Another Top Balance Screwup

[Bob B]

Sometimes it's because people set the power supply to 3.65 volts after connecting to the cells. The power supply needs to be set to 3.65 volts before connecting to the cells.

This is the most common thing people seem to do ..... they don't understand why it matters to only set the voltage with the leads disconnected. They just want the current flow to be greater.

Maybe you could make a video of what happens when you do this .... and show how the voltage will come up as the current drops off and the battery voltage comes up. So, demonstrate why doing it wrong results in damaged cells .... why the voltage drops when the power supply is connected to the battery, show how much difference it makes if you use the tiny crappy alligator clips and tiny wire that comes with many of the supplies .... so by making better leads, the current increases without cranking up the voltage on the power supply.

 

[dRdoS7]

Hi,

Yeah, sorry about deleting the posts. I need to think of a sticky post or video that can solve this problem forever.

I am not sure how I can help here.

The old adage about leading a horse to water?

You're wasting your time trying to stop people making mistakes. Besides, that's how to learn lessons, weed them out, and advance the rest.

dRdoS7

 

[GSXR1000]

Or maybe it's not my responsibility? I always blame myself when people make mistakes on here. I feel like I should have a video covering every topic possible. I might need to calm down and understand that mistakes do happen.

If you guys think that something I post here is causing a problem, let me know. I really want people able to build their own systems, and understand how every part works. But sometimes that's not possible.

people only half listen.. you have in your videos turned ps past 3.65 to "speed things up but not a good idea"
I world stress a top balance at 10 amp power supply could take up to 1 day per 270ah cell in parallel and you won't see much of a voltage change for 80% of that time... too many think that it isn't working

 

[FilterGuy]

Even with crappy leads, if you set the Power Supply to 3.65V (or whatever your target voltage is)....it will eventually get top balanced. Crappy leads will just make it slower.

It is easy to say to go buy a better power supply, but for many people they don't see a need for it beyond the one time of top balancing so they are hesitant to purchase a $200+ power supply and frankly, I can't blame them. Particularly if they can replace the leads and do just fine.

Even with the best leads, there *will* be a voltage drop. Furthermore, to get leads that wont have a healthy voltage drop between the power supply and the cells, you will probably have to replace the leads on even really good power supplies..... The reason is that the biggest resistance is at the connections. At 7 or 8 amps, a really good lead with an alligator clip will probably be worse than a crappy lead with a ring lug.

I'll say it again: The most important tool for top balance is patience.
If you don't understand ohms law and are just following directions, the 2nd most important tool is faith that the people that say not to turn up the power supply know what they are talking about.
If you think you won't forget because you are "smarter than those dumb-$#|+s on the forum" ... I can't help.

 

[Sojourner1]

I'll say it again: The most important tool for top balance is patience.
If you don't understand ohms law and are just following directions, the 2nd most important tool is faith that the people that say not to turn up the power supply know what they are talking about.
If you think you won't forget because you are "smarter than those dumb-$#|+s on the forum" ... I can't help.

Great advice!!!

 

[Metalophile]

Not that I recommend this, but one way I sped up top balancing was to wire up my battery in its final 8S configuration and charged the pack using a large solar panel. I depended on my BMS to cut off when one cell reached 3.65V. I did watch the individual cell voltages like a hawk once it got close to make sure the BMS did its job. After that I did the top balancing bit after the cells were at a high state of charge. It did save me some time, especially since I didn't have a good constant low voltage power supply capable of putting out a lot of current.

 

[GSXR1000]

Not that I recommend this, but one way I sped up top balancing was to wire up my battery in its final 8S configuration and charged the pack using a large solar panel. I depended on my BMS to cut off when one cell reached 3.65V. I did watch the individual cell voltages like a hawk once it got close to make sure the BMS did its job. After that I did the top balancing bit after the cells were at a high state of charge. It did save me some time, especially since I didn't have a good constant low voltage power supply capable of putting out a lot of current.

I would goto 3.5 (14v pack) and then switch to top balance.. that 3.65v cell could but higher ah then top balance stop right at 3.65 and the other 3 have slightly less..

 

[John Frum]

I would goto 3.5 (14v pack) and then switch to top balance.. that 3.65v cell could but higher ah then top balance stop right at 3.65 and the other 3 have slightly less..

Why do you think that?
The bms measures voltage at the terminals.
Probably more than fast enough to be very deterministic.
State of charge correlates well to voltage at the extreme ends of the curve.
Absorption doesn't really apply when you measure at the terminals.
And at reasonable current levels consistency is more important than actual mv.
To clarify its more important that the cells are charged to consistently close to 3.65 than it is for them to be as close as possible to 3.65 volts.

 

[Luthj]

. I do not understand how someone could over charge cells when using a psu set to 3.6V

It's simple the PSU was not set at 3.6 V. No other questions or answers need be provided. As I explained in the beginning of this thread the person who did this was aware of the risks and opted to take them because they were in a rush.

Honestly there's only so much you can do. There are 100 ways to cause a fire or injure yourself when working with things like batteries. You cannot possibly educate somebody on all the details. You either need to follow the directions or take the weeks and months it takes to fully educate themselves.

Example is a user posted up a thread recently where they sat on top off a battery bank and damaged a terminal. In addition to this they covered the top of the battery pack with a piece of wood which started on fire.

