Calculating amps per 18650 cell

I'm still in the 18650 harvesting/testing stage but am starting to think about the layout of my system. I think I've watched every HBPowerwall video and every DIY Tech and repair video. And I'm sure my question was answered somewhere.

Using a 48v 4kwinverter,If I want my packs 80 cellsin a series of 14 packs (like Pauls), would the amps per 18650batterybe:

(4000/48) / 80 based on the cellcount in each pack

or would it be

(4000/48) / (80*14) based on the cellcount in the wholearray


Im thinking its the first equation,What is a good rule of thumb regarding amps per 18650?

Thanks!

If we base the current on your nominal voltage. Nominal is 3.7V*14 = 51,8

4000W/51,8 = 77A

So on 4kW you need 77A from the battery bank when at nomninal voltage. If we go by lowest voltage instead that may be down to 3V then you get:
4000/42= 95A.

And if we calculate that a 4kW inverter easy can do 8kW shorter times you need whopping 190A current from your pack!

Laptop cells should not go above 1A if not really needed. So basically your 80p packs are fine as long as you dont go above 4kW. Also note that you need beefy busbars for it.

Neither

It is the first one actually, but not quite accurate. 4kW is the output of the inverter. If you are using the full 4kW then the input to the inverter will be more than 4kW depending on its efficiency. Let's assume 5kW at 51.8V nominal. That equals about 96.5A and these 96.5A are then spread across the 80 cells. 96,5A/80 cells ~1.2A per cell.

OK thanks..
Thats what I was afraid of. Looks like Ill need more batteries, or rethink the system.

Yeah i never took the loss in account Though 5kW for doing 4kW is really really bad

If You mix cell capacities, You will need even more cells in parallel to keep individualcell current below 1 Amp.

Using a 80p pack I would keep pack current below 60 Amps.

ChrisD

ChrisD5710 said:

If You mix cell capacities, You will need even more cells in parallel to keep individualcell current below 1 Amp.

Using a 80p pack I would keep pack current below 60 Amps.

ChrisD

Click to expand...

So I should really use the Pack builder and start putting all my cells mAh into a spreadsheet? Doing an 80p14s setup 48v.

Packbuilder does not give you any upper advantage as such. I doubt you willl see a big difference but if you want the cell capacity 100% based on your numbers the repacker is a very good tool!!
But since i guess you used an opus or other tester that can diff 5-10% easy that sorting may not be 100% in the end

This is how I arrange all my packs:
Its not as exact as the Packbuilder here though. Its based on 100mAh ranges kind of.


And its also 14s80p unless you tend to make 80 strings of 14 cells in series

Sauron_Gorthaur said:

ChrisD5710 said:

If You mix cell capacities, You will need even more cells in parallel to keep individualcell current below 1 Amp.

Using a 80p pack I would keep pack current below 60 Amps.

ChrisD

Click to expand...

So I should really use the Pack builder and start putting all my cells mAh into a spreadsheet? Doing an 80p14s setup 48v.

Click to expand...

You can do that, but cells come with different mAh ratings as well asdifferent rated discharge rate.
I have seen 2200 mAh cells that is rated at 420 mA only. These cells will be able to supply the 1 Ampere You want but at a reduced capacity.
These cells will have tobe omitted from Your Banks, unless You are prepared to draw only 34 Amps.

In a previous post I suggested classifying the cells depending on their remaining life, but I forgot to include the rated Discharge Current as well.

Never mind, nobody thought that was a good idea, but You are free to use my input.

What I am trying to say is, You can indeed use the cells rated at only 400 or so milliAmps but You need to put at least 200 of these cells in parallel if You want to safely feed this 4KW inverter.

best of luck

ChrisD


Strange, this Board system will not let me enter two individual answers, it insists on listing them in succession inside one answer, therefore this next part was meant as a separate answer to daromer:

---

daromer said:

Yeah i never took the loss in account Though 5kW for doing 4kW is really really bad

Click to expand...

Yes, that is bad, and that is because there is a small error in YourMath.

Look here:

Thelabel on the PIP 4048MSD says:

Inverter Mode

DC input 48 VDC 93 A
AC output 230 V 22A, 1 phase.

DC input 48V*93A= 4464 Watts

TheAC output of 230V 22Ais in VoltAmperes, NOT in Watts.

Rated output in Watts is 4000.

