Two questions

pipedream

New member
Joined
Apr 22, 2021
Messages
6
I am very new to the forum so please let me know if I have posted this in the wrong section. I just had two questions relating to building and designing a battery pack:

1. is there a particular reason that packs are designed with the positive and negative leads coming from alternative sides (top and bottom) of the battery pack?
What are the downsides for building a pack with the positive and negative both coming from the top of a pack? below are examples of what I mean.



2. What would you consider is too low of a capacity when using 18650 batteries in a powerwall? Is there a general rule of thumb or is it just preference?
I have just reached enough 18650 batteries for a 7s 80p powerwall, however the capacities vary from 1400mAh - 2800mAh.
I have input the capacities into repackr to start balancing out the packs, however the deviation is approx 350 - 450 between packs. Would this be acceptable?

Apologies for all of the questions, I am still quite new to this.
 
>1 .... with the positive and negative leads
I go further and put + and - on opposite ends AND opposite sides of the pack (as in the bottom picture). This increases safety as its practically impossible to accidentally drop something or have a wrench make contact (when tightening screws) across the + and -. But it means you have to bolt things on each end - which is a little harder when packs are horizontal on a shelf.

>2. ... capacity
Its not about 'too low' capacity - its about % of original specs. If you have 1000mah speced cells at 1000mah then that's 100% capacity but of course it will take more physical space than 2000mah cells.

Setting aside physical space. Low capacity % (relative to new) means the pack will not last as long. I go down to 85% of original with 90% more of 'the average' and I'm nearing 700 cycles at 50% DOD on many packs with no issues. *There isn't really 'precise' data on all of this as temp, DOD, C, $, and you're design goals all play a part. I don't read about anyone going below 70% or even 80% of original for longer-term powerwalks - probably because we want the packs to last a long time :)

I do read about low % capacity cells being used in non-Powerwall apps... maybe things that aren't cycled daily.
 
Last edited:
1. is there a particular reason that packs are designed with the positive and negative leads coming from alternative sides (top and bottom) of the battery pack?
What are the downsides for building a pack with the positive and negative both coming from the top of a pack? below are examples of what I mean.

The main reason is to ensure the same electrical resistance, regardless of which cell the current flows through. Otherwise, the cells closest to the connector will receive/provide a somewhat higher current and degrade somewhat faster. The severity depends on the thickness of the busbar vs the amount of current flowing.
Here's a more detailed explanation... it's using parallel 12V lead-acid batteries, but the concept is the same.

But it's a not an absolute "must" thing. The practical effect can be greatly reduced by using beefy bus bars and/or limiting to low current charge/discharge. Eg my powerwall doesn't follow this rule, but I limit the current to just 1/3~1/2 the current the busbars are rated for.

2. I have input the capacities into repackr to start balancing out the packs, however the deviation is approx 350 - 450 between packs. Would this be acceptable?

So taking the average of your numbers, the packs have 168Ah of capacity, +/- 0.2Ah or so.
That's totally fine, as long as you don't try to fully drain your battery down to 2.5v or whatever.
Set the cutoff to 3.4v or so, and that tiny difference becomes totally inconsequential. The higher cutoff voltage also dramatically increases the lifespan (cycle count) of the cells, and is strongly recommended for powerwalls.

Apologies for all of the questions, I am still quite new to this.

No need, it's the raison d'être for this forum
 
The main reason is to ensure the same electrical resistance, regardless of which cell the current flows through. Otherwise, the cells closest to the connector will receive/provide a somewhat higher current and degrade somewhat faster. The severity depends on the thickness of the busbar vs the amount of current flowing.
Here's a more detailed explanation... it's using parallel 12V lead-acid batteries, but the concept is the same.

But it's a not an absolute "must" thing. The practical effect can be greatly reduced by using beefy bus bars and/or limiting to low current charge/discharge. Eg my powerwall doesn't follow this rule, but I limit the current to just 1/3~1/2 the current the busbars are rated for.
An electrical engineer friend gave me the opinion that at 200-500ma/cell - in an 100p pack w/6awg buss (1 cell on each side - so dual buss for 4x wide holder) - the affect of current flowing (charge/discharge) from terminals at one end vs 'thru the pack' are negligible and can be ignored.

My powerwall averages 200ma/cell with an absolute max of 400ma/cell - so this was not even a consideration (for me) and why I listed safety as the primary issue.

@ajw22 - I know you are technical, do you (or anyone) have idea at what ma/cell it becomes an issue?
 
Thanks everyone for the responses, definitely making more sense now :)
Building a modular design would be nice, however I am really struggling to source enough cells to add on more in the future. Plus I dont intend to use the powerwall for anything more than powering some lights and sockets for an offgrid shed.

I dont really want to check every cell in the database to determine their capacity % from new as this would take forever.
Instead could I ask what peoples lowest capacity cells are in their powerwalls just so I have an idea of a ball park?
Would 1500mAh as a minimum be acceptable? this should reduce the deviation per pack by a bit.
Also completely agree on the cutoff limit being around 3.4V.
 
An electrical engineer friend gave me the opinion that at 200-500ma/cell - in an 100p pack w/6awg buss (1 cell on each side - so dual buss for 4x wide holder) - the affect of current flowing (charge/discharge) from terminals at one end vs 'thru the pack' are negligible and can be ignored.
Will take a few days, but I'll try running a few SPICE simulations to see how the Amps get distributed in each configuration.

Of course, a practical long term test with a few hundred charge/discharge cycles would be required to see if it makes any difference to the cell health in the real world... if only there were a person who likes doing such tests... :unsure:

this should reduce the deviation per pack by a bit.
Don't worry about such small capacity deviation. The measured cell capacities aren't that accurate anyways.
Most people forego Repackr and just use the much quicker "100mAh bins" method when building such large packs, and the results are good enough.

I dont really want to check every cell in the database to determine their capacity % from new as this would take forever.
Instead could I ask what peoples lowest capacity cells are in their powerwalls just so I have an idea of a ball park?
The quickest and very accurate way to determine the health of a cell is to use a dedicated "IR checker" (Internal Resistance) tool. The ones built into the Opus and such chargers are basically crap.

Going just by capacity is often misleading. You could have a barely used 1200mAh powertool cell that initially had 1300mAh, and you could have a near dead 2000mAh laptop cell that initially had 3500mAh.
 
Back
Top