This video explains it well.

 

Thanks @razzor13bt

That’s an interesting video. Some takeaways would be that for large battery banks connected in parallel (eg rack mount), and where maximum discharge performance is required, a busbar is the way to go (and where the demands on the bank are quite low, a standard parallel wiring is fine). For a two-battery setup where one isn’t sailing close to the wind in terms of discharge it seems that a straight battery 1 to battery 2 connection can be used (not even using a loop back from the second battery), with no ill-effect, and still get the maximum capacity out of the battery bank. On his string of batteries the first two were shown to be fully charged and the next three dropped off considerably with their respective SOCs (at that point in time). Also interesting explanations around internal resistance. 
 

But - back to the wiring diagram in my earlier post (last one on previous page) and doubling up on 35mm cables between DC disconnect and the first / last battery respectively, is that the way to go then for this 2-battery setup paired with a 8KW SunSynk? 

10 hours ago, Kalahari Cruiser said:

Would this work - given the Sunsynk battery connector size limits? (now with new neg to first battery, and new pos return from second battery). Or just nonsensical?

 

I see it working as long as the + and - of each battery is not connected to the same point at the bottom of each fuse as per drawing. Hahaha

Only comment. I still like each battery to have it's own fuses to prevent one battery feeding into a fault be it inverter or other battery. 

11 minutes ago, Scorp007 said:

I see it working as long as the + and - of each battery is not connected to the same point at the bottom of each fuse as per drawing. Hahaha

Oh dear. Yes well spotted @Scorp007, was a very hasty mockup earlier, of course no mixing of + and - at the fuse  

 

The video that @razzor13bt shared earlier also had protection between the batteries. What fuses (and ratings) are typically used for that? I've not seen this in real life to know. 

9 minutes ago, Kalahari Cruiser said:

Oh dear. Yes well spotted @Scorp007, was a very hasty mockup earlier, of course no mixing of + and - at the fuse  

 

The video that @razzor13bt shared earlier also had protection between the batteries. What fuses (and ratings) are typically used for that? I've not seen this in real life to know. 

I would use 4 fuses where you have 2. Each battery on it's own +&- and then bridge the 2x+ and 2x- at the top before they go to the inverter where your 70 mm^2 go to the inverter. Size of fuse 1 size higher than the max charge/discharge current for each battery. Should a fuse blow on 1 battery the fuse would also blow on the 2nd battery and thus you will be forced to investigate knowing there is a problem. Each set then supervises the other. 

Thanks very much for your advice and explanation, @Scorp007

23 hours ago, Kalahari Cruiser said:

Oh dear. Yes well spotted @Scorp007, was a very hasty mockup earlier, of course no mixing of + and - at the fuse  

 

The video that @razzor13bt shared earlier also had protection between the batteries. What fuses (and ratings) are typically used for that? I've not seen this in real life to know. 

Hi Kalahari Cruiser,

your installation looks good, it will work, but it's a terrible overkill. You mentioned 200A fuses. Be aware that those wont protect anything. Remember what I posted in my previous post: A 200A fuse will stand 300A for approx 20 minutes. But your 100Ah packs probably wont. If you have a short on the 48V bar in your inverters have your fire extinguisher at hand, the 200A fuse will not prevent anything. The max fuse that would perhaps protect is 100A.

The double wiring with 35mm² is overkill too. I wrote in another thread that such attempt to keep all leads at same length is not necessary. The difference of internal resistance between packs due to manufacturing inaccuracy is likely to be higher than the difference of resistance in leads of different length, even with 25mm².

Sorry to say that. I wish you all the best.

10 hours ago, Beat said:

You mentioned 200A fuses. Be aware that those wont protect anything.

I think you misunderstand the purpose of the fuse.  It is ONLY there to protect the wire.  The correctly sized fuse for SANS rated wire will 100% guarantee that the fuse will blow before the wire insulation breaks down/catches fire.  That is its one and only purpose.  Equipment is protected by its own over current protection devices (eg BMS / internal fuses).

Also worth noting, that a 100A fuse will blow in around 240s (4 minutes) at the maximum inverter draw of 190A.  So be sure you are happy with maximum load times before picking smaller fuses (the current-time curves are in the fuse data sheets).

11 hours ago, Beat said:

The double wiring with 35mm² is overkill too.

35mm² will operate at at least 85°C at max continuous draw - so you had better be 100% sure that it is suitably rated high temperature wire, and that it is not touching any low temperature plastics (like PVC conduit).  While you  technically can do it (legally only if the cable is not covered with anything), I certainly would not recommend it.

11 hours ago, Beat said:

The difference of internal resistance between packs due to manufacturing inaccuracy is likely to be higher than the difference of resistance in leads of different length, even with 25mm².

Please see the video posted above, it is very easy to demonstrate that it is in fact quote important to match the cable lengths if you want to maximise the life of the battery.

3 hours ago, JustinSchoeman said:

While you  technically can do it (legally only if the cable is not covered with anything), I certainly would not recommend it.

Not recommend what?

3 hours ago, JustinSchoeman said:

Please see the video posted above, it is very easy to demonstrate that it is in fact quote important to match the cable lengths if you want to maximise the life of the battery.

