diy solar

diy solar

Runner cell "trigger" voltage

Except, there it says to set the voltage at 3.65. If I do that, I can only get some 3.5A out of the supply. I want 10 if I can, don't I?

Build better cables. It has been discussed at length and in detail many times.

You *WILL* damage your cells, then try to blame Amy for it.

There is a REASON the guide explicitly states to set the voltage to 3.65v when not connected.
Nobody wants to listen to experience, they always know better and wind up damaging things, and try to blame it on anything other than "I did what I should not have done, despite being told not to do that".
 
Interesting thread - can OP please report back and show us an image of the JK-BMS app with pack voltage at 27.5V, hopefully by now the active balancer will be bringing the cell voltages into line and demonstrating that an active balancer is an alternative to top balancing in parallel.
 
Interesting thread - can OP please report back and show us an image of the JK-BMS app with pack voltage at 27.5V, hopefully by now the active balancer will be bringing the cell voltages into line and demonstrating that an active balancer is an alternative to top balancing in parallel.

It really just depends, and in the vast majority of cases, it's not, unless you're very patient.

Unfortunately, active balancers, while they may claim 2, 4, 5, 6 or even 10A capability, that's usually based on AT LEAST a 1V difference between cells, which is exceedingly rare, balance at far less than their rated current. A 2A balancer with only a 0.05V different is only going to see 100mA of balance current - about 1.5-2X the typical passive balancer.

I have 6A balancers on my 14S NMC battery. Even with a pretty substantial voltage deviation, I never saw more than 0.4A of balance current, and I had to rely on the more robust 0.5A resistive balancers of the Batrium to get it into balance over several days as the total imbalance was on the order of 60Ah on a 400Ah battery.
 
Last edited:
Interesting thread - can OP please report back and show us an image of the JK-BMS app with pack voltage at 27.5V, hopefully by now the active balancer will be bringing the cell voltages into line and demonstrating that an active balancer is an alternative to top balancing in parallel.
I have tried 3 different active balancers (including the JK BMS) and it will work. However it will take time and effort, since the active balancers don't move nearly as much energy as they claim. You must get all of the cells above 3.6v, don't just have one cell hit 3.65v, let the active balancer work and think you are done when they all measure 3.4v, then you add more energy (preferably with reduced current) and continue again until they are all equal above 3.6v.
 
can OP please report back and show us an image of the JK-BMS app with pack voltage at 27.5V

At 27.7, I now have 6 cells in the (almost) 3.5s and two (#3 and #6) in the 3.3-3.4s.

voltage_cell01 3.491 V
voltage_cell02 3.494 V

voltage_cell03 3.337 V
voltage_cell04 3.478 V
voltage_cell05 3.494 V

voltage_cell06 3.338 V
voltage_cell07 3.494 V
voltage_cell08 3.495 V
average_cell_voltage 3.453 V
delta_cell_voltage 0.158 V
current_balancer 0.519 A
battery_voltage 27.624 V
battery_t1 30.0 °C
battery_t2 29.4 °C
mos_temp 30.1 °C
percent_remain 99 %

capacity_remain 277.2 Ah

They have been so for more than a week. As soon as charge stops, they drop into the 3.3s, quite balanced.

voltage_cell01 3.308 V
voltage_cell02 3.306 V
voltage_cell03 3.306 V
voltage_cell04 3.308 V
voltage_cell05 3.306 V
voltage_cell06 3.306 V
voltage_cell07 3.306 V
voltage_cell08 3.306 V
average_cell_voltage 3.306 V
delta_cell_voltage 0.005 V
current_balancer 0.256 A
battery_voltage 26.45 V

[EDIT]
I hardly ever use the phone app any more. I use
#!/bin/bash
konsole --hold --geometry 500x400+20+20 -e /bin/bash -c "jkbms -p C8:47:8C:E2:8D:26 -P JK02 -c getCellData | sed -n '7,14p ; 31,33p ; 58,65p'"

Where jkbms is part of this suite. So I just click an icon on my computer screen and get the relevant information.


-
 
Last edited:
They have been so for more than a week. As soon as charge stops, they drop into the 3.3s, quite balanced.

Nope.

Have you already forgotten?

 
How about individually charging those two 3.3xxV, and the 3.47x, to 3.495?

Balanced is what balanced does at charging voltage, not resting voltage.
As it stands now, you can charge your pack to something less than 27.7V without getting a runner.
Stopping charge with lowest cell at 3.337, it could drift lower without getting topped off. Or rather, balancing only occurs as bleed-off of cells presently 3.495V

After bring up those three lower cells, you should be able to charge to something less than 28.02
But still not good enough for a top-balance. Would like to charge each cell to 3.65V once, so all well above knee.
Then, when pack is charged at the now-higher pack voltage (something under 29.2V) without a runner, balancing can occur.
If no bad, excessively self-discharging cells, balance should then be maintained.
 
Imagine filling 8 bottles simultaneously, with 8 hoses putting juice into each at the same rate:

1660659113718.png

The "runners" are the ones where level shoots up the narrow neck while others still have level at the wide part of bottle.
Nothing wrong with any of the bottles, just some started out less full than others.

The bottles are only filled in parallel, but charging cells in series means current that charges one is same as current that charges all the others. That is why you need to fill each of the cells once (individually, or wired in parallel). And fill up the the same level in the neck, not declare that levels are the same down at the wide part.
 
