some strange fortunes I have ended up with 4 separate bench power supplies, all CV/CC adjustable output. Three can go up to 10A and one goes up to 20A. By adjusting them all to 3.65V I can connect them in parallel and have a 3 x 10 + 20 = 50A charger. I don't think I want to push the cheap 10A chargers all the way to 10A, so maybe it would be more like 45A max.

Most have no problems putting out the full 300 watts they are capable of for indefinite times. If you are talking about top balancing, then sustaining 37 watts is not a problem. I am sure they all use parts from the lowest bidder, so there will obviously be some failures, but on the whole, many different brands are identical in design.

Buy cheap supplies and test them, you will likely get your money's worth. $50 to $60 is pretty reasonable price for top balancing, if you're worried it isn't powerful enough, for approximately $200, you can get a high amp supply that will do double duty as a pack charger.

The ability to parallel supplies depends on the design, and your willingness to risk $60 to see. But you can get much higher amps by buying one for about the price of 3 cheap supplies. I paid $190 for a 40 amp supply, or about what 3 of the cheap supplies costs.

 

[Horsefly]

Most have no problems putting out the full 300 watts they are capable of for indefinite times. If you are talking about top balancing, then sustaining 37 watts is not a problem. I am sure they all use parts from the lowest bidder, so there will obviously be some failures, but on the whole, many different brands are identical in design.

Maybe, but I have experienced a failure. One of the reasons I have so many supplies: I bought one of those cheapies (this one), and about 3 weeks after I got it I had it charging a cell at 8A. I came back a couple of hours later and the display was blank. It looked like the power had gone out. I started contacting Amazon, who had me contact the seller. I tried 4 times over the course of two weeks, but they never responded. So I posted a nasty review and ordered another cheapy (this one), as well as a bit more expensive one (this one). When the new cheapy showed up I opened it up and looked at the circuit board compared to the one that blew. Like you said, they all have basically the same design. I found I had a blown thermistor. I ordered a replacement part, and fixed the blown supply.

About then the seller of the failed supply finally responded to me and said they were sending me a new one, and they didn't want the old one back. They didn't even ask me to edit my review. In the end, I had 4 power supplies.

The ability to parallel supplies depends on the design, and your willingness to risk $60 to see.

I guess I don't understand that statement. My MSEE degree is pretty dated at this point, but I know of no risk introduced by putting power supplies set to the same voltage in parallel. No greater risk than putting multiple LFP cells in parallel.

I've been doing either 2 or 3 tied together for several weeks now.

Edit to add: By the way, no one should ever trust the voltage displayed on any of these power supplies, including the $200 bench supplies. Use a good multimeter (that hopefully has been verified against a voltage reference) and adjust the supply voltage using the meter. Once you get it set I can almost guarantee that the display of the power supply will show something other than what you now know the voltage is set to.

 

[Just John]

Maybe, but I have experienced a failure. One of the reasons I have so many supplies: I bought one of those cheapies (this one), and about 3 weeks after I got it I had it charging a cell at 8A. I came back a couple of hours later and the display was blank. It looked like the power had gone out. I started contacting Amazon, who had me contact the seller. I tried 4 times over the course of two weeks, but they never responded. So I posted a nasty review and ordered another cheapy (this one), as well as a bit more expensive one (this one). When the new cheapy showed up I opened it up and looked at the circuit board compared to the one that blew. Like you said, they all have basically the same design. I found I had a blown thermistor. I ordered a replacement part, and fixed the blown supply.

About then the seller of the failed supply finally responded to me and said they were sending me a new one, and they didn't want the old one back. They didn't even ask me to edit my review. In the end, I had 4 power supplies.

I guess I don't understand that statement. My MSEE degree is pretty dated at this point, but I know of no risk introduced by putting power supplies set to the same voltage in parallel. No greater risk than putting multiple LFP cells in parallel.

I've been doing either 2 or 3 tied together for several weeks now.

Edit to add: By the way, no one should ever trust the voltage displayed on any of these power supplies, including the $200 bench supplies. Use a good multimeter (that hopefully has been verified against a voltage reference) and adjust the supply voltage using the meter. Once you get it set I can almost guarantee that the display of the power supply will show something other than what you now know the voltage is set to.

The more expensive one you bought looks like the Tekpower 40 amp supply I bought, pretty sure they are all manufactured by Wanptek with just different brands on them. Mine has been very reliable. The cheap ones you linked, I also had two of them, tested them extensively for long periods at full output and other common voltages without issues. I am not sure if some of the buck converter models (like the Riden) can be paralleled, likely they can. I will test that this weekend.

