Battery pack for robot 4

[waveboy]

Hi,
We intend to use a 6S/6P AA LiPO cell array for our robot.
Is it best to have a block of 6 cells in parallel, and then put 6 of these in series?
Or is it best to have 6 series strings...and then put these 6 in parallel?

 

[3DDave]

Draw a schematic of the two choices.

 

[davidbeach]

If you open circuit one cell, how much do you lose in each case?

 

[jimkirk]

Six in parallel, then put those parallel sections in series.
Be sure to match cell voltages before connecting them in parallel. (You can put them in parallel with each one having a low value resistor (1 ohm or so)and letting the voltages equalize for a while. When the voltages are within a few millivolts of each other, THEN connect them in parallel.)

Use a Battery Management PCB to maintain cell charge balance , prevent under/over voltage...

Are you using battery holders or wiring them directly? Best to weld connections directly to cells. If you must solder, be as quick as possible.

 

[itsmoked]

What jimkirk said.

Are you scratch building these or just specifying a pack to buy?

And yes, you'd want a BMS Battery Management System included. You would need 6 of them if you went the wrong way with six series strings then put in parallel.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waveboy]

Are you scratch building these or just specifying a pack to buy?

We are scratch building them

Are you using battery holders or wiring them directly? Best to weld connections directly to cells. If you must solder, be as quick as possible.

We eventually wish to weld them. However, we would like to be able to disconnect a cell or two if necessary. So the permanence of the weld is an issue.

Draw a schematic of the two choices.

as attached

https://files.engineering.com/getfi...7&file=Battery_cell_array_connection_6s6p.jpg

If you open circuit one cell, how much do you lose in each case?

Now that's interesting. I will look into this. You loose less with the "six paralleled arrays in series" i would have thought?
------------------------------............................___________________________
By The Way, what are the methods that you know about, for diagnosing whether or not any cell(s) in a large series/parallel array of LiPO cells,
have become damaged , or "malfunctional"? That is, damaged such that the lifetime of the LiPO array is made shorter than it would otherwise be?

 

[itsmoked]

There aren't many ways of diagnosing individual cells without instrumenting each of them. A good BMS will let you know which battery of the pack is showing lower capacity. Then you have to diagnose that battery's parallel cells to find the culprit(s). If you're going to build the packs with new cells that you've correctly matched via charge/discharge tests and then you're not heavily thrashing the packs you shouldn't have any problem. If you're using used/culled cells then you'd be more likely to have issues.

If you can spare the room and using used cells I'd start with standard battery holders you can plug and unplug cells out of.

With many cells in parallel you need a fuse per cell so a failed cell can 'check-out' on its own.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jimkirk]

These folks have an interesting non-welding battery construction kit. Also Battery Management boards, cells, and related stuff...

https://vruzend.com

 

[jimkirk]

Also, if you're scratch building them, how are you sourcing your cells? If you aren't getting known new, guaranteed in spec cells, you'll want to evaluate them for capacity, internal resistance, & leakage, and then sort them to get roughly the same characteristics per parallel block of cells.

Something like this can help...

https://dc-workshop.com/opencart/index.php?route=product/product&product_id=50

 

[itsmoked]

You'll even need to do that with new ones!

Keith Cress
kcress -

http://www.flaminsystems.com