At 27.7, I now have 6 cells in the (almost) 3.5s and two (#3 and #6) in the 3.3-3.4s.

voltage_cell01 3.491 V
voltage_cell02 3.494 V

voltage_cell03 3.337 V
voltage_cell04 3.478 V
voltage_cell05 3.494 V

voltage_cell06 3.338 V
voltage_cell07 3.494 V
voltage_cell08 3.495 V
average_cell_voltage 3.453 V
delta_cell_voltage 0.158 V
current_balancer 0.519 A
battery_voltage 27.624 V
battery_t1 30.0 °C
battery_t2 29.4 °C
mos_temp 30.1 °C
percent_remain 99 %

capacity_remain 277.2 Ah

They have been so for more than a week. As soon as charge stops, they drop into the 3.3s, quite balanced.

voltage_cell01 3.308 V
voltage_cell02 3.306 V
voltage_cell03 3.306 V
voltage_cell04 3.308 V
voltage_cell05 3.306 V
voltage_cell06 3.306 V
voltage_cell07 3.306 V
voltage_cell08 3.306 V
average_cell_voltage 3.306 V
delta_cell_voltage 0.005 V
current_balancer 0.256 A
battery_voltage 26.45 V

[EDIT]
I hardly ever use the phone app any more. I use
#!/bin/bash
konsole --hold --geometry 500x400+20+20 -e /bin/bash -c "jkbms -p C8:47:8C:E2:8D:26 -P JK02 -c getCellData | sed -n '7,14p ; 31,33p ; 58,65p'"

Where jkbms is part of this suite. So I just click an icon on my computer screen and get the relevant information.


-
Thanks for updating. The delta V 0.158v is still indicating moderate state of being unbalanced. Aim for < 10mV at 27.7V Depending on how many hrs per day you can " float" at 27.7. On the average off-grid setup this may yet take weeks to months, but you have a useable system. Make sure delta V on your JKBMS is 10mV or less and balancing comes on when first cell is > 3.4V
 
0.9V is a big voltage drop?

Yes, it is.

I checked the leads - after hours of charging, they're barely warm.

Doesn't really matter, does it? The fact they are warm at all (10 amps at 4 volts = 40 watts) tells you something if you just pay attention.

I am watching the whole process like a hawk, you know.

Famous last words. Everyone watches like a hawk and find out that people with experience aren't tell you things for the fun of it, but BECAUSE IT IS IMPORTANT.

I get 10A - which is all that supply does, the leads are reasonably thick, they're not hot...
-
Ignore those with experience at your peril. You need better cables and connections. Do NOT set the power supply above 3.65v.
How much did those cells cost you? That might be the price of experience for you is why I ask.
 
I mean, obviously, I don't have enough bars to connect 8 cells in parallel, so some of them I connected with (reasonably thick) wires with (reasonably big) alligator clips. I have enough of those.

I would ignore you too if you told me this. You actually need to listen. Alligator clips have no place in top balancing cells.
 
Just in case you're wondering, all went rather well.
All cells are now quite balanced - have been for months (0.01 delta V, consistent, when charging at 27.8V).

No cells ever went over 3.65, even when top balancing at 4.
No animals were harmed during the process either (unless you consider a few swatted flies, it being summer ;·)
'saul goodman.
-
 
Make sure delta V on your JKBMS is 10mV or less and balancing comes on when first cell is > 3.4V
Emphasis on the > sign.

Anyone trying to balance here at 3.4v exactly (or float for that matter), is asking for trouble, because this is the knife-edge between achieving a full charge (no matter how long it takes), or stopping well short or even go into a "rest" state while the other cells are going higher. It all depends on the characteristics of the cells themselves at how they respond to 3.4v. Some might transition, while others will naturally fall back while straddling this knife edge - guaranteeing an an always out of balance condition for cells that can't make up their mind which way to go.
 
There is a REASON the guide explicitly states to set the voltage to 3.65v when not connected.
THIS ^^

Inexperience with variable bench chargers has killed many an expensive battery, and one returns to an accordion on the bench the next day because the owner doesn't even have a multimeter.

You can also kill a supply if the current knob hasn't been taken off the dead-stop just a little, so that when you do attach and there is no charging current - even if small - the cheaper supplies blow the crowbars assuming a dead-short because the supplies are not battery-chargers that take this into account.

There are *some* that are battery-charger qualified, but you have to look for them. Mastek makes a series of those (with and without battery charger ratings), but the consumer doesn't know and goes the cheapest route. Poof!
 
Just in case you're wondering, all went rather well.
All cells are now quite balanced - have been for months (0.01 delta V, consistent, when charging at 27.8V).

No cells ever went over 3.65, even when top balancing at 4.
No animals were harmed during the process either (unless you consider a few swatted flies, it being summer ;·)
'saul goodman.
-
Yes, I apologize since I've now read more of this thread and see I've already replied.

I haven't paid a lot of attention to the forum lately (since February) and somehow this was showing as a new thread for me.
Yes, my wife is from Kherson, so I've been a bit distracted (her twin sister was living in Kherson still, but managed to escape to Kyiv on June 1st).
 
The "runners" are the ones where level shoots up the narrow neck while others still have level at the wide part of bottle.
Nothing wrong with any of the bottles, just some started out less full than others.

BTW, great analogy. It makes it easier for people to understand and also demonstrates the rapid climb in voltage at the end.
 
Back
Top