why 80p for powerwall?

wattwatt

Member
Joined
May 21, 2018
Messages
55
I notice a lot of DIYers online (including YouTube) make 80p cellpacks/modules(/strings?) for their 24v or 48v power wall batteries. Why is 80p so common?
 

Crimp Daddy

Well-known member
Joined
Feb 21, 2018
Messages
964
There is no hard or fast rule regarding this and its 100% preference / customizable.

Your parallel count should be determined by your capacity requirements and max load of the systems design.

Your series count should be determined by your equipment operating voltage range.

To answer your question, I suspect people just like to copy other people who have made systems and dont really take much of the above into consideration.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,422
If you figure max amp draw of about 500mA per cell, then 80p would be 40A. This is usually the max most ppl would draw from the strings. However, there are many who make larger packs. Mike has 400p, I think. There are several with 120p.

Also, it's usually based on pre-defined holders. If you get the 4x5 holders, then 1 would be 20p. Then 4 of them together is 80p. The next would be 100p. So this is probably another reason for this common arrangement.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,590
The 80p comes from that it basically is 10kwh. A round and even Number based on the cells available att that time. The size also did fit a bench om the depth. It hvae nothing todo with average or max current.

Then many People followed us when we started.
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,522
I've stumbled into the idea of building to a specific ah per pack rather than specific number of cells.

I shoot for 130ah per pack. In the early days I bought 2100mah cells it took 62'ish to make 130,000mah. Lately I've been buying2600mah cells where it takes 50 to get to 130,000mah. But all mypacks are the same physical size - so in terms of mah they are interchangeable.

Another reason for pack size is the physical space that will house them. As you can see in the picture below - my shelves are only so large.

I also used the 4 x 5 cellspacers as protection for the+ and - terminals... so a few spaces a 'wasted'for this as well...

image_wxqoxh.jpg

As you can see in this picture - the center pack (grey) is one of those 50cell packs and the one above (pink) and one below (blue) are of the 62'ish cell size.


However, as mentioned above - this totally up to you and I wish you good building :)
 

Oz18650

Active member
Joined
Oct 2, 2017
Messages
200
OffGridInTheCity said:
I've stumbled into the idea of building to a specific ah per pack rather than specific number of cells.

I shoot for 130ah per pack. In the early days I bought 2100mah cells it took 62'ish to make 130,000mah. Lately I've been buying2600mah cells where it takes 50 to get to 130,000mah. But all mypacks are the same physical size - so in terms of mah they are interchangeable.
By referring to the packs as interchangeable, it sounds like you are saying that you put the 62 cell packs into strings along with the 50 cell packs?

If so, the current per cell (stress on cells) will be higher for the 50 cell packs.

Common design is that each cell should not be providing more than 500ma of current.

Using that number, your 62 cell packs could comfortably deliver 31A of current.
If you have a string of 62P cells and you substitute in a 50P cell pack and draw 31A from the string, you will be taking 620ma per cell from the 50 cell pack.
 

wattwatt

Member
Joined
May 21, 2018
Messages
55
Interesting. I was starting to think 80p was the go to because it's as big as you can physically go before the pack started suffering structural issues.

Good point about the physical location of the battery - I don't even have shelves to hold 80p packs (16" to 18" deep).


Oz18650 said:
Common design is that each cell should not be providing more than 500ma of current.

Why is 500mA the suggested limit? Is this for continuous or for burst/surge/spike? If the former, what is the suggested max for burst/surge/spike?
 

Oz18650

Active member
Joined
Oct 2, 2017
Messages
200
crashintoty said:
Interesting. I was starting to think 80p was the go to because it's as big as you can physically go before the pack started suffering structural issues.

Good point about the physical location of the battery - I don't even have shelves to hold 80p packs (16" to 18" deep).


Oz18650 said:
Common design is that each cell should not be providing more than 500ma of current.

Why is 500mA the suggested limit? Is this for continuous or for burst/surge/spike? If the former, what is the suggested max for burst/surge/spike?



You can also "stack" the common 4x5 holders.
E.g. 3 deep and 2 high,
Or stand them so they have the 5 cells high and the 4 cells deep.

