A Question of Balance

Kingfish

100 MW
Joined
Feb 3, 2010
Messages
4,064
Location
Redmond, WA-USA, Earth, Sol, Orion–Cygnus Arm, Mil
In the early days of my electric discoveries I balanced the small pack with a pair of Blinky bleeders. This was fine for 6 or 8 brick-sized batteries from time to time, but then difficult and cumbersome when the pack grew to include another 12-14 bricks.

In resolution I purchased the Version 4.0.8 of the LVC/HVC/Parallel Adapter last year (June 2010) just as I was starting the Going to California trek. The truth is that I didn’t have the means to complete the assembly on the road, and it remains boxed and unassembled to this day. The reality is that the last time I ran out of power was on the last day – just one block shy of the Amtrak station after traveling 173 miles.

When I returned to Redmond the pack was divided in half, electing to keep 18 bricks on the bike for commuting. Today I have a total of 72 bricks (15S24P) in preparation for the next big odyssey. With so many lil’ lipos the Question of Balance is an intriguing one.

A cursory study suggests that I can parallel the JST-XH connectors for the balancing application. The new BMS tendered by TTPacks appears to manage this with increased ability. However I cannot afford purchase a BMS for every brick, and even if I were to divide 72/4 to equal 18 – the cost is beyond scope. :cry:

So we return to the Question. Certainly I would like to level the voltages across the pack, and it is easy to understand that cells in series could be of different potentials. What happens though if we were to parallel these cells, first as groups of 3 or 4, or perhaps more? Should we limit this parallel process to the clustered series? Example: 5S1P x 9; with 3 in series it becomes 15S3P == 3 JST-XH connectors in parallel with 3 bricks each. What prevents us from tying all the cells into parallel? Would the resulting disparities of potential resolve themselves?

Admittedly padded clothing and a face shield might be required when creating this linkage if the individual cells were way out of balance. Though once hooked up, wouldn’t they float together in parity? What if we left them that way whilst charging/discharging?

If I ran in the fat of my pack, never too high and never too low with cells linked in this manner – would that preclude the need for a balancer?

Reference two interesting balancing concepts:

Deeply curious, quite possibly really dense… :roll:
…and somewhat Moody, KF
 
Are you referring to parallel balancing?
If so, this is what i have my 5s packs plugged into:

parallel_6x_1.jpg


And then the charge/discharge leads are all plugged into parallel harnesses too.
They are thusly, for reasons of charging, balancing, and discharging, large 20 amp hour packs. The cells keep each other even, and the pack voltage as a whole keeps itself even too.

The only thing that can throw them severely out of balance is if there is one cell in the parallel pack that is super wonky. In that case, i disassemble the pack and cycle each one to find the weak pack.
 
Yes! This is precisely the method I had considered with the 6X connection as indicated.

Therefore in my instance, I have my pack broken into three sections:
  • Front = 15S6P = 3x6 = 3 JST-XHx6 parallel harnesses which in turn could be combined into ONE. The pack is too far away to combine further.
  • Rear-Left & -Right on the Trailer, each as 15S9P = 3x9 = 5 JST-XHx6 parallel harnesses which could be combined into ONE.

A very edgy layout, no doubt. Are there many members that use this technique?

Muchly appreciative, KF
EDIT: typo
 
A lot of people with larger lipo setups do this, it's actually quite commonplace. Can't think of anyone of the top of my head tho.
RC guys have been doing it forever, i am sure there is an article on rcgroups.com about it :)

Here's a pic of my 10S 20ah pack..

20ahparallel.jpg


My 5S pack is the same deal, it doesn't sit in here though.
Pretty damn convenient, huh? :lol:
 
Dang yes it is! Geez, the more you learn the less ya knows! :D

I think I just grew a new wrinkle in the ol' gray matter.
Very pleased, KF
 
For sure, you don't want to go around connecting up packs that are much different voltages, protective gear or not.