We can Emphasize the danger of doing this type of thing, but there's only so much that can be done.


Even in this overcharge event, there really isn't any significant risk of fire even at five volts.

As for toms, still waiting for any references for your claims.

 

[DerpsyDoodler]

The only way to make parallel top balancing more foolproof would be to insert a disconnect device that would disconnect above 3.65 volts.

I personally didn't parallel top balance mine ... I balance with the pack fully built and the BMS fully active.

You mean like a BMS?

 

[GSXR1000]

Why do you think that?
The bms measures voltage at the terminals.
Probably more than fast enough to be very deterministic.
State of charge correlates well to voltage at the extreme ends of the curve.
Absorption doesn't really apply when you measure at the terminals.
And at reasonable current levels consistency is the more important than actual mv.
To clarify its more important that the cells are charged to consistently close to 3.65 than it is for them to be as close as possible to 3.65 volts.

partly because not a real difference in top balancing between 3.5 and 3.65.
you can have 2 SOC at the same voltage. it would be a bigger issue at 13.6 (3.4v) because it is so flat there... think of it like water. at 32F it can be ice or water depending on energy it contains

 

[John Frum]

partly because not a real difference in top balancing between 3.5 and 3.65.
you can have 2 SOC at the same voltage. it would be a bigger issue at 13.6 (3.4v) because it is so flat there... think of it like water. at 32F it can be ice or water depending on energy it contains

You lost me.
I thought temperature was ultimately a measure of energy.

 

[Luthj]

3.5V is perfectly fine at low C rates. That should get you within 1%. For those with enough experience, series charge with bms till cutout, then top off any low cells.

Remember, for every screwup, there are hundreds of trouble free experiences.

 

[Hedges]

You lost me.
I thought temperature was ultimately a measure of energy.

There's energy required to melt a crystal structure. Some materials have a distinct melting point; Water, for instance, melts at 0C and both ice and liquid water coexist at that temperature. But the energy to melt ice is about he same as energy to raise water from 0C to 100C, or to go from water at 100C to vapor. So as you pump in energy, ice warms up to 0C then sits there for an extended time as it melts. Same happens at 100C until enough energy has been put in to converter it all to steam. So ice is a good way to keep food at 0C, and water is a good way to keep a (non-pressurized) cooling system at 100C.

Good old 60/40 solder has a well defined liquidus temperature. Lead-free alloys have a wider temperature range.

At a particular pressure, water, ice, and steam coexist.

Triple point - Wikipedia

en.wikipedia.org

 

[John Frum]

There's energy required to melt a crystal structure. Some materials have a distinct melting point; Water, for instance, melts at 0C and both ice and liquid water coexist at that temperature. But the energy to melt ice is about he same as energy to raise water from 0C to 100C, or to go from water at 100C to vapor. So as you pump in energy, ice warms up to 0C then sits there for an extended time as it melts. Same happens at 100C until enough energy has been put in to converter it all to steam. So ice is a good way to keep food at 0C, and water is a good way to keep a (non-pressurized) cooling system at 100C.

Good old 60/40 solder has a well defined liquidus temperature. Lead-free alloys have a wider temperature range.

At a particular pressure, water, ice, and steam coexist.

Triple point - Wikipedia

en.wikipedia.org

I must have missed this day at the academy of floral design.
I'm off to wikipedia.

 

[GSXR1000]

You lost me.
I thought temperature was ultimately a measure of energy.

A total of 334 J of energy are required to melt 1 g of ice at 0°C, which is called the latent heat of melting. At 0°C, liquid water has 334 J g−1more energy than ice at the same temperature. This energy is released when the liquid water subsequently freezes, and...

that is why we use ice and not cold water in ice chest

 

[Bob B]

You mean like a BMS?

Can you provide us with a link to a BMS that will work with the pack on a parallel 3.65V configuration?

 

[Hedges]

Can you provide us with a link to a BMS that will work with the pack on a parallel 3.65V configuration?

My thought is, assemble your 4s, 8s or whatever pack.
Wire it to keep a relay closed. Use the relay in series with a CV/CS supply set to 3.65V or less, wired to a single cell.
You now have "Belt and Suspenders". If BMS gets unhappy with any cell voltages, it disconnects the single-cell charger.
Top-balance one cell at a time. This won't take any longer than top balancing all cells in parallel (at least as far as machine time; you may lose time getting around to moving charger to next cell.)

 

[Bob B]

My thought is, assemble your 4s, 8s or whatever pack.
Wire it to keep a relay closed. Use the relay in series with a CV/CS supply set to 3.65V or less, wired to a single cell.
You now have "Belt and Suspenders". If BMS gets unhappy with any cell voltages, it disconnects the single-cell charger.
Top-balance one cell at a time. This won't take any longer than top balancing all cells in parallel (at least as far as machine time; you may lose time getting around to moving charger to next cell.)

That would probably work .... except some BMS will cut off if the differential between cells gets too large. I'm not sure how prevalent that feature is, but would have to be considered.

 

[Hedges]

That would probably work .... except some BMS will cut off if the differential between cells gets too large. I'm not sure how prevalent that feature is, but would have to be considered.

Or how big a differential.
You would want to get all cells up to the start of the knee by charging in series, which is faster. Then bring each individually up to some top-balanced voltage.