Both of those supplies are very nearly identical inside, one has a 2 amp USB circuit, the other doesn't.
Both are solid choices if you don't want to spend money.

Cheap power supply for top balancing (tested and works well).

First, a link to the cheapest 30 volt, 10 amp supply I could find on Amazon. There are others on occasion that go on special for a dollar or two less, but this one is pretty consistently the cheapest I could find. https://smile.amazon.com/gp/product/B07512KQDW/ An alternative that is not as...

diysolarforum.com

The RD6018 is a step up (programmable):

133.0US $ 20% OFF|Rd6006w Rd6012w Rd6018w Rd6006pw Dc Dc Bench Power Supply Multimeter Whole Set Not Assembled Adjustable Buck Mode Converter - Multimeters - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

Lots of reviews on YouTube and the web for various models of the Riden.

You can also hook them up as "solar input" to your charge controller and get a lot more amps, just FYI. Especially the Riden.

Unfortunately the Tekpower 40 amp supply I bought from Amazon for $190 is out of stock.

Wanptek is the company I think either made the Tekpower, or copied their design. They have some reasonably priced high amp supplies.
Pick the right voltage and they make a good charger as well.
I have seen reviews of a couple of these, and they were all favorable.

146.85US $ 31% OFF|Bench Source Laboratory Dc Stabilized Power Supply Adjustable Variable Digital Regulated Power Supply 30v 20a 60v 10a Wanptek - Switching Power Supply - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

 

[Substrate]

@PeterBC - Thank YOU! You reminded me to look once again at the Marine HowTo and it solved a major sticking point for me, one of which was critical.

DIY LiFePO4 Batteries On Boats - Marine How To

DIY LiFePo4 On Boats Last edit;3/26/23:The base of this article was written a number of years ago (2010) but this does not mean the information here is outdated. We have been keeping it updated and have added to it when ever we had the time. This article deals

marinehowto.com

You listed how he showed the "Parallel Step-Method Top Balance" in msg #8 - aka "PSMTB"

Essentially a 3-step process, going from 3.40v to zero tail current, to 3.50v and zero tail, then to 3.6v (or whatever you want) and then zero, or perhaps less than zero depending) and you are done.

What makes this so appealing is that very first step to 3.40v - as described by the author in the Marine How-To article, is that the first step of getting to 3.40v and a tail-current of zero, can take as long as you need! Time here is not a problem - so if you have only a 10A charger, but 800ah of cells to charge. No worries if that first step takes an outrageously long time on a huge bank.

I had a spirited discussion about the damaging effects of time using very very low currents to initially charge up a very high-capacity bank from a low state of charge, and this 3-step solution is what I know I'll being doing from now on when getting cells directly off the boat.

Once this is done, the next two step-ups will be relatively fast, reducing my concerns about the overall time spent doing this procedure if one merely set a single high CV point and twiddled fingers for 5 days.

OT: I commend this Marine HowTo author. Thanks for making me revisit his work. I mean where else can you find a howto on which direction you should place your fla banks in? One look at the pic explains it. Makes me want to own a boat, because I'd trust him.

 

[Gazoo]

What makes this so appealing is that very first step to 3.40v - as described by the author in the Marine How-To article, is that the first step of getting to 3.40v and a tail-current of zero, can take as long as you need! Time here is not a problem - so if you have only a 10A charger, but 800ah of cells to charge. No worries if that first step takes an outrageously long time on a huge bank.

I tried the step method tp parallel top balance my 8 EVE cells and it didn't work for me. I am attaching a photo of my 12 amp power supply set at 3.40 volts and it's only putting out 5.84 amps, and it's in CV mode. I used 12awg cables. The PS wouldn't put out the the full 12 amps until I set it above 3.50 volts. Then the PS went into CC mode and put out the full 12 amps. So I ended up setting it at 3.65 volts and cut the time it took to top balance in half.

After about 4 days of twiddling my fingers, the PS went into CV mode and it only took about 2 hours to finish the balance. I let the tail current drop to .100ma's. I was using the charging feature of the Riden and I had tested it on another battery before parallel top balancing so I slept while while it was doing it's thing. But I was happy to be awake when it was finishing up the charge so I could watch the end of charge.

There sre several methods to parallel top balance and in the end they all accomplish the same thing. I guess I am set in my old ways which is simply connecting all the cells in parallel, and setting the PS to 3.65 volts before connecting it to the cells. Of course using a traditional PS one should monitor the voltage and current of the PS and monitor the voltage at the cells.