A few questions in preparation for cells

[beckkl]

I ordered 16 304AH cells from Amy Zheng. I had already purchased the 48V Overkill 16S BMS. I had a couple of questions if anyone would have the time to comment:

1.) How do I size a class t fuse? Is 200A sufficient?
2.) What is the best place to put a fuse?
3.) Is putting some sort of smart shunt on the battery necessary? I see a lot of folks use the victron unit. I am mostly just looking for the easiest way to view capacity over LAN. I am guessing the growatt inverter or BMS would not be able to do this.
4.) Beyond top balancing, are there any other steps I should be doing before assembling the battery? Is a capacity test of each cell necessary or recommended?

 

[FilterGuy]

1.) How do I size a class t fuse? Is 200A sufficient?
2.) What is the best place to put a fuse?

The fuse size will be determined by your load current.... not battery size.
1) Since you have a 16S BMS I am assuming a 48V system.
2) For example purposes, I will assume you have a 3000W inverter that is 90% efficient.
3) That means the the battery will have to supply 3000W/.9=3333W when the inverter is maxed out.
4) That means the current at max continuous load if the battery is almost empty will be 3333W/48V=69.4A.
5) You never want to run a fuse at it's full capacity so you need to add 25% so the fuse size should be at least 69.4*1.25=86.8.
Note: A class T fuse is very fast acting. That means surge currents from inductive loads have the possibility of blowing the fuse. Therefor
25% is the *minimum* headroom I would advise.
6) I have not shopped for a 90A class T, but it might not be available so you will probably use a 100A class T.
7) *Any* wire down-stream of the Fuse that does not have an additional fuse on it must be large enough to handle the trip current of the fuse. (For 100A, that is a 4 AWG wire

I would advise a review of this paper:

DC Fuse and breaker Sizing and Placement

To get the paper, click on the orange button at the top of the page The attached presentation gives the fundamentals for picking fuse/breaker sizes and describes where in the circuit they should be placed. You can download the presentation by...

diysolarforum.com

3.) Is putting some sort of smart shunt on the battery necessary? I see a lot of folks use the victron unit. I am mostly just looking for the easiest way to view capacity over LAN. I am guessing the growatt inverter or BMS would not be able to do this.

NO. If your inverter or some other device can provide the data you want in the way you want it, an additional shunt is not necessary. In some cases, adding a shunt is necessary to get the the data you want in the way you want it. Since we don't know the other components of the system, we don't know if SOC is available over LAN from them.

Beyond top balancing, are there any other steps I should be doing before assembling the battery?

Only the obvious: Preparing whatever case you will use, preparing whatever compression (if any) you will use, Acquiring all the incidentals like wire, lugs, fuses and tools you will need. etc

4.) Beyond top balancing, are there any other steps I should be doing before assembling the battery? Is a capacity test of each cell necessary or recommended?

I used to ALWAYS do a capacity test on individual cells. However, after a while I realized the results did not change what I did next...which was to assemble the battery. Consequently I never do per-cell capacity tests any more... the extra data is not worth the extra effort to me (Many people have the opposite opinion on this...it is up to your personal value set). Heck, I have built whole batteries without even doing a battery capacity test before I put them into service. Granted, the data of capacity testing is interesting....but it rarely changes what I end up doing.

 

[beckkl]

The fuse size will be determined by your load current.... not battery size.
1) Since you have a 16S BMS I am assuming a 48V system.
2) For example purposes, I will assume you have a 3000W inverter that is 90% efficient.
3) That means the the battery will have to supply 3000W/.9=3333W when the inverter is maxed out.
4) That means the current at max continuous load if the battery is almost empty will be 3333W/48V=69.4A.
5) You never want to run a fuse at it's full capacity so you need to add 25% so the fuse size should be at least 69.4*1.25=86.8.
Note: A class T fuse is very fast acting. That means surge currents from inductive loads have the possibility of blowing the fuse. Therefor
25% is the *minimum* headroom I would advise.
6) I have not shopped for a 90A class T, but it might not be available so you will probably use a 100A class T.
7) *Any* wire down-stream of the Fuse that does not have an additional fuse on it must be large enough to handle the trip current of the fuse. (For 100A, that is a 4 AWG wire

I would advise a review of this paper:

DC Fuse and breaker Sizing and Placement

To get the paper, click on the orange button at the top of the page The attached presentation gives the fundamentals for picking fuse/breaker sizes and describes where in the circuit they should be placed. You can download the presentation by...

diysolarforum.com

NO. If your inverter or some other device can provide the data you want in the way you want it, an additional shunt is not necessary. In some cases, adding a shunt is necessary to get the the data you want in the way you want it. Since we don't know the other components of the system, we don't know if SOC is available over LAN from them.


Only the obvious: Preparing whatever case you will use, preparing whatever compression (if any) you will use, Acquiring all the incidentals like wire, lugs, fuses and tools you will need. etc


I used to ALWAYS do a capacity test on individual cells. However, after a while I realized the results did not change what I did next...which was to assemble the battery. Consequently I never do per-cell capacity tests any more... the extra data is not worth the extra effort to me (Many people have the opposite opinion on this...it is up to your personal value set). Heck, I have built whole batteries without even doing a battery capacity test before I put them into service. Granted, the data of capacity testing is interesting....but it rarely changes what I end up doing.

Thanks for the response and the link. I may just capacity test a couple at random to be sure they aren't a bad batch. But if one pulls 300 vs 304 I really am not going to go through the trouble. My overall strategy is to oversize the array so I don't have to sweat milking every last AH from the battery.