I did not easily find reference to 500ma per cell just now.
I believe there are several reasons to aim for ow current draw
- longer run time. if the cells can supply 2000mah and you draw 1A from each cell, then your pack will be flat I 2 hours, if you draw 500ma then the pack will last 4 hours
- less stress which equals longer lifespan for the packs
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,422
There's a Battery University link that shows that <1A draw increases life cycles. I don't recall the link, or which post(s) has it. This has specific relation to low drain (LD) cells, like those used in laptop packs and such. They are not designed to run at or greater than 1C for very long (1C = rated mAh capacity / 1000, generally speaking). A lot of LD cells are rated at .5C. So if a cell is rated at 3000mAh, then 1C = 3A, and 1.5A is .5C.
Generally speaking, the less aggressive the current draw, the longer the cell will last. 500mAh is a pretty good balance between usable power and size. Some DIY builds go as low as 250mA, and I think there are a few even lower. However, spike current could still be around 1A.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,590
The 500mA comes from a generic saying. If you test charge/charge at 500mA thats also your max current. If you test at 1A thats then your max. Its not based on any university page stating it prolongs life span either

Once again 80p choosen because of 10kWh and that was what Peter among others had in total cell count. The size of the pack was also workable to move around. Nothing special.

The max current to draw on above depends on the tested current.

There is no magic over this :)
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,522
Oz18650 said:
OffGridInTheCity said:
I've stumbled into the idea of building to a specific ah per pack rather than specific number of cells.

I shoot for 130ah per pack. In the early days I bought 2100mah cells it took 62'ish to make 130,000mah. Lately I've been buying2600mah cells where it takes 50 to get to 130,000mah. But all mypacks are the same physical size - so in terms of mah they are interchangeable.
By referring to the packs as interchangeable, it sounds like you are saying that you put the 62 cell packs into strings along with the 50 cell packs?

If so, the current per cell (stress on cells) will be higher for the 50 cell packs.

Common design is that each cell should not be providing more than 500ma of current.

Using that number, your 62 cell packs could comfortably deliver 31A of current.
If you have a string of 62P cells and you substitute in a 50P cell pack and draw 31A from the string, you will be taking 620ma per cell from the 50 cell pack.

>Common design is that each cell should not be providing more than 500ma of current.
I wouldn't disagree with this - but its not a fixed rule. This is more about testing on the cells you put into your packs. If you test at 1,000ma (and cells are designed for this) then 1,000ma is OK. I (like many others I'm sure) use 500ma discharge tests because that's what OPUS supports easily :)

>If so, the current per cell (stress on cells) will be higher for the 50 cell packs.
*Removing my original comments here about different packs discharging at sightly different ma rates/cell as this is 'incorrect logic'. Thanks to @Daromer for correcting the discussion in following posts.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,590
All the cells are in parallel so the 50p pack doesnt get more current per cell than the 62p pack gets unless you fucked up...
Think someone missed a lesson in ohms law there :D

Once again dont overcomplicate things here. There aint no magic behind or hidden things ongoing :)
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,522
daromer said:
All the cells are in parallel so the 50p pack doesnt get more current per cell than the 62p pack gets unless you fucked up...
Think someone missed a lesson in ohms law there :D

Once again dont overcomplicate things here. There aint no magic behind or hidden things ongoing :)
You're saying thatall cells (50 + 62 = 112 in this example) deliver the same ma/cell to meet the overall load on the battery - correct? I'm asking so I can correct my own thinking :)
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,590
Yes you are correct. As long as you didnt do something wrong with wiring so the resistance in busbars and wires causes issues.

those 50+62 is just like having a larger pack of 112... "basically" because once again there is of course resistance everywhere but for a normal user here that really doesnt change that much. The internal resistance in the cell causes larger differences in the same pack than between packs.
 

wattwatt

Member
Joined
May 21, 2018
Messages
55
WOW! I'm so glad y'all mentioned amps per cell - I totally forgot to double check my amps per cell calculations. I was going to build a quick & dirty 14s30p powerwall (3 kWh, 1+ hour of run time - simply because I'm so eager to try things out) and that would of been a disaster!!! Based on our average energy usage of 2.5 kWh that would of equated to 1.8 A per cell continuous draw and based on a "better safe than sorry" guesstimated surge/spike/burst of 8 kWh that would of equated to 5.6 A draw per cell. And as y'all said laptop batteries are low drain so they would not have been able to manage those draws.

