Top balancing - issues to look out for

[Gould]

I’ve read somewhere between 3 and 300 threads on top balancing and each one starts in the same way, green folks like myself thinking hey, how hard can this be? A simple question about how and what power supply to use quickly shifts into a comprehensive technical discussion. This is an amazing community with brilliant folks who understand this stuff inside and out. Personally, stumbling onto this site is the only reason I‘m able to even entertain the solar build that I’m planning.

I’ve watched Will’s how-to video too many time to recall and it does seem pretty straight forward. New cells will be arrive with a high SOC. Connect in parallel, set your bench top power supply to 3.6V, grab a coffee and wait for the amperage to drop to 100ma.

My planned approach is to follow Will‘s recommendation. Parallel charge until the amperage hits 100ma, let settle overnight, connect in a 4S configuration, attach BMS and done. I’ll likely run through some capacity tests to understand what my ranges are but that’s it. Is it really that simple? Here are the couple of units that I’m looking at. I’m not looking for something complicated - cost is not the concern, just looking for “simple”. I have far too much to learn about the end-to-end system that I need to pick my battles wisely! I’d like to understand what are the common issues/conditions/scenarios that I need to be mindful of. I’ve read a lot about the concerns of overcharging related to the very flat charge curve, dropping off very quickly. How safe are these cheap power supplies from that perspective? If set at 3.6v can you trust that it will not exceed? How on point does a person have to be, watching for the moment that they hit 100ma?

As always, really appreciate the input, and patience

DC Power Supply Variable, 0-30 V / 0-10 A LW-K3010D Adjustable Switching Regulated Power Supply Digital, with Alligator Leads US Power Cord Used for Spectrophotometer and lab Equipment Repair: Amazon.com: Industrial & Scientific

DC Power Supply Variable, 0-30 V / 0-10 A LW-K3010D Adjustable Switching Regulated Power Supply Digital, with Alligator Leads US Power Cord Used for Spectrophotometer and lab Equipment Repair: Amazon.com: Industrial & Scientific

www.amazon.com

https://www.amazon.com/TACKLIFE-DC-Power-Supply/dp/B08HLRDHDB

 

[FilterGuy]

I successfully use a very similar cheap china supply. However, your concern about the voltage drifting has some merit. The 'constant voltage' does drift on these. Mine drifts by plus or minus .02V. I top balance to 3.65 so I do watch it closely at the end. However, even if it goes to 3.67 the cells can take it.

If you top balance to 3.60V, you have a lot more room for the voltage to drift up before there is an issue.

Having said that, the first time you use *any* power supply to top balance, you should watch it closely to make sure it is behaving. After you have seen it operate a couple times, you will get a good feeling for it's quirks.

Top Balancing LiFePo4 Cells using a low cost benchtop power supply.

To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...

diysolarforum.com

 

[snoobler]

That's not Will's recommendation, and it's a waste of time.

Download the PDF and follow it exactly:

Top Balancing LiFePo4 Cells using a low cost benchtop power supply.

To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...

diysolarforum.com

 

[HRTKD]

@Gould, the Longwei power supply you provided a link to is the same brand that I used for my top balancing. Different model, however. The difference between your link and mine appears to be precision in the readouts and I have an additional readout indicating watts. I don't know that the precision is a big deal. My power supply worked like it was supposed to. I even used the OEM leads.

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

 

[zorlig]

I get confused when people say they will "follow Wills recommendation" but then what they say doesn't match those recommendations. Do the serial charge to save (a lot of) time.

The big danger is making a mistake or getting impatient and making a mistake. They aren't toys. I put my cells together in series exactly once and charged very slowly with a decent active balancer because I wanted to minimize chances of a short, minimize the work and keep them compressed at high SOC. It did work.

 

[FilterGuy]

The big danger is making a mistake or getting impatient and making a mistake. They aren't toys. I put my cells together in series exactly once and charged very slowly with a decent active balancer because I wanted to minimize chances of a short, minimize the work and keep them compressed at high SOC. It did work.

Interesting..... I am not a big believer in active balancers, but this may be a good use for them. Use them to get the series cells balanced reasonably quickly and then take them off.

I put my cells together in series exactly once

There is wisdom in this!! Putting screws in-and-out of the aluminum threads of the cell pads tears up the threads and leads to stripping them out.

 

[John Frum]

New cells will be arrive with a high SOC.

They usually arrive ~3.29VDC which is ~50% state of charge.

***Most important thing***
DO NOT ADJUST THE CHARGE VOLTAGE AFTER THE CHARGING LEADS ARE ATTACHED TO THE BATTERY!
********************************

 

[Gould]

@Gould, the Longwei power supply you provided a link to is the same brand that I used for my top balancing. Different model, however. The difference between your link and mine appears to be precision in the readouts and I have an additional readout indicating watts. I don't know that the precision is a big deal. My power supply worked like it was supposed to. I even used the OEM leads.

