12 hours ago, plonkster said:

The inverter has built-in fuses for that purpose, at least according to the manual. They still recommend that you put in overcurrent protection. It's up to you, I personally doubt that it makes much sense to spend extra money on the fast-blow fuses. When the axie wants to blow its FETs it will do so, likely without even exceeding current anywhere else.

I disagree with that. With parallel strings, the current doesn't divide equally between the two strings. Each string contributes relative to its own capacity, you may find that sometimes the one string provides more than the other and they might even switch roles during the discharge cycle. In other words, two 125A fuses don't add up to 250A. To avoid nuisance blowing, those two fuses (and the cables) should be rated at slightly more than half the total current. The main fuse should be sized for whatever is downstream. It's perfectly acceptable for the main fuse to be lower than the sum of the two smaller ones.

Also, with two strings a short in the one string can cause current from the one string to sink into the other one, and the string fuses also protect against that. You could size all three fuses at 125A and that would be perfectly justifiable.

Thanks, this is perfectly clear. I agree the 2 fuses do not add up to 250A as its a parallel setup.

 

3 hours ago, Chris Hobson said:

I settled for a 100A fuse. There is a bit of spare capacity in the x25% calculation. In my mind 125A is too high  50% over your max of 83A. I have 100A here in the Karoo and they have not blown in 2½ years constant use. I do like your Mersen battery disconnect.

This was my thinking since the last post i made. I'll get the 100A fuse per string.

I liked the Mersen as well as it served 2 purposes... fuse and disconnect

 

12 hours ago, plonkster said:

It goes between the negative busbar and the fuse, in other words, it goes BEFORE the fuse. Let me explain why you do this and why it is not dangerous.

1. The thin red positive wire that runs between the shunt and the battery positive has its own fuse to protect it. This line is used only to power the BMV and do voltage measurements.
2. There is a fuse after the shunt that protects that path.
3. If you put a fuse between the shunt and the battery, the voltage drop across the shunt will affect the BMV measurements.

I hope that helps.

Like this

21 minutes ago, PurePower said:

Like this

Prezactically.

One more thing.... i understand that the cables lengths per string must be equal up to the busbar... what about the cable between the busbar and the inverter? I'm guessing that it should also be equal?

1 minute ago, PurePower said:

what about the cable between the busbar and the inverter? I'm guessing that it should also be equal?

Doesn't matter. The paths must be equal length. Think about the path that each electron travels from the battery negative pole down the cable through the inverter other side out up the red cable to the positive terminal. When there is more than one battery string, the total length of that path must be the same for both, otherwise one string works harder because those electrons have less "distance" to go (aka resistance).

When there is only one path (between the busbars), there is no need that they be the same length.

4 minutes ago, plonkster said:

Doesn't matter. The paths must be equal length. Think about the path that each electron travels from the battery negative pole down the cable through the inverter other side out up the red cable to the positive terminal. When there is more than one battery string, the total length of that path must be the same for both, otherwise one string works harder because those electrons have less "distance" to go (aka resistance).

When there is only one path (between the busbars), there is no need that they be the same length.

That was very well articulated. Thank you!

Very interesting post this.

Can I just get some clarification on fuse sizes. With a 5000W inverter and a 2 x 150AH 48V battery bank (300AH), I want to fuse the individual strings and also the main feed using a disconnect switch/fuse. If the inverter at constant full load pulls ~100A I should make the fuse ~125A to compensate for temperature etc?

Here is the question, with the inverter's capability to 'burst' to 10kW, should I not make the main fuse 200A and each individual string 125A each? Cables used are 35mm2 if I remember correctly.

My thinking is that in a dooms-day scenario there will be much more than 200A flowing in a very short time period which will hopefully blow the fuse(s) near the fault condition.

Thanks,

Ingo

On 2018/11/08 at 9:40 AM, Ingo said:

Can I just get some clarification on fuse sizes. With a 5000W inverter and a 2 x 150AH 48V battery bank (300AH), I want to fuse the individual strings and also the main feed using a disconnect switch/fuse. If the inverter at constant full load pulls ~100A I should make the fuse ~125A to compensate for temperature etc?

Yes

On 2018/11/08 at 9:40 AM, Ingo said:

Here is the question, with the inverter's capability to 'burst' to 10kW, should I not make the main fuse 200A and each individual string 125A each? Cables used are 35mm2 if I remember correctly. 

I think the "short burst" capacity normally refers to the AC throughput - I know that my inverters immediately switch to grid if I exceed 4kW on battery.

If you want to go the 3 fuse route I would make them all 125amp - having said that I would love someone to explain the logic of fusing individual strings in co-located parallel strings. I would use the shortest possible cable of the same length from each string to a single fuse or DC Breaker. If the strings are far apart then sure, fuse each string, but rather don't locate them far apart if it can be avoided.

Edited November 9, 20187 yr by pilotfish

12 hours ago, pilotfish said:

explain the logic of fusing individual strings in co-located parallel strings

I think it has to do with a short developing in one cell of one string. Now the other strings are dumping current into this string because it is now at a lower potential. In particularly severe cases -- and one can probably debate how likely that is but I'd expect from three strings upwards it becomes quite possible -- you could be cooking and heating batteries in that bad string. It would be better if there was a fuse that could disconnect just that one string. That's the idea at least, the theory.

In a BYD rack for example, each module has a 85A breaker, but the rack as a whole would have another 125A breaker. Same idea :-)

Victron has a busbar system with individual fuses for exactly this kind of setup. Of course it isn't exactly cheap, but it explains the general idea. Jono modifies a "Battery In" unit here, from around 7:00, to use fuses. There is also a distributor module that also has LEDs to show blown fuses.

 

On 2017/11/03 at 11:58 AM, PurePower said:

Thanks, this is perfectly clear. I agree the 2 fuses do not add up to 250A as its a parallel setup.

 

This was my thinking since the last post i made. I'll get the 100A fuse per string.

I liked the Mersen as well as it served 2 purposes... fuse and disconnect

 

Like this

OK so here goes nothing I hope not...

I did my install as per drawing above, (without the mid-point) I had only 4 x 12v = 48v = 200amp battery bank and added a second bank to this with busbars the works, with bmv-702 battery monitor with mid-point and ha02 battery balancer as well, my batteries DOD was normally between 20-25% for the 4 battery string, now that I have added the second string my DOD is still between 20-25%

I have set my bmv to 400 amps, inverter to charge at 40 amps

the question is should my DOD not now be less as I have doubled my total battery amps, what am I missing here, (still a nooby here and still learning new things everyday)

2 hours ago, Samsed said:

the question is should my DOD not now be less

it should be. Maybe you can have a look at the Ah consumed as measured by the bmv. This should give you an idea what is going on. Maybe your consumption has increased. Another possibility is, that the batteries were not fully charged.

11 hours ago, Fuenkli said:

it should be. Maybe you can have a look at the Ah consumed as measured by the bmv. This should give you an idea what is going on. Maybe your consumption has increased. Another possibility is, that the batteries were not fully charged.

Thx Fuenkli,

found the problem, my bmv did not take the setting from 200amp to 400amps when I initially change the settings