Max charge current to have balance on?

rg12

100 kW
Joined
Jul 26, 2014
Messages
1,584
Am testing the new version of the ANT BMS.

Basically, if you aren't aware ANT has been amazing so far (to me at least) and then since the chip shortage their main chip went up x10 in price so they have changed it to a slower one and then have to redesign the whole app (which to begin with was problematic to many).

Now with the new app they have added many BS parameters that don't make sense but there is one in particular that is pretty useful.

Up until the new version it wouldn't balance if you charge at over 15A but now this setting can be tweaked (comes at 10A by default).

My question is, why is it bad to have balance on when charging at say 20A?
Why wouldn't it balance at any charge speed?
 
The ANT BMS has a problem with heat , when it has to balance a few cells at the same time over a longer time. So balancing during the whole charge procedure with higher currents could end in a thermal problem.

The active BMS systems so not have this restriction.
 
dominik h said:
The ANT BMS has a problem with heat , when it has to balance a few cells at the same time over a longer time. So balancing during the whole charge procedure with higher currents could end in a thermal problem.

The active BMS systems so not have this restriction.

I'm setting the balance start voltage at 4.10V
Besides, a normal balance procedure can take from hours to days so the BMS needs to be built for that and it is, you just have to know it.

ANT BMS (previous version) heats to 60c very fast when set to the default 180mA and that's why I always limit it to 100-120mA max and even when 20 rows are balancing it wouldn't go above 55c on a hot day.
The 60c limit protected it but it actually makes things slower running at 180mA than 120mA as it stalls for a few seconds every time it hits 60c, waits for a few seconds to drop a few degrees and then restarts balancing.
It seems that the later units of the previous versions are handling the 180mA much better and it gets to 60c and holds there without stopping the balance procedure.

* Have installed about 80-100pcs of them over the last two years so I know all of the ANT BMS guts :D
 
Does your ANT BMS allow you to bottom balance via resistance discharge before MAX Bulk Charge Current? We'll use a 14S7P pack for the following ANT BMS WHAT-IF feature possibility ...

Controller discharge cutoff at say 46.2V (3.30V avg p-group) with bounceback after 2-3 hrs to 50.0V (3.57V avg p-group). Is your ANT BMS able to bottom balance the 14 p-groups within 25-30mV of each other using resistance discharge? Then MAX Bulk Charge to 58.8V (14 x 4.10V = 57.4V). Will your ANT MAX allow you to top balance the 14 p-groups within 0.050V (e.g. 4.083V to 4.113V = V variance of 0.050V) if need be or even within 0.025V of each other ?

Are you able to bottom balance charge before bulk charging so all 14 parallel groups are within 25-30mV of each other? Does your ANT BMS have a feature for bottom-balancing whether via charging or via resistance discharging ?
 
The entire concept of giving a robot total control over your battery is one issue few people even consider.

At the most simple level you really only have a limited number of cell groups and for anyone with a reasonable intelligence it's pretty easy to manage everything manually.

After all if you know each cell group and bottom balance as well as charge with knowledge of the weak cell you can eliminate the need for balancing. (reverse all that with top balancing which is inherently more dangerous on discharge)

What purpose does balancing even serve?

And I do very hard charging in the area of up to 38 amps at near 50 volts so I know what I'm talking about. (1500 watts is my circuit breaker limit)

If you are seriously hardcore about getting the maximum charge current you probably don't want a stupid robot running things.
 
rg12 said:
Why wouldn't it balance at any charge speed?
Top balancing is intrinsically harmful to longevity.

Only a good idea if the device gets the job done both quickly and gently.

High current charging is causing imbalance, only after the main charge is nearly done should balancing start.


rg12 said:
I'm setting the balance start voltage at 4.10V

Besides, a normal balance procedure can take from hours to days so the BMS needs to be built for that
Wow that is just abusive. You need a balancer that can finish in well under an hour, ideally say 10min.