I find at times that some folks worry too much about balance, fretting over a difference of .05v or less. .05 different is the same in my book, and ok to connect paralell. Admittedly, I consider the lipo to be my "entertainment budget" batteries, and will treat em a bit harsh. Just how I am with a toy.

A paralell harness like neptronix shows is often a very good way to go, particularly with a less that totally permanent battery setup. After getting em all pretty much balanced, you can add or subtract packs easily from such a set up. So you can have the huge pack for an expedition, and then next week unplug half of it for more normal daily use.

Once you do paralell a big pile of packs, balancing them that way can take nearly forever with a typical RC charger. They are designed with much smaller packs in mind. So once you do get out of balance eventually, you might want to break up the packs and balance charge one by one. Doing that once in a while will also help you identify the stinkers you want to weed out of the pack. Depending on your use, this could only be needed 2-3 times a year.
 
Thanks Dogman :)

For some reason when I think about this issue it reminds me of Apollo 13. And then I wonder how do they do it on the ISS? :?

I have the Battery Medic Booster kit though not the Battery Medic; HK has them on back-order. :cry:

Adding to the confusion is the choice between the MKS 3in1 Balance/Discharger System 6S and the Hobby King Battery Medic System 6S. HK says the latter is newer. Both are out of stock. However – this appears to be a convenient method in lieu of the Zephyr BMS design.

The choices for a strong balancer are few. :|

~KF
 
Great thread by a member also here on ES by the same name: http://www.rcgroups.com/forums/showthread.php?t=932319

Also a more elegant approach is use these from Digikey or the 5s version on a pcb from the shack: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=455-2252-ND

But yes, tying all your lipos together, if in similar condition, does a good job of balancing. Not sure what would happen if they were not in similar condition, but I'm sure it ain't pretty. I tend to keep my packs in sets of 18s2p and either carry multiple of this layout for longer trips or just one if staying local. And I don't mix individual packs between layouts.
 
number1cruncher said:
But yes, tying all your lipos together, if in similar condition, does a good job of balancing. Not sure what would happen if they were not in similar condition, but I'm sure it ain't pretty. I tend to keep my packs in sets of 18s2p and either carry multiple of this layout for longer trips or just one if staying local. And I don't mix individual packs between layouts.

If they're in different conditions, you will notice that one cell group tends to have notably less amp hours and drops to a lower voltage..
It is important to fully cycle large packs like this from time to time in order to notice this.

If you have a weak cell, it will wear the nearby cells funny. So yes, that's definitely a problem.
I think rotating your lipo every once and a while would be a good idea if you are using large parallel-serial packs for this reason.

And definitely.. testing the packs before you put them into parallel is imperative.
 
Hey KF,

If you haven't then I'd suggest you give David at Progressive RC a call. He is in Seattle and has the 6 to 1 parallel adapters plus, he does custom wiring. Great guy who goes out of his way for his customers, highly recommended.

ProgressiveRC
6906 15th Ave NE
Seattle, WA 98115
(443) BATTERY[443-228-8379]
info@progressiverc.com
http://www.progressiverc.com
 
newb said:
Hey KF,

If you haven't then I'd suggest you give David at Progressive RC a call. He is in Seattle and has the 6 to 1 parallel adapters plus, he does custom wiring. Great guy who goes out of his way for his customers, highly recommended.

ProgressiveRC
6906 15th Ave NE
Seattle, WA 98115
(443) BATTERY[443-228-8379]
info@progressiverc.com
http://www.progressiverc.com

I'm one step ahead of you partner... Order #005826 late last night :D
Nod to Nep for the lead...
Cheers, KF
 
A Dead-Simple Balancing Circuit :idea:

OK, so let’s say I get these balancing cables. The risk firstly is when a person connects the first two bricks together: Will it be a spark, a flash, a pop, a fire, or something benign and unnoticed?