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

Appreciated.

 

[Gould]

I get confused when people say they will "follow Wills recommendation" but then what they say doesn't match those recommendations. Do the serial charge to save (a lot of) time.

The big danger is making a mistake or getting impatient and making a mistake. They aren't toys. I put my cells together in series exactly once and charged very slowly with a decent active balancer because I wanted to minimize chances of a short, minimize the work and keep them compressed at high SOC. It did work.

Completely up for that if it simplifies matters. Challenge as I understand is this could take months.

 

[Gould]

Interesting..... I am not a big believer in active balancers, but this may be a good use for them. Use them to get the series cells balanced reasonably quickly and then take them off.


There is wisdom in this!! Putting screws in-and-out of the aluminum threads of the cell pads tears up the threads and leads to stripping them out.

Is this the argument to move to welded posts?

 

[FilterGuy]

Is this the argument to move to welded posts?

Yes. It is at least one good argument for it. Way too many of us have stripped the tiny aluminum threads in the pads. I wish there were options other than Fortune for large welded posts.

Challenge as I understand is this could take months.

Maybe not. I have not tried balancing with an active balancer. However, based on my experience with cells, If you have an active balancer with a couple amp balance current and do a slow charge over a day or two, it seems like there would be a good chance it will be reasonably well balanced at the end of the charge. This is assuming they are not ridiculously out of balance to start with (if you buy from a reasonable supplier they are usually not too far out of balance.)

Hmmmmmm, now I have to talk myself down from buying a set of active balancers..... or figure out how to convince my wife it is important for me to have a set....

 

[Gould]

Yes. It is at least one good argument for it. Way too many of us have stripped the tiny aluminum threads in the pads. I wish there were options other than Fortune for large welded posts.

Maybe not. I have not tried balancing with an active balancer. However, based on my experience with cells, If you have an active balancer with a couple amp balance current and do a slow charge over a day or two, it seems like there would be a good chance it will be reasonably well balanced at the end of the charge. This is assuming they are not ridiculously out of balance to start with (if you buy from a reasonable supplier they are usually not too far out of balance.)

Hmmmmmm, now I have to talk myself down from buying a set of active balancers..... or figure out how to convince my wife it is important for me to have a set....

Haha, I’m still at the stage of convincing my wife the batteries are needed! Thankfully she can’t fall asleep without her TV and uses a hair drier religiously!

Just ordered 8 271ah CATL cells from Michael. Super nervous but excited. I’ve gone with M8 posts.

 

[FilterGuy]

I’ve gone with M8 posts.

M8 is a lot better than the M6 or even M4 some of the cells come with.

Ever since I striped some threads, I have gone to using grub screws and nuts.

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8 at Fastenere.com

www.fastenere.com

I screw them in once with a little red loctite and never take them back out. If I need to take bus bars on and off, I can take the nuts off without taking the grub screws out. (I *always* hold the grub screw with an Allen wrench when I am tightening or loosening the nuts)

If the cells come with regular headed screws.... I throw them into my big box of metric nuts and bolts that I will probably never use but can't bring myself to throw away because I might, just maybe, need them someday.

 

[HRTKD]

Ever since I striped some threads, I have gone to using grub screws and nuts.

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8 at Fastenere.com

www.fastenere.com

I screw them in once with a little red loctite and never take them back out. If I need to take bus bars on and off, I can take the nuts off without taking the grub screws out. (I *always* hold the grub screw with an Allen wrench when I am tightening or loosening the nuts)

Grub screws, set screws, or whatever you want to call them, can still strip the terminal threads. Put too much force on the nut that holds the bus bar or cable lug down and the screw can pull out of the terminal.

An advantage of grub/set screws is you can thread them in by hand with two fingers and get them all the way to the bottom of the terminal without any risk. That ensures that you have as many terminal and screw threads holding onto each other as possible. The fewer threads that are engaged, the more likely it is that the terminal threads will be stripped.

 

[fafrd]

M8 is a lot better than the M6 or even M4 some of the cells come with.

Ever since I striped some threads, I have gone to using grub screws and nuts.

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8

M8-1.25 Metric Socket Set Screws Cup Point Din 916 Stainless Steel 18-8 at Fastenere.com

www.fastenere.com

I screw them in once with a little red loctite and never take them back out. If I need to take bus bars on and off, I can take the nuts off without taking the grub screws out. (I *always* hold the grub screw with an Allen wrench when I am tightening or loosening the nuts)

If the cells come with regular headed screws.... I throw them into my big box of metric nuts and bolts that I will probably never use but can't bring myself to throw away because I might, just maybe, need them someday.