Nothing should be heating up, definitely a sign you're using the wrong tool for the job.

Are you using the BMS for initial balancing before pack assembly?

Or are these like 100Ah cells?

Or crapped out secondhand scrap?

Use the BMS for protective functions only, get a dedicated balancer.
 
SafeDiscDancing said:
What purpose does balancing even serve?
Capacity utilisation.

And that is all, very marginal

if both HVC and LVC are based off cell/group level voltages.

With great imbalances say a delta of 0.1V or higher

AND pack level voltage cutoffs

then you need to sacrifice more capacity, be much more conservative in order to protect the cells.

But if for longevity's sake you are being very conservative, sacrificing say 25% cap utilisation anyway then you are both causing less imbalance and can afford to worry about it less.
 
I dont know. My thoughts. Mostly uneducated conjecture based on minimal experiences.

I think that is more than normal, ( new revision ANT algorithm).

I dont know why the old ANT would limit charge current / bleed voltage selection, it makes sense to me that it does not. 10A? Silly. I can charge the batteries in my garage at 2C, 100A, and it will be fine. Rated to take that. An Ant would be worthless here with its 10A.

Now, I know we are talking bout BMS... but bear with me for a little guys... : I like to begin balance at 3.4v, and maximize charge current. However, if you set balance " precision" to close, the BMS will sit there and hunt, all day. A hunting BMS is hot.... a " not hunting" bms is cool. So I set precision to a reasonable level, usually 2 or 3 mV for my big, fine, high cost cells. A small pack this may be way to small. you need 40-50mV or even less precision: for the balance hysteresis, encountered. This is adjustable in the new software?

RG: I propose this: Try two baseline tests: One, balance from 4.1v up, like you describe. The next, begin the balance at 3.4v/cell. Consider the time, the balance bleed current ( resultant in dissipate power ie heat) , and the SOC when charged for result. For each scenario. Measure the heat displayed by the BMS in both scenario.

ANT BMS (previous version) heats to 60c very fast when set to the default 180mA and that's why I always limit it to 100-120mA max and even when 20 rows are balancing it wouldn't go above 55c on a hot day.

What gets hot? This is too general a statement. I do not know if you have a FLIR, RG, but I have looked at the BMS the whole cycle watching on the camera: It is CERTAIN ORDER of bleeding sets, and they toggle, on and off, all the time, through the populated string. You can see it quite clearly on a FLIR camera. They exchange the hots and colds in order, in the bleed bank: If it is getting excessibly hot, I think the balance precision should be lowered ( increased numerically). Hunting.

Do you heat sink your BMS? RG? Have you a chance to try that? Thermal transfer. Might help a little, some of these BMS are just out of limits for the design capability of the resistance bank ( wasting 1w on a resistor made for 500 milliwatt (mW) ma.... Just cannot do it, bleed the voltage, at this current.. ) .

john61ct said:
rg12 said:
Why wouldn't it balance at any charge speed?
Top balancing is intrinsically harmful to longevity.

The ANT BMS has a problem with heat , when it has to balance a few cells at the same time over a longer time. So balancing during the whole charge procedure with higher currents could end in a thermal problem.

... or, give more time to dissipate the talked about dissipate power. This is how i see it?

rg12 said:
I'm setting the balance start voltage at 4.10V

Besides, a normal balance procedure can take from hours to days so the BMS needs to be built for that
Or crapped out secondhand scrap?

Use the BMS for protective functions only, get a dedicated balancer.

I kinda disagree with alot that has been put forth in this thread.

I think the PID loop in alot of modern BMS can use some work ( or changeable variables, of the P, I, or D, for that, but you cant get that with bleed resistors easily)... The PID loop will overshoot, undershoot, repeat. Cause excess heat, that the developer hopes ( hope is a great thing) is good enough, and within the heat capacity of the design.