To preclude this hazard, can we not conjure up a dead-simple leveling circuit that balances the disparity between cells in a calm professional manner? Why not replicate a simple resistive circuit much like the one that inhibits the spark when connecting the battery pack to the controller that assuages the surge of current as the controller caps feast to charge? :)

To undertake the study and solution, we need to comprehend a few basic features:
  • Balance the Brick by leveling the cells within the series prior to connecting to another brick.
  • Balance between Bricks in parallel using the same method. This method can continue ad infinitum until all bricks are leveled.
  • To build the basic circuit we need to know the limit of the wire; how much current can it take.
  • Also the limit of the battery cell to accept charge which should be more constrained than discharge.

    Which of these is greater: Cell charge or wire capability?
Match the resistor wattage to meet or exceed the limit of the system, say 150% current handling capability. Aside from the resistive throttling, what else do we need to worry about? The current flows bidirectional so a protective diode won’t help. We’re not looking for limiting the high/low voltage; if that were the case we’re already in deep trouble. We just want to balance within a reasonable physical constraint the differences between potentials, and move on.

I’m not an EE, so I need a bit of hand-holding in sorting out the math. The two units I’d like to construct is the X-cell brick balancer, and the brick-to-brick balancer; the difference between the two is purely functional in wiring.

Does this sound reasonable, or am I out of my tree … again. :roll: :)

One brick, two brick… KF
 
Just balance them with an RC charger before you hook 'em up!

I set my iCharger to balance everything to about 3.85v before i hooked it up in parallel.
There was a difference in cell voltage between 0.01v +/-.
Connecting them together in parallel on the discharge leads and then the balance leads produced no electrical bravado :)

Easy peasy.
 
I don't own an iCharger. I think a resistor would be cheaper :)

~KF
 
Do you have a balancing charger of any sort?
 
neptronix said:
Do you have a balancing charger of any sort?
Never.

When I first got into the hobby I had LiFePO4 with a BMS. Then I converted to LiPo and bought the blinkies because I already had a charger. Then I upgraded to Meanwells S- and SP-units. And now the HRP-600 series. I've left my packs connected and topped off from time to time. The only problem I had ever was on the first trip to California when I ran into a windstorm and drew the pack down to the LVC mark. That killed off 5 bricks.

I don't want to go through that again. And the Blinkies take too long. Can't get a Battery Medic cos they're on backorder. Can't source a reasonable BMS without falling back to cumbersome technology that doesn't support 15S. I am left with only Balance as an option. It's not rocket science; just simple math, yes? :)

  • A cursory review of JST-XH connectors suggests that the leads are 22-18 AWG, with the majority of posts leaning 18 AWG. Once source suggests that 18-guage wire can handle 18A @ 12V.
  • Most bricks can handle 2C charge, and if they are 5Ah, then we're looking at 10Ah. If each cell was max'd out to 4.3V, that would be 4.3 * 10 = 43Wh.
Am I on the right path? :)

~KF
 
Oh dear.
You need some kind of balance charger. Not just because they are great at balance charging and all that ;), but because they can also be used to analyze batteries too.
The factory balance is absolutely horrible, so each pack needs to be balanced from the get go.
A 1010b+ would be great since you could do double duty on those 5s packs.

The leads are about 24awg at the battery themselves. They would hopefully act like fuses if you connected significantly misbalanced packs to each other.

4.3v/cell will drastically shorten their life span for a small capacity gain. Don't recommend it.
 
You're misreading what I am explaining:

For calculations, I presume the worst. I do not charge to 4.3V/cell; I use 4.3V as the worst-case to determine potential maximum current :wink:

I have zero-interest in purchasing something I consider not meeting my requirements; the iCharger was ruled out long ago. That would be a step-backward when We need to focus forward. :)

Back to the engineering solution:
43Wh/1h = 43W. Closest overrated Resistor is 50W. Went to TheFind and sourced a few out of curiosity more to understand what industry uses a 50W resistor.

What is the relationship of resistance to current? If I want to slow the flow and yet dissapate safely, is this a 1:1 relationship?

Little help plz, KF
 
KF - Do you have any CellLog8s or other trusted monitors? Or DVM, buts that's just a hassle. If the cells are within .05V, packs within -+1V, it would be okay to hook up without a resistor.

