diy solar

diy solar

Can you have too many fuses?

Aphers

New Member
Joined
Nov 17, 2020
Messages
212
I've been looking at different installations and seeing it's quite common to use MRBFs on the battery terminals, then a big class T, and then smaller fuses for individual circuits.
At the same time I've read that fuses can run hot, which sounds inefficient. So is it best to minimise the number?

For my own system, I want to use a class T so that I have something with high enough interrupt current to handle a catastrophic short. But they're *so* expensive. Would there be anything wrong with adding a cheaper fuse in series, at a lower rating? An ANL or Mega fuse is about 20 times cheaper, so I can carry plenty of spares. And if the short is bad enough to exceed its interrupt current, then the class T will blow as well.
 
The typical interrupt rating capability of the main fuse connected to the battery positive for the various types of fuse are,

Class T 20kA at 160v dc
MRBF 10kA at 14v, 5kA at 32 v dc
ANL 6kA at 32vdc
Mega 2kA at 32 v dc

Note the ratings differ with different manufactures, Cooper for example rate their ALN fuse at 6kA at 80v dc

Where a FET based BMS is in the negative path of the battery, the fault current has to find its way through the BMS which I suspect will limit the fault current somewhat. ( assuming the over current protection in the BMS has failed)


then smaller fuses for individual circuits.
the fuse protects the cable and whatever is on the end of that cable from fault currents in excess of the cable rating. In practice the cable sizes will vary is a system, thus each leg may/will need a separate fuse. Apart from circuits like lighting where a single fuse/breaker may feed numerous lamps, most major appliances, chargers and inverters will need their own fuse or DC breaker.

with adding a cheaper fuse in series,
with a serious fault the less able fuse may pass current and the class T will still fail.


Mike
 
Thanks Mike.
To be honest I think going with the class T is kind of overkill for me. With a 4S/270Ah pack the highest current I'm likely to see in a dead short, with no losses anywhere, is something like 10C, i.e. less than 3KA. An ANL fuse should handle than no problem, especially as the figures are usually given for a higher voltage.
Anyway to play on the safe side I've ordered a class T, just wondering if it will introduce losses if I also have ANL fuses in series with that.
 
I've been looking at different installations and seeing it's quite common to use MRBFs on the battery terminals, then a big class T, and then smaller fuses for individual circuits.
At the same time I've read that fuses can run hot, which sounds inefficient. So is it best to minimise the number?

For my own system, I want to use a class T so that I have something with high enough interrupt current to handle a catastrophic short. But they're *so* expensive. Would there be anything wrong with adding a cheaper fuse in series, at a lower rating? An ANL or Mega fuse is about 20 times cheaper, so I can carry plenty of spares. And if the short is bad enough to exceed its interrupt current, then the class T will blow as well.
I'm not sure how much is "too expensive", but Don Rowe has decent prices on class T fuses. Yes, $20 for a replacement fuse is expensive.

Only you can decide if safety is worth the price. Try to avoid blowing them! :)


That's what I used, so far haven't needed the replacement fuses I purchased. MRBFs are a "belt and suspenders" type of thing, certainly it is located where it does the most good, but if you have a class T inside your battery case, probably isn't needed.

Fuses do indeed generate heat, that's how they work. Yes you can have too many fuses, but a main one for each battery is a good idea.
 
Thanks Mike.
To be honest I think going with the class T is kind of overkill for me. With a 4S/270Ah pack the highest current I'm likely to see in a dead short, with no losses anywhere, is something like 10C, i.e. less than 3KA. An ANL fuse should handle than no problem, especially as the figures are usually given for a higher voltage.
Anyway to play on the safe side I've ordered a class T, just wondering if it will introduce losses if I also have ANL fuses in series with that.
With LiFePO4, one thing to consider is the AMAZING amount of current they can supply to a dead short. Way beyond the 10C you think.
That's why people use class T.

Will lesser fuses work? Probably.
 
Thanks Mike.
To be honest I think going with the class T is kind of overkill for me. With a 4S/270Ah pack the highest current I'm likely to see in a dead short, with no losses anywhere, is something like 10C, i.e. less than 3KA. An ANL fuse should handle than no problem, especially as the figures are usually given for a higher voltage.
Anyway to play on the safe side I've ordered a class T, just wondering if it will introduce losses if I also have ANL fuses in series with that.
How do you figure?
Not saying you are wrong just want to see how you do the math on this.
 
I'm not sure how much is "too expensive", but Don Rowe has decent prices on class T fuses. Yes, $20 for a replacement fuse is expensive.

Only you can decide if safety is worth the price. Try to avoid blowing them! :)


That's what I used, so far haven't needed the replacement fuses I purchased. MRBFs are a "belt and suspenders" type of thing, certainly it is located where it does the most good, but if you have a class T inside your battery case, probably isn't needed.

Fuses do indeed generate heat, that's how they work. Yes you can have too many fuses, but a main one for each battery is a good idea.

At $20 I wouldn't bat an eyelid.
But the usual price around here is this:

(in case the link doesn't work, that's £98, or $135)

With a bit of digging around the lowest price I can find for somewhere that actually has stock is about £45 (=$62).

It's not a big price to pay for safety, but I wouldn't want to do something stupid and blow it, and I wouldn't be planning on carrying more than one spare. Hence the idea of more fuses down the line at a lower rating. Leave the Class T as the emergency fallback for a dead short.
 
With LiFePO4, one thing to consider is the AMAZING amount of current they can supply to a dead short. Way beyond the 10C you think.
That's why people use class T.

Will lesser fuses work? Probably.
Interesting, thanks. I had read that between 7 and 10 C was what could be expected. Couldn't find any other figure. And of course it will depend on what route the short has to take, in real life it's going to be constrained by the cabling, the voltage drop, any components and connections etc.
 
How do you figure?
Not saying you are wrong just want to see how you do the math on this.
Well I could be wrong but as I understand it:
cells are 270Ah
A 10C dead short would generate 2700A
Good quality ANL fuses claim an interrupt current of up to 6000A

My 10C figure could well be wrong, but I haven't found another figure.
 
Well I could be wrong but as I understand it:
cells are 270Ah
A 10C dead short would generate 2700A
Good quality ANL fuses claim an interrupt current of up to 6000A

My 10C figure could well be wrong, but I haven't found another figure.
Where do you get the 10C from?
 
There will be some voltage loss through any connection. Also, any connection is subject to eventual partial failure. The fewer connections, the better, at least in theory.
 
I agree to stick with the minimum. If the current source exceeds the wire capacity a fuse is needed. And again at branch circuits where the wire is reduced. Can be a bit more complicated if there are multiple power sources feeding a circuit. And sometimes I would just skip the fuse if the connection is short and well protected from damage.
 
OK found it again- was from a discussion on this forum: https://diysolarforum.com/threads/l...it-current-and-main-circuit-protection.14343/

It's got me wondering about a MRBF instead of the class T. Less than half the cost of class T and a very tidy, compact solution. Should handle up to a 37C short for my size of battery. Is that enough protection?
Can you link a specific post as their are 225 posts in that thread?
I searched the the thread for "10c", "10 c", "10C" and "10 C" to no avail.
I also posted ~10 times in that thread.
 

This investigation