Overshooting and undershooting ( PID LOOP BALANCER GATE CONTROL) is more important to look out for.. as opposed to charge current itself. We have the hard limits of the charge current in the datasheets: I dont suppose you guys like to ignore them. So we must be in bounds.

You guys see it. Charge a battery at 25A and it goes so far out of balance.. Charge it at 50A and does it goes out faster? I dont really agree with this especially if you never go full empty or full full: I dont know. A cell is capable of a ( rated?) max charge current,. You will know when you go over it ( the BMS isnt gonna be hot.. but the cells will proportionally get hot, and hotter, when you go over the max charge).

Do the math. My best advice. Time is measured in min, they are 1/60 of an hour. Capacity is measured in wHours, or watt days, or watt minutes. Watt seconds, watt weeks, watt months. Percentage is " per hundred".. ie.... 1 percent is 1/100.

How many watts are you wasting? How many watt hours is that per cell string? How long does this, or any specific string, under this balance? You guys ignore the process: They come up, they get in line, they go up more, they top out. For me.

A day, symbol d, is defined as 86 400 seconds. The Second is the base unit of time in SI units. … the duration of 9 192 631 770 periods of the radiation corresponding to the transition between two hyperfine levels of the ground state of the caesium 133 atom.

I begin balancing when the SOC vs voltage comes together and balance the whole time until full charge. This depends on the cells, is usually around 3.4v for my recycled EV cells. I begin balancing there.

From the way you guys describe, you have never done a middle balance. I found a middle balance is the best way to begin the baseline the cells. The log also ends there. I'm weird though. Negates some of the variables but thats my charger, very configurable.
With great imbalances say a delta of 0.1V or higher

Lol.... I dont think I ever saw such imbalance with the cells I use. 1/10? 10%? lol hell no. Most I see is... 0.001v.... ( 1/100) and is usually less, in the range of half a millivolt. Poor BMS trying to balance 10% variation over the pack. Lol. I can only imagine. Sure I have seen that, in beat ToolPacks, but those are small and the balance is bought back into line very low in SOC for my datalogging of baselines.... but when I balance the whole charge ( from 3.2v/cell to 4.25v/cell) they come in line fast. For they are so small. Same with th larger cells, but the variation is not great to start with.


Top balancing is intrinsically harmful to longevity.
High current charging is causing imbalance, only after the main charge is nearly done should balancing start.

This is a blanket statement that I wholly disagree with, from my experiences. I have some of the oldest batteries, strongest, you can find, all top balanced regularly to 4.28v/cell on the big charger.... With its 1A balance current. Mostly bulk charged to 4.28v and settles at 4.195v...

Whether it is 0.5C (7.5A), 1C ( 15A) or 2C ( 30A).. they logs show the same imbalance, from the start of the charge, ( where the charger stopped cause one of the string was going low, empty,, for longevity) and the same coming together... at the 3.4v point for my cells I like to use... They love to come together for me at 3.4v/cell.... right at the beginning of the CC mAh input... When charged from empty.

Never go full empty on any cell string. This is what throws them out. I can stop a discharge datalog at 3.0v/cell... and bulk charge right up to capacity full SOC and max v....

Empty them to 2.8v..... unilaterally.. and some will suffer greatly.. and get even more need for balancing current... ... when you go through the climbs of IR at zero SOC and 100% SOC....

Empty to 2.5v/cell and you see it more.

Shoot for zero (0v?) on a string and then a recharge.. see what is still in balance, and what is not.... ( if the cell strings have not been destroyed... yet)

IR climbs like hell on the 0% SOC and at the 100% SOC... Go for it.. Try to measure the IR of a dead battery.. on discharge...

Or measure IR of a full battery... on charge..... when it goes over 4.4v/cell..