Do you have any 24V meanwells? Maybe charge each pack to something like 23.4V (3.9v per cell) and see what the cells look like. If similar you're good to go.

I both admire and respect your decision not to use a balance charger. I only use mine to charge ALL the other crap in my house not ebike related. Like the nicad cheese grater. It had a wall wart that did not perform right. Sat in my attic for 5 years before its resurection. My $30 turnigy fixed it. Sold my wife on the benfits of the ebike hobby.

If not for the bike, maybe everything else around the house I recommend: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=15167 (out of stock atm) and http://www.sureelectronics.net/goods.php?id=59 (wow, these have got a little pricey). The 12V DC supply comes in handy for a multitude of tasks, but I digress...

Good luck and have a safe journey along the Left Coast! :mrgreen:
 
number1cruncher said:
KF - Do you have any CellLog8s or other trusted monitors? Or DVM, buts that's just a hassle. If the cells are within .05V, packs within -+1V, it would be okay to hook up without a resistor.

Do you have any 24V meanwells? Maybe charge each pack to something like 23.4V (3.9v per cell) and see what the cells look like. If similar you're good to go.

I both admire and respect your decision not to use a balance charger. I only use mine to charge ALL the other crap in my house not ebike related. Like the nicad cheese grater. It had a wall wart that did not perform right. Sat in my attic for 5 years before its resurection. My $30 turnigy fixed it. Sold my wife on the benfits of the ebike hobby.

If not for the bike, maybe everything else around the house I recommend: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=15167 (out of stock atm) and http://www.sureelectronics.net/goods.php?id=59 (wow, these have got a little pricey). The 12V DC supply comes in handy for a multitude of tasks, but I digress...

Good luck and have a safe journey along the Left Coast! :mrgreen:
No cellLogs. I have an S-350-27 CC that I think can be turned down pretty far, but that's a good idea - just zap them individually before connecting :wink:

Left Coast: No doubt about it. I've often heard some particular cities described as being like a bowl of granola: Remove the fruits and nuts and yer left with a bunch of flakes :lol: Maybe that's why I always ate Cap't Crunch with Crunchberries...

Checkin' fer cavities :shock: KF
 
Balancing Plug

I’ve gone ahead and drawn up this little schematic just for grins. Essentially it shorts all the positive pins together, equalizing the potentials between the cells. With two or more these can be tied together in prior to attaching to the parallel cable.

BalancingPlug0.png


The only thing missing is a fuse or micro-breaker. :lol:

What about an inline LED, one in each direction: The LED would light up until the potential is equalized. :idea:

Full of ideas, or at least full of something… :roll:
~KF :)
 
bump.

Also another question; forgive me ~ I am a noob:

  • Would you leave the balancing cables attached whilst in use? I am not talking about the entire pack, just the 6-to-1 bundles.
Thanks, KF
 
Kingfish said:
bump.

Also another question; forgive me ~ I am a noob:

  • Would you leave the balancing cables attached whilst in use? I am not talking about the entire pack, just the 6-to-1 bundles.
Thanks, KF

Any packs hooked to the 6-to-1 can not be series connected to each other, otherwaise you get KFF. Is this what you are asking?
 
number1cruncher said:
Kingfish said:
bump.

Also another question; forgive me ~ I am a noob:

  • Would you leave the balancing cables attached whilst in use? I am not talking about the entire pack, just the 6-to-1 bundles.
Thanks, KF

Any packs hooked to the 6-to-1 can not be series connected to each other, otherwaise you get KFF. Is this what you are asking?
Yes, exactly, thank you :)

I will need to re-strategize my thinking on how to maintain this in the field.

FWIW, I recently checked my extra reserve pack; this was actually the original pack, though after the road trip last year I cycled the batteries out with the “new” batteries and have been running those for close to a year. Anyway – this pack has seen more abuse and was last charged some months ago, although every cell was still within 0.03V within the 5S1P unit, and 0.08V between the high and low of all cells.

Am I worrying too much? KF :roll:
 
Back
Top