Ok. Let me start with a simpel boating example. You drop your 5 ton anchor out into the ocean. Anchor is nice and strong. And so is the anchor chain, but the shackle connecting the anchor to your chain is (only) rated for 1 metric ton. This is the weakest link and that will snap before anything else.
Now back to batteries, wiring, inverters, Battery Management and fusing. Let’s assume that your prismatic cell can be charged and discharged with 157A. Your 70mm2 wiring can handle 200A easy. Your BMS can do 100A or 150A max.
So you want to protect your DC battery cables. Cool. Let’s use 200A. Which is what most people do anyway these days. But hold on! Won’t that become a potential problem for the installed BMS, and possibly the inverter as well?
Most of today’s battery housing comes with breakers preinstalled. Sometimes rated for 200A, up to 300A. Switching capacity mind you. But hold on! Won’t that become a potential problem for the prismatic cells and or DC wiring?
In short, too many people only look at what the DC wiring can handle. Yeah yeah. Fuses should (only) be rated to protect the wiring. Really?
So when the installed BMS cannot even handle it, you still want to protect the wiring with a 200A fuse?
Also. So when your prismatic cells cannot even handle it, you still want to fuse the wiring with a 200A fuse?
What happened here to honouring the weakest link? You know. That 1 metric ton shackle?
Yes. A 200A fuse is cool… when everything else in the chain can handle at least that 200A as well.
Now please tell me what type of fuse you use, where your fuse(s) is / are installed, what rating the fuse(s) is / are, and what you think about all this.
Fuse for the weakest link as a primary design criteria.
Unless all links are designed to carry more load than the actual load is, then fuse for the actual load.
My system is : cable 200a+, 2x bms 200a+, 2x fuses 200a +175a, actual max seen load 140a.
Fuse sizes differ due to originally having one battery and fused as recommended by spec sheets as actual load was unknown, 2nd battery fused for actual load / to allow single battery to run loads if required.
We are talking about hundreds of amps and protecting cables…
But let’s take a look at an even more interesting, practical and real, aspect.
I am talking here about the Multi RS, but the Multipluses have the same H-Bridge configuration.
And we know that on a H-Bridge, only two diagonal paths are active at some point. Red and green on the picture below.
Multi RS can handle 100A. Cable OK, fuse OK, connection OK, PCB traces… hmm… let’s say OK, but what about the fact that all those 100A are going through only 4 (four!!) FETs per path, as seen on the picture???
An those FET pins are no more than 1.5mm2 (less than that internally), so a total of less than 6mm2 for 100A!!!
Cables and fuses are stationary and very little chance to touch each other once they are properly installed, so the sole moment when a plus(+) and minus(-) will “touch” is when a semiconductor will fail, so, in my opinion, the weakest link are the semiconductors, that will fail spectacularly with a show of light and sound…
For a fuse to blow or disconnect it may take a few seconds, but the semiconductors will blow literally in milliseconds.
LE: Sorry, I’ve pointed out the gate pin, but the drain and/or source pins are the same size, so no matter, the cross section is the same…
I have been fighting this kind of thinking for years. It started with the lack of proper battery fusing. Now solved with the introduction of gBat Class-T fuses. However. Open 100 off the shelves batteries and you’ll find none of them with any kind of internal fuse (I kind of understand that as well). Same for Battery Management Systems. No real fusing whatsoever (as a last resort when a fuse is not there). That’s the harsh reality of how bad it still is (in the world of DIY which is what we are discussing here).
So yeah in an ideal world you could go about fusing based on your wiring. But most of the (DIY) battery boxes are still being sold without a fuse. And in that case fusing based on what your wiring can handle, could potentially end in a disastrous event. And that is what I want to bring under people’s attention.
Note: a fuse in a battery container should be A: a class-t fuse and B: be installed before the BMS. This to be certain that a short inside won’t toast your stuff.
I know a customer who had a short somewhere in one of his 3 phase inverters, and the lack of fuses burned down all three of them. Happened years ago, but still. Better safe than sorry.
So did you install (a) fuse(s) or not? I did and had one that failed for some weird reason. Why would a 100A fuse fail on a MP 4K6 that won’t ever do 100A?
I did, but somehow for the reason of easy disconnect when I need it. An 125A one.
I know for sure, that if something will go south, for sure the PCB traces and/or the FETs are the first casualties.
I can’t imagine something inside the Multi RS that could be “tougher” than a 35mm2 cable…
So the cable will handle it alright, but an eventual fire, God forbid, it will be because of the arching, for sure.
And also, for sure, the 125A breaker/fuse will not disconnect an arching in progress, like happened on your 3 phase customer’s inverters…
I didn’t see up until now a reasonable and affordable AFDD for 125A DC current…
Hi there. Thanks for your reply. Yeah I know Will and his videos. Usually good entertaining stuff. And you’ve said it yourself. A good BMS. All are ‘good’ (enough) until one fails. A dead short near a BMS or on the battery side might be rare incidents. Sometimes even due to some human failure
Anyway. I think that we can all agree about one thing, like not risking anything.
I just watched a video this morning from Dogan batteries and they showed how they work. The first thing I noticed was that all used busbars had black lines on them. Like the nut was torqued. But before they even screwed the nut on the terminals No idea what was going on there, but it looked very odd to me. Seen way too many weird things over the course of my battery mission.
Looks like we have similar feeds as I saw the Docan Panda mega battery I hope those bus bars are copper but they don’t look great to me but I’ve never designed a battery and I don’t intend to either lol
To be honest the fusing thing is a topic not only for DC but for AC too as in the UK they are talking about changing the RCBO’s from 32amps to 40amps for 7kW EV chargers bearing in mind that the EV chargers that they want this on have a max up 32amps so 40amps won’t protect it from overload, as some of the cheaper brands RCBO’s are burning out due to the sustained 32Amp load for a couple of hours.
They are relying on the device to honest think we are getting to lazy for convenience as a society. Even some charger manufacturers are specifying a 40amp RCBO’s to mitigate the failure.
Hi there. Nice. Sure thing. Fuse should be installed on the positive side of the battery. Either inside your battery box, or in Victron’s Class-T Lynx Power In. On a busbar. You name it. As long it is there.Just some examples after all.
You can also use a EFI NH1 fuse holder with a 80V gBat fuse. Fuse holder acts as a battery disconnect as well. The fuse holder is a bit expensive, but looks professional.
Yes the NH gBAT fuses are my favourites,as I have them installed,tiogether with the trible holder.It looks like the ideal solution to keep sparks not going all over the place
Atefuse in china sell them for a decent price,the problem are the taxes….
No idea what was going on there, but it looked very odd to me. Seen way too many weird things over the course of my battery mission.