Let us know what kind of heat you make. Why sit there hammering the topped out cells with current in and out ( charge and balance bleed, same time) while waiting for the lesser string groups to " catch up".... Nah . I like them all to come together much much earlier. Around teh place the natural SOC begins to fill the cell... So it can bleed the entire time of the charge.. From 3.4v up. Quick termination at chosen CV value ( whether it be 3.8v/cell, 4.0v/cell, 4.25v/cell, ect). A BMS that cannot hack the balance will sit there and get hot as the charger pounds the current into the last cells as they (dont) want to top balance... the good ones suffer. As the bad ones catch up.

Ok that is all.
 
WOW!!

Managed to read (and reread) between the lines and get what you're getting at and basically agree. That said some of your comments could be misunderstood :wink:
DogDipstick said:
you need 40-50mV or even less precision: for the balance hysteresis, encountered.
Ah so true ... as most Active Balance Boards don't even begin balancing until say a 20S p-group/row variance is more than 100mV and then only balances the 20 parallel groups back to within say 95mV (0.095V) of each other.

80% of the 20S p-groups(rows) may be within 45mV of each other with only 20% (4 p-groups) more than +60mV variance. IMO Active Balance Boards are only useful for packs of [usable] salvaged cells and packs near EoL. Even so (IMO) the usefulness of a relatively inepensive (cheapy?) ABB is debateable.

The problem with an ABB is when 1 or 2 p-groups are out of balance by at least 0.10V (100mV) more than the other 19 or 18 p-groups (e.g. 20S). Trying to balance a pack with 1 or 2 parallel groups(rows) considerably weaker(Grade B) than the other 19 or 18 p-groups is basically a no-win game.
DogDipstick said:
Never go full empty on any cell string. This is what throws them out. I can stop a discharge datalog at 3.0v/cell... and bulk charge right up to capacity full SOC and max v....

Empty them to 2.8v..... unilaterally.. and some will suffer greatly.. and get even more need for balancing current... ... when you go through the climbs of IR at zero SOC and 100% SOC....

Empty to 2.5v/cell and you see it more.

Shoot for zero (0v?) on a string and then a recharge.. see what is still in balance, and what is not.... ( if the cell strings have not been destroyed... yet)
Is DD reflecting on the herd mentality of some speed demon RC pilots? - :wink:
In simple English don't discharge li-ion parallel groups(rows) less than 3.2V.
DogDipstick said:
Ok that is all.
Worth reading a couple times to get the gist of it. You needed to get ALL that off your chest :thumb:
 
On either end if you try to balance all cell groups to be even it will then do the OPPOSITE when the pack gets to the other end.

So the whole thing becomes a kind of absurd comedy where energy is being either bled away or shuttled back and forth with no real benefit and actually it probably means more abuse on the weakest cell group because it's getting jerked around on both ends.

If you only do Top Balancing then you are very vulnerable to the weak cell getting murdered with reversal at near end of charge.

If you only do Bottom Balancing then you eliminate the reversal risk but need to focus harder when charging to not overcharge the weak cell group. All I do is hard charge my pack while watching the weakest cell and when it hits 4.19 volts I end charging. This means the other cell groups can be as low as 4.10 volts because they are underused.
 
SafeDiscDancing said:
On either end if you try to balance all cell groups to be even it will then do the OPPOSITE when the pack gets to the other end.

So the whole thing becomes a kind of absurd comedy where energy is being either bled away or shuttled back and forth with no real benefit and actually it probably means more abuse on the weakest cell group because it's getting jerked around on both ends.

If you only do Top Balancing then you are very vulnerable to the weak cell getting murdered with reversal at near end of charge.

If you only do Bottom Balancing then you eliminate the reversal risk but need to focus harder when charging to not overcharge the weak cell group. All I do is hard charge my pack while watching the weakest cell and when it hits 4.19 volts I end charging. This means the other cell groups can be as low as 4.10 volts because they are underused.
If everyone treated their expensive new 20S Grade A packs with care(non-abuse) there should be no weak parallel groups/rows having abnormal variance from the other p-groups/rows.