Well, that guy states clearly that for solar application you can do it the way it is shown in the video. Further more in his analysis he assumes  that batteries of same type, manufacturer and capacity have identical internal resistance. My experience is that this is not the case at all. I have written about it in another thread.

19 minutes ago, Beat said:

3 hours ago, JustinSchoeman said:

While you  technically can do it (legally only if the cable is not covered with anything), I certainly would not recommend it.

Not recommend what?

I would not recommend using only 35mm² wire.

 

24 minutes ago, Beat said:

Well, that guy states clearly that for solar application you can do it the way it is shown in the video.

No. He says you can do it for a low C rate application.  The application being discussed here will run the batteries at close to 1C, for which he says this wiring is completely unsuitable.

Thanks @Scorp007 @Beat @JustinSchoeman all very much appreciated. It's clear that there are different views on things as much as there are also different interpretations on SANS guidelines or requirements. I've also chatted to others (e.g. DC specialist), and have decided to reconfigure things a bit schematically. In fact what @Scorp007 mentioned with the individual battery fuses etc. was strongly advised by someone else too (in fact, when I showed my earlier schematic I was threatened with being taken outside, hanged drawn and quartered). 

The point emphasised was that each battery "must" be individually fused (and not to rely on the BMS or joint fuses to provide all the safeties). And then to join the two batteries via busbars before running a 70mm cable to the inverter. This means I need to find or have suitable busbars made up (the Victron guide is useful on busbar thickness calculations) and to get a seond Keto for the second battery. Schematically it'll probably look like the diagram below.

While the + and - cables to the bus bar should practically end up being the same length, I'm told it's important that the set of + cables must be the same length, and the set of - cables also (but less important that + and - are of exactly same length). And that the run to the inverter should connect to the busbar between the cable connections coming from the batteries. 

The run from busbar to inverter will probably be a more direct than laid out in the diagram. 

 

Edited March 13, 20233 yr by Kalahari Cruiser

5 hours ago, Kalahari Cruiser said:

The point emphasised was that each battery "must" be individually fused

That makes sense, therefor many battery manufacturer such as Leoch have in the packs built in 125A breakers. That also helps when connecting or disconnecting a pack. If you need external fuses, it's good enough to have only one pole fused.

Why the loop of the negative cable? A difference of + to - feeder length to the inverter has absolutely no importance and no effect. Make each as short as possible.

5 hours ago, Beat said:

it's good enough to have only one pole fused.

Ok, so I could in theory leave a second “Keto” and just run the respective pos cables from the battery via the fuse disconnect, to be of similar effect? I don’t think the sunsynk batteries are fused internally but don’t know. I know in vehicle aux electrics it’s often just + fused, and I’ve seen sometimes once at the load, and again at the battery pole. 
 

5 hours ago, Beat said:

Why the loop of the negative cable?

You’re right. As mentioned earlier it’s just a draft diagram and not representative of the final layout runs. Actually, it’s going to be a straight-up from the busbar box/es to inverter which is sitting directly above the batteries. At the time it wasn’t clear yet how things will be mounted. 
 

thanks again, @Beat

I only have the draft regs, but:

Both conductors should be fused:

There should also be a disconnect device:

A fuse disconnect should suffice.

I highlighted the wording 'battery arrangement'. A withdrawn draft referred separately to a 'battery system'. I would take this to mean that modular batteries that are intended to be stacked can be stacked and connected as per manufacturer's recommendations, with a single fuse/disconnect for the entire stack/system/arrangement. (Although I would not personally do this, unless each battery had an internal overcurrent protection.

 

 

12 hours ago, JustinSchoeman said:

I only have the draft regs, but:

Both conductors should be fused:

Do these "regs" apply to systems with voltage below 50V? If yes, then all automotive electric equipments are illegal. So would be flashlights. Technically that doesn't make much sense.

30 minutes ago, Beat said:

Do these "regs" apply to systems with voltage below 50V? If yes, then all automotive electric equipments are illegal. So would be flashlights. Technically that doesn't make much sense.

It is part of OHS regulations, and applies to all premises wiring below 1000V AC or 1500V DC, including LV and ELV wiring.

Cars are explicitly excluded but caravans are included.

 

13 hours ago, JustinSchoeman said:

Both conductors should be fused:

Thanks - this is also what I was told over the weekend with reference to the draft regs. I've made space provision to have two Ketos alongside each other but will still have to source a second one then. 

 

13 hours ago, JustinSchoeman said:

There should also be a disconnect device

Is a separate disconnect for the battery/batteries compulsory, or can one disconnect / isolate them by removing the fuse links (not under load / battery off etc...). I don't really have space for another two isolators outside of the Keto fuse links. I've not noticed any separate disconnect devices in installs I've seen (beyond fuses and COS or isolators in the DB boards). 

Keto devices are rated for disconnect under load, so it is a single device sufficient for both the fuse and disconnect purposes.

Thanks very much for that 👍🏻

2 hours ago, JustinSchoeman said:

It is part of OHS regulations, and applies to all premises wiring below 1000V AC or 1500V DC, including LV and ELV wiring.

Cars are explicitly excluded but caravans are included.

Thanks - good to know.