4464 Watts in gives You 4000 Watts out (4000 * 100)/4464 = 89,6 % conversioneffectivity.

I do not think10 % conversion loss, is allthat bad. MPP has hardly used any Rocket Science when they designed their Inverters. Maybe that is also the reason for all these cooling fans they have put inside

Though, I still think the PIP inverters are reasonable value for money, if wewant a more efficient inverter, we must be prepared to shell out a lot more money.

ChrisD

And the solar power we put in to the inverter is (almost)for free in the first place, so who cares about a few %'s more loss in the inverter, if it is a cheap(er) one.

He never said that this is actually a thing, it was just a reply to what I said because I factored in the conversion losses and he didn't. And I did that by assuming 80% efficiency. You know, I don't know the efficiencies of all inverters on the market from the top of my head So I knew it will be off, but it didn't matter as an example, because that only leads so light oversizing of the battery and that is never a bad thing. So 80% is always viable for rough calculations and it is very easy to calculate as well.

So max is about 1A, for longer life what is a good value? Joe fused his cells with 0.5A fuses.

What are decent upper/lower voltage ranges for longer life and less stress? max is 2.8-4.2?

Looking at this pic and considering I am using a mix of old cells, 3.6 or3.7 might even be the lower cutoff point?





[edit] Fixed image

Wrong image?

3.4 to 3.96 as min depth of charge
I go 3.2 to 4.05 or so

If you stop at 3.6 under load you have not used much at all. You need to calculate a voltage drop. Lets say you havr 14 s and set low limit to 3.6V. thats 50.4

So now lets remove some voltage drop. You will have up to 0.2v per cell and 1v on the wires unles you go really big. That would be 4.8V in total drop. Thats down to 3.8V. Thats 3,87V per cell. And then you havent used much on the cells

Important to diff on resting voltage to loaded voltage.

Just an FIY

For windy or days where you can say for sure you got less sun you might want to be able to utilize larger DOD so always design so you can go full out if needed.

Bigger wires means less voltage drop. Is bigger always better? Are there any electrical downsides to oversized wires?

Some cells have 2.75 in the specs others 3.0 as min voltage.

As for design, lets get the first string up and running first (thinking about 14s80p although the p's are not set in stone) and then expand with another string in time to be able to draw more current and/or have more kWh's.

There will be a point where there is diminishing returns. As in, too stiff wire to bend, too expensive, over-kill. So yeah, you can go to big. Electrically, no, not really an issue. But there is a peak point where going larger doesn't make any negligible difference in drop.

As Korishan said its just about how much you want to spend on it. Generally its a cost factor in end.

By now, You have gotten a lot of suggestions to base You decisions upon.

Each suggestion is made based on the conditions each contributor is operating under, so instead of telling You my findings, I will give You an idea to how You can make Your own decision based on exactly the cells You have and the conditions, You want to use.
This suggestion may not be better, just different.

You need: 2 cells of the brand/make/model You want to utilize, one Foxnovo 4s charger / Opus / IMax or similar.
One power supply with individual adjustments for voltage and current, f.eks this DP 50V 5 A progrmmable step down power supply" and one loadbank say a "DC 5V USB Battery Capacity Tester / Battery Life / Internal Resistance Analyzer" or the larger model "CC Electronic Load 60W Batterie Discharge Kapazitt Tester Detection DC12V TE553".
On these load testers You can adjust the discharge current as well as the cut-off voltage.

Now You decide on what voltage range You want to check, say 3.3 to 3.85 Volts.

Proceed as follows:

Using Your Foxnovo/Opus/ Imax charger, fully charge Your 2 cells to the rated 4.21 Volts, make a capacity test but do not recharge Your cells. Just record the mAh recordings for reference, this is the max power You will be able to get.

Now use Your power supply set to 3.85 Volt and .5 A to charge both cells to 3.85 V.

Adjust Your Electronic load unit to the wanted discharge rate, say 500 mA and adjust the cut-off voltage to the 3.3 Volts You have settled upon. Now the load bank will tell You exactly how many mAh You will get from exactly the cells You have.

Note this is the maximum You will get, not taking any wiring into considerations.

Your findings may be a surprise for You, showing much lower values than You initially thought, but this is really WYSIWYG

Is Your findings too low, look for what options You may have at hand, can You increase the charge voltage to say 4.0 V? Or can You go for a lower discharge cut-off voltage?

Best of luck

ChrisD