Problems arise with DIY [inferior] pack builds due to faulty workmanship and/or Grade B cells.
 
eMark said:
Problems arise with DIY [inferior] pack builds due to faulty workmanship and/or Grade B cells.

Quality control and getting different batches are a problem.

Unless you buy all the cells out of the same batch made at the same time you will have some differences.

Then if you spot weld your cells (not using pouches) those welds will introduce slight deviations as the welding does some damage.

I also find that within different batches their "Voltage Stepping" is different as you charge. Some cells jump up in voltage early then mellow out later on. My guess is different mixtures cause that. (batches are never exactly the same)

So there are always random factors causing differences.

For me I simply "accept the reality" and bottom balance then watch the weak cell on charging. That never fails. (it can't)

There is something to be said for simply eliminating the chance of failure... and I used to charge and pray but after reversing enough cells I decided to take better control over everything.
 
Last year, I had two unbalamced batteries. Thought a balance BMS upgrade would fix them. It got them balanced, but the weak cell groups still sagged and had to be coaxed back on every recharge. Meanwhile, the BMS would heat up. I don't like heat, so I replaced the weak cell groups. Post testing showed high self discharge, One cell went to zero on its own. Good riddance.

I can see why some manufacturers don't use balance BMS on inexpensive ebike batteries. If the cells are good, and battery is properly sized, they tend to stay balanced. If it goes bad, they're better off shutting down the battery and hoped it lasted past warranty. Let the user eat the cost.
 
SafeDiscDancing wrote: ↑Feb 27 2022 12:02pm
For me I simply "accept the reality" and bottom balance then watch the weak cell on charging. That never fails. (it can't)
Yes, am an advocate of bottom balancing before bulk charging with no need to top balance my 10S3P 30Q pack. When the the time comes that i need to top balance will use resistance discharge on the healthier p-groups until all p-groups are at least within 0.050V (if not 30mV) of each other.

After pack discharge and bounce back voltage (after a few hours) i've gleaned/noticed how to determine when a parallel group/row may have a weak cell. Via the use of balance leads i take p-group voltages before bottom balancing p-groups. Then bottom balance p-groups within 10-12mV of each other before bulk charging to 41.0V.

Each winter i disassemble my experimental Vruzend V.2 10S3P 30Q pack to test all 30 individual cells. Just testing the p-group voltage doesn't tell you the condition of each cell in a p-group.

So far any problem p-group has been due to just 1 cell out of the 3 cell p-group ... due to high self-discharge. The first winter i replaced 9 out of the 30 "141" 30Q cells with 9 "KH1T" 30Q cells in 2020. In hindsite i probaby would've only needed to replace 2-3 of the cells in 2020, as i recharge my pack within 24-48 hrs after discharge (so very little if any p-group discharge has occurred).

This winter i replaced another 6 "141" 30Q cells that had 250 c/d cycles so far. Probably would have only needed to replace 2-3 (as prevously explained). All 9 of the "KH1T" 30Q cells after 125 cycles in 2021 are AOK (no signs of high self-discharge). Hoping the same will be true with these "KH1T" 30Q cells after another 125 c/d cycles this year.

Here's the problem as i see it: By the time most DIYers locate a bad p-group whether 10S5P or 20S10P it's more likely due originally to just 1 weak p-group cell. Possibly a Grade B cell that has been slowly dying over weeks. Then eventually adversely affecting the other Grade A cells in that p-group and finally alerting the DIYer to a problem.
docw009 said:
If the cells are good, and battery is properly sized, they tend to stay balanced. If it goes bad, they're better off shutting down the battery and hoped it lasted past warranty. Let the user eat the cost.
If its been said before (which it has), its been said a thousand times :bigthumb:

FWIW i don't use a BMS. I may come to regret it, but until then i'll carry on :thumb:

As far as MAX Charge Current ... are you serious or in too much of a hurry ... haste makes waste :flame:
 
I have yet to see a protective BMS that does anything but top balancing, using resistance

so I take that as the topic here.