Did you retap all of your terminals from M6 to M8? If so, do you have an M8 tap you can recommend?

I just stripped my first thread and was going to get 15 M6 grub screws and a single 1/4@ grub screw I was hoping would self-tap into the stripped terminal.

Tapping all of the used-thrice terminals for M8 grub screws may be a superior solution your post is making me think about...

 

[FilterGuy]

Grub screws, set screws, or whatever you want to call them, can still strip the terminal threads. Put too much force on the nut that holds the bus bar or cable lug down and the screw can pull out of the terminal.

Yes.... but the advantage of grub screws is that you can put them in once and leave them. Every time the stainless steel threads go in and out of the aluminum threaded hole, it will tend to tear up the aluminum threads a little. The difference in hardness between the stainless steel and the aluminum is large enough that it does not take very many times to significantly loosen the aluminum threads. (I learned the the hard way)

 

[zorlig]

Maybe not. I have not tried balancing with an active balancer. However, based on my experience with cells, If you have an active balancer with a couple amp balance current and do a slow charge over a day or two, it seems like there would be a good chance it will be reasonably well balanced at the end of the charge. This is assuming they are not ridiculously out of balance to start with (if you buy from a reasonable supplier they are usually not too far out of balance.)

Hmmmmmm, now I have to talk myself down from buying a set of active balancers..... or figure out how to convince my wife it is important for me to have a set....

Mine was 2 days after the first cell got to 3.4. I had 2P24S and it had to do 0-5ah for each of the 24 (and it could balance 48ah a day, 2a*24 hours). I did a bit by hand too. I did stop the charge for a bit twice to keep the highest cell under 3.45 but in retrospect that might not have been necessary. I was charging at .75 amps, so it would have been easier if I had control of that, both higher at the start and lower towards the end.

This is my test setup, when I do a full house one I'll buy many more cells + real inverters I can control. I'll probably just charge up to a voltage that has the first cells at 3.4 and let the balancer work, then come back once a month for a while to increase that voltage until they are all 3.4 and balanced.

 

[FilterGuy]

Did you retap all of your terminals from M6 to M8? If so, do you have an M8 tap you can recommend?

I just stripped my first thread and was going to get 15 M6 grub screws and a single 1/4@ grub screw I was hoping would self-tap into the stripped terminal.

Tapping all of the used-thrice terminals for M8 grub screws may be a superior solution your post is making me think about...

First: Welcome to the "Ah Shit!" club.

I wrote up what I did in this resource:

How to drill and tap aluminum cell terminal to modify from M6 to M8 threads

To get this paper, click on the orange button at the top of the page I stripped out the M6 threads in the terminal hole of an Aluminum Case LiFePO4 Cell. This deck shows what I did to drill out the hole and tap for an M8 hole. Revision 2...

diysolarforum.com

The problem is that the hole is soooo shallow that the larger pitch of the M8 gives you fewer threads to grab. Consequently, it is trade off between larger diameter threads and fewer threads. I would drill them out deeper, but some of the comments on the forum imply there is very little aluminum under the hole so there is a big risk of punching through. Someday I might dissect an old cell to see what is there.

 

[FilterGuy]

I'll probably just charge up to a voltage that has the first cells at 3.4 and let the balancer work, then come back once a month for a while to increase that voltage until they are all 3.4 and balanced.

No matter how you do it....Patience is the key. Don't try and rush it.

 

[fafrd]

Yes.... but the advantage of grub screws is that you can put them in once and leave them. Every time the stainless steel threads go in and out of the aluminum threaded hole, it will tend to tear up the aluminum threads a little. The difference in hardness between the stainless steel and the aluminum is large enough that it does not take very many times to significantly loosen the aluminum threads. (I learned the the hard way)

I think that’s what I learned as well.

Threaded once for top-balance (parallel).

Threaded a second time for low-discharge-rate series capacity test of all 16 cells (serial).

Threaded a third time the top-balance the 8 best cells.

All 3 of these were by hand and well below 4Nm / 35inch-lbs.

Then threaded a fourth time today for high-discharge-rate series capacity test of 8S / 24V battery using 3 inverters and stripped one thread. I used a new torque-wrench for the first time today (which is how I know the first 3 threadings were tightened nowhere near 4Nm / 35inch-lbs) and 15 of the terminals tightened from 30inch-lbs to 35, but one stripped at 30inch-lbs...

I may have had a defective terminal / thread (didn’t inspect too carefully before use) but I don’t believe these soft aluminum threads can withstand 4Nm (and am certain that is true with repeated use).