I guess I'll stick to my original configuration of 14s300p (30 kWh, 10+ hours run time). This will give me 180 mA draw per cell continuous and 560 mA draw per cell during a surge/spike (refrigerators, furnace, AC unit, etc.)... are these safe draws or should I add more cells in parallel to get the surge/spike draw under 500 mA too?

Also, would a 1s300p config (20 cells deep x 15 cells wide) be physically unmanageable and/or a b*tch to troubleshoot/maintenance or should I simply throw four 1s80p packs together in parallel and call it a day??
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,422
crashintoty said:
Also, would a 1s300p config (20 cells deep x 15 cells wide) be physically unmanageable and/or a b*tch to troubleshoot/maintenance or should I simply throw four 1s80p packs together in parallel and call it a day??

Just all depends on what you need for you space. If you can put 300p on a shelf and you can handle it just fine, go 300p. If it won't fit on the shelf and/or you can't manage it, moving it around, lifting it, etc, then go smaller and put two strings in parallel instead of making 1 large one.

Personally, even if I can manage it, and it does fit the space, I'd go with 2 strings. This allows you to bring a string down to work on it if need be and your system is still operational. As daromer mentioned, N+1 configuration. This also allows you to add a 3rd string easily as you've already built for 2 strings to be connected in parallel.
The down side is, you'll need more bms monitoring (longmons if going Batrium) doing the N+1 route. But in the end, worth it, imho.
 

LithiumSolar

Administrator
Staff member
Joined
Oct 8, 2016
Messages
2,388
crashintoty said:
Also, would a 1s300p config (20 cells deep x 15 cells wide) be physically unmanageable and/or a b*tch to troubleshoot/maintenance or should I simply throw four 1s80p packs together in parallel and call it a day??

No, not at all. This is exactly what I'm doing and it works GREAT. I even wrapped mine with heat shrink and the compression of the shrink creates a rock solid pack.

There's some info in this video...
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,522
mike said:
crashintoty said:
Also, would a 1s300p config (20 cells deep x 15 cells wide) be physically unmanageable and/or a b*tch to troubleshoot/maintenance or should I simply throw four 1s80p packs together in parallel and call it a day??

No, not at all. This is exactly what I'm doing and it works GREAT. I even wrapped mine with heat shrink and the compression of the shrink creates a rock solid pack.

There's some info in this video...

45g/cell * 300cells = 13,500g = 30lbs. 30lbs is doable (my 'units' are 20lbs)but reaching an upper limit (for me) if you have to stoop over or do any kind of awkward placement/hookup. Another consideration is BMS balance - Batrium is perfectly capable of 1s300p balancing (in a reasonable time frame) but I'm not sure how many other BMSs are beefy enough.
 

Oz18650

Active member
Joined
Oct 2, 2017
Messages
200
daromer said:
All the cells are in parallel so the 50p pack doesnt get more current per cell than the 62p pack gets unless you fucked up...
Think someone missed a lesson in ohms law there :D

hmmm....
I think we are talking about different scenarios.
--
There was the section which talks about physical size and mah being the same for the packs and the word "interchangable".

"But all my packs are the same physical size - so in terms of mah they are interchangeable"

Where the word "interchangable" was used, I took this to mean that a 50P pack could and would be swapped into a string instead of a 62P pack (with the justification that this is ok, because each packs capacity is the same).

eg for 7S

-- 62P -- 62P -- 62P -- 62P -- 62P -- 62P -- 62P --

becomes

-- 50P -- 62P -- 62P -- 62P -- 62P -- 62P -- 62P --

In this "interchanged" scenario the 50P pack is delivering more amps per cell than the 62P packs.

I think you were talking about adding 50P packs to the 62P packs effectively making 112P packs.
 
Top