Yes there are dozens of combinations, of other technology methods, strategies and devices - yes manual middle as delivered, or bottom are better than top - but that's for another thread.

Balancing must only be done at your selected voltage / SoC point

singular, ONE point. Balancing at multiple points is not just a waste of time and unnecessary wear on all your gear, but also lifespan shortening of your pack.

There is nothing to be achieved going for a delta lower than 20-30mV

If your new high quality pack does go to say under 5mV delta within say 10-20 minutes, of course that's great, no harm there

but heroic efforts to get there, buying better balancers or letting a pack sit for hours at high SoC is just a waste, accomplishes nothing useful.



SafeDiscDancing said:
On either end if you try to balance all cell groups to be even it will then do the OPPOSITE when the pack gets to the other end.

If you top balance, then you must accept your highest imbalances will be at the end of discharge cycle. As stated above you are not supposed to try to "fix" that, unless you switch to manual bottom balancing **instead** of using a BMS at the top.


> If you only do Top Balancing then you are very vulnerable to the weak cell getting murdered with reversal at near end of charge.

Wut? Maybe you wrote this wrong, please rephrase it, makes no sense as is.

> If you only do Bottom Balancing then you eliminate the reversal risk but need to focus harder when charging to not overcharge the weak cell group.

No "extra effort" needed, either base charge termination on per-cell voltage - stop at first cell hitting 4.15V

or if pack-level, stop at 4.10Vpc and periodically verify at charge termination that no cell has hit 4.20V


> All I do is hard charge my pack while watching the weakest cell and when it hits 4.19 volts I end charging. This means the other cell groups can be as low as 4.10 volts because they are underused

Which is fine, and a perfect strategy when bottom balancing.

 
Wow, all of these replies answered so much more than I asked.
Thanks for all the great info :D
 
rg12 said:
Am testing the new version of the ANT BMS.
Me too. $72 BMS and i needed to buy a $100 android to use it, or sign up for WeChat which seems like you need to fly to china and marry someone to sign up. Lol.

Basically, if you aren't aware ANT has been amazing so far (to me at least) and then since the chip shortage their main chip went up x10 in price so they have changed it to a slower one and then have to redesign the whole app (which to begin with was problematic to many).

Now with the new app they have added many BS parameters that don't make sense but there is one in particular that is pretty useful.
Yes, and many are locked behind an administration password that so far i cant find, like current calibration and clear logs, reset timers. I got lvl5 set which at least lets me play around with it.

Up until the new version it wouldn't balance if you charge at over 15A but now this setting can be tweaked (comes at 10A by default).

My question is, why is it bad to have balance on when charging at say 20A?
Why wouldn't it balance at any charge speed?
I saw that setting and was trying to figure out what it was. I was looking for the “Balance only when charging” setting but that might be embedded in the “ChargingCellBalStart” voltage.

I have my 15s mock-up, let the BMS balance it at the top, as it should be, then turned off all balance functions and drained it with 2500 watts of water kettles which pulled 2a with this 48v 5ah pack.
8F22BEA4-1EE6-4B5D-A9D7-6B28EB3717BB.jpeg

Later ill be using the Ant for a 20s 5-6P 64v pack for my electric scooter (my avatar).

This pack has cells from 3 different sellers and they each behave differently. But with my bms i can test up to 24 cells at a time in just 3 hrs at 2a instead of 24 individually at 0.75a.

During my first discharge test i noticed the AH reported is only 1/2 the actual value. It’s showing the same error in all ios and Android apps i tried using, so it’s probably one of those changes that you mentioned that firmware software hasn’t caught up with because its stored onboard the BMS.

I saw a lot of nonsense about bottom balancing which anyone who understands the BMS knows that wont work. Its a carry over from those who manually manage their pack, but will not let the BMS do its job.