 

[FilterGuy]

My overall strategy is to oversize the array so I don't have to sweat milking every last AH from the battery

That is my approach as well. Doing cell level capacity testing would find a really bad cell earlier, but getting a cell that is so bad I won't use it is rare enough that I feel it is OK to not find it till late in the build process. Part of this comes from the fact that I buy cells from china and any realistic hope of doing a return for even a really bad cell is extreamly low....

 

[chrisski]

I bought 16 cells from Amy also. For capacity testing, that would take me three and a half weeks. 16 batteries charged at 3.65 at 10 amps take a long time to charge, and then individually capacity testing them, and then charging them again.

At best what a top balance would do would be group like capacity cells. She already sent a data sheet. I capacity tested eight 25 ah batteries and it took days. These larger cells have 10+ times the capacity.

IMO since all cells are going in the same battery for you, less of a factor than my build.

I put my fuses the first thing after the battery terminal as close to the battery as can practically be mounted. On my current RV build, the fuse is a couple feet from the Battery at the closest point i could screw the fuse holder in. The fuse is just before the battery cutoff switch. On my smaller portable solar “generator” the fuse is 12” from the Bartery, after a 50 AMP quick disconnect and before a master cutoff switch. For my lithium upgrade, the fuses will be inches after the BMS for each battery, and another larger fuse after both batteries are combined.

 

[beckkl]

I bought 16 cells from Amy also. For capacity testing, that would take me three and a half weeks. 16 batteries charged at 3.65 at 10 amps take a long time to charge, and then individually capacity testing them, and then charging them again.

At best what a top balance would do would be group like capacity cells. She already sent a data sheet. I capacity tested eight 25 ah batteries and it took days. These larger cells have 10+ times the capacity.

IMO since all cells are going in the same battery for you, less of a factor than my build.

I put my fuses the first thing after the battery terminal as close to the battery as can practically be mounted. On my current RV build, the fuse is a couple feet from the Battery at the closest point i could screw the fuse holder in. The fuse is just before the battery cutoff switch. On my smaller portable solar “generator” the fuse is 12” from the Bartery, after a 50 AMP quick disconnect and before a master cutoff switch. For my lithium upgrade, the fuses will be inches after the BMS for each battery, and another larger fuse after both batteries are combined.

Thanks. Yeah, its all going in a battery box of some sort so it shouldn't be a problem to mount the fuse close. This thing is going in an unheated, uninsulated pole barn in the woods in the U.P. of Michigan, so most of my mental cycles are going into designing an insulated box I can heat with a seed mat/heat pad. I'm thinking of slightly compressing the cells via a plywood box, and then setting that box in another plywood box that is 2" bigger on all sides. Then I could slide in some 2" foam during the winter months, and take them out in the summer.

 

[FilterGuy]

At best what a top balance would do would be group like capacity cells.

Do you mean top ballance or do you mean capacity testing.

Top balance doesn't group like capacity cells. Top balancing is done to make different capacity cells behave better at the top of the charge curve so you can get the most out of the pack....in many casses, the cells are already reasonabbly ballance so it does not make much difference, but if the cells are way out of whack, it can makes a substantial difference. Even though I rarely do capacity testing, I always do a top balance.

These resources are my feeble attempts at explaining top balance:
Beginners explanation of Top and Bottom Balancing
Top Balancing LiFePo4 Cells using a low cost benchtop power supply.

Just as an aside. I have noticed that in several of the @Will Prowse recent videos he has not top balanced his cells. Instead, he builds the battery and does a capacity test and only does a top ballance if the capacity test comes out bad. Even then he sometimes just lives with it. My guess is he is setting his charge voltage somewhat conservatively or he would be getting high voltage disconnects. The other possibility is that he is getting high voltage disconnects from the BMS but does not worry about them. Unless the cells are way way out of balance, the high voltage disconnect disconnect is going to happen when the SOC is nearly 100%... and with the BMSs he is using, the charging will re-start when appropriate....so who cares if the BMS is stopping the charge instead of the charger? I can't really fault that logic too much. The only downsides are 1) you might loose a couple of percent capacity till the BMS can get them ballanced over time and 2) if the BMS is stopping the charge on a cell over-voltage event, there is only one line of defense. (I am assuming the over-voltage is not set to a damaging level)

 

[chrisski]

Do you mean top ballance or do you mean capacity testing.

I meant capacity test. Not on my A-Game today. I agree on top balancing the batteries.

Amazing how when I use the wrong terms like that I derail the conversation.

I have not seen the video where he does not top balance. My only guess is if he uses an active balancer, or a BMS with an active balancer, but he did not like active balance earlier. I would think a good active balancer and being careful not to overcharge the bank or cells the first time a battery is assembled would be just as good as top balancing.

 

[FilterGuy]

I have not seen the video where he does not top balance.

Watch the recent video reviews of the falsely advertised 310Ah Cells. (he has several videos involving them). He plays with an active ballancer and then gives up on it. It is kinda funny to watch, he knows he should do the top balance but *really* does want to hassle with it. With as much storage capacity as he has.... I probably wouldn't bother either.

 

[beckkl]

Yeah, I mean I get it. These are a huge investment. Batteries historically were finicky and expensive, so taking EVERY precaution to get every bit of value out of them was important. I think now its fairly easy to just keep "leveling up". I really don't imagine that I will be pulling even 3K cycles from this before I replace them with whatever is the new thing. So I plan on setting my voltages a bit conservatively and just moving on to the next thing.