To your original question, i think the charge rate being 20a you don’t want to try to balance with 0.2a bleed because it will be totally ineffective and also it could push a brick with higher IR to a higher voltage than a brick with lower IR and lower capacity. So you just making more work. Lol.

The test im running now is how much imbalance i get starting with perfect top balance and discharge until one cell hits 2v, then charging at 2a until one cell hits 3.65v.

This is where the whole system come together if top Balanced, when the charger is able to use its taper from CC to CV, because it closes in on CV voltage. If bottom balanced, it will still be on CC when the balancing begins, and then soon enter 300 cycles of charging, bleeding, over and over.

But if you can let the balancing start after the taper in current, and after the flat part of the battery SOC curve, then every time it’s fully charged it will have a reasonable amount of time balancing at 0.2a because it’s tapering down 2a—>1a—>0.5a before it pushed the first brick to the OVP limit.

So on this test im charging and set it to stop everything after the first cell gets to 3.65v. Release voltage is as low as allowed 3.3v. But i think it wont decay to that value. Then i can let it settle out and see what type of balance current is needed.

I used 0.2a for initial top balance and lowered it 0.1a—>0.05a—>0.03a—>0.02a for the final stages of balance because it seems too much current when its almost balanced makes it keep going for no reason. Lol

I am running mybms_v1.3.0, tried 1.3.25 but it doesnt fit my display. If you know the admin password or any other app versions that work let me know so i can try them,
 
BareKuda said:
rg12 said:
Am testing the new version of the ANT BMS.
Me too. $72 BMS and i needed to buy a $100 android to use it, or sign up for WeChat which seems like you need to fly to china and marry someone to sign up. Lol.

Basically, if you aren't aware ANT has been amazing so far (to me at least) and then since the chip shortage their main chip went up x10 in price so they have changed it to a slower one and then have to redesign the whole app (which to begin with was problematic to many).

Now with the new app they have added many BS parameters that don't make sense but there is one in particular that is pretty useful.
Yes, and many are locked behind an administration password that so far i cant find, like current calibration and clear logs, reset timers. I got lvl5 set which at least lets me play around with it.

Up until the new version it wouldn't balance if you charge at over 15A but now this setting can be tweaked (comes at 10A by default).

My question is, why is it bad to have balance on when charging at say 20A?
Why wouldn't it balance at any charge speed?
I saw that setting and was trying to figure out what it was. I was looking for the “Balance only when charging” setting but that might be embedded in the “ChargingCellBalStart” voltage.

I have my 15s mock-up, let the BMS balance it at the top, as it should be, then turned off all balance functions and drained it with 2500 watts of water kettles which pulled 2a with this 48v 5ah pack.


Later ill be using the Ant for a 20s 5-6P 64v pack for my electric scooter (my avatar).

This pack has cells from 3 different sellers and they each behave differently. But with my bms i can test up to 24 cells at a time in just 3 hrs at 2a instead of 24 individually at 0.75a.

During my first discharge test i noticed the AH reported is only 1/2 the actual value. It’s showing the same error in all ios and Android apps i tried using, so it’s probably one of those changes that you mentioned that firmware software hasn’t caught up with because its stored onboard the BMS.

I saw a lot of nonsense about bottom balancing which anyone who understands the BMS knows that wont work. Its a carry over from those who manually manage their pack, but will not let the BMS do its job.

To your original question, i think the charge rate being 20a you don’t want to try to balance with 0.2a bleed because it will be totally ineffective and also it could push a brick with higher IR to a higher voltage than a brick with lower IR and lower capacity. So you just making more work. Lol.

The test im running now is how much imbalance i get starting with perfect top balance and discharge until one cell hits 2v, then charging at 2a until one cell hits 3.65v.

This is where the whole system come together if top Balanced, when the charger is able to use its taper from CC to CV, because it closes in on CV voltage. If bottom balanced, it will still be on CC when the balancing begins, and then soon enter 300 cycles of charging, bleeding, over and over.

But if you can let the balancing start after the taper in current, and after the flat part of the battery SOC curve, then every time it’s fully charged it will have a reasonable amount of time balancing at 0.2a because it’s tapering down 2a—>1a—>0.5a before it pushed the first brick to the OVP limit.

So on this test im charging and set it to stop everything after the first cell gets to 3.65v. Release voltage is as low as allowed 3.3v. But i think it wont decay to that value. Then i can let it settle out and see what type of balance current is needed.

I used 0.2a for initial top balance and lowered it 0.1a—>0.05a—>0.03a—>0.02a for the final stages of balance because it seems too much current when its almost balanced makes it keep going for no reason. Lol

I am running mybms_v1.3.0, tried 1.3.25 but it doesnt fit my display. If you know the admin password or any other app versions that work let me know so i can try them,

In the new version manual they say that the Level 5 password is 123456789abc and if it doesn't work to try 000000000000
 
rg12 said:
BareKuda said:
If you know the admin password or any other app versions that work let me know so i can try them,

In the new version manual they say that the Level 5 password is 123456789abc and if it doesn't work to try 000000000000

Yes, the level 5 password is pretty easy to find but it won’t allow access to certain settings. I need the Admin password which says it needs 15 character or so.
 
BareKuda said:
rg12 said:
BareKuda said:
If you know the admin password or any other app versions that work let me know so i can try them,

In the new version manual they say that the Level 5 password is 123456789abc and if it doesn't work to try 000000000000

Yes, the level 5 password is pretty easy to find but it won’t allow access to certain settings. I need the Admin password which says it needs 15 character or so.

Talk to this guy, I think he is the manufacturer:
https://www.aliexpress.com/store/2344164?spm=a2g0o.productlist.0.0.25e84036jkGbgA
I have order about 150pcs from him over the years.
 
rg12 said:
BareKuda said:
rg12 said:
BareKuda said:
If you know the admin password or any other app versions that work let me know so i can try them,

In the new version manual they say that the Level 5 password is 123456789abc and if it doesn't work to try 000000000000

Yes, the level 5 password is pretty easy to find but it won’t allow access to certain settings. I need the Admin password which says it needs 15 character or so.

Talk to this guy, I think he is the manufacturer:
https://www.aliexpress.com/store/2344164?spm=a2g0o.productlist.0.0.25e84036jkGbgA
I have order about 150pcs from him over the years.
I got software download from his website
https://www.rjxzstech.com/download/ANT-BMS-APP-20211116.html

But there are other later versions else’s so i’m not sure how he fits in. Did you order any new ones from him?

If you are a big customer he might tell you the admin password before me, lol
 
BareKuda said:
rg12 said:
BareKuda said:
rg12 said:
In the new version manual they say that the Level 5 password is 123456789abc and if it doesn't work to try 000000000000

Yes, the level 5 password is pretty easy to find but it won’t allow access to certain settings. I need the Admin password which says it needs 15 character or so.

Talk to this guy, I think he is the manufacturer:
https://www.aliexpress.com/store/2344164?spm=a2g0o.productlist.0.0.25e84036jkGbgA
I have order about 150pcs from him over the years.
I got software download from his website
https://www.rjxzstech.com/download/ANT-BMS-APP-20211116.html

But there are other later versions else’s so i’m not sure how he fits in. Did you order any new ones from him?

If you are a big customer he might tell you the admin password before me, lol

Just sent him a PM, will let you know once he replies 8)
 
rg12 said:
His reply 8)
Untitled.png
S778c7d3c0d7b4b6db7e3dda5a1c2f45ba.png
I can understand some of that being locked but current calibration isn’t one of those. Every other bms lets the user have access to the current calibration, and warns them they can get there bms abnormal if they calibrate it wrong. Lol.

I also wanted to try different baud rate to see if the lag reduces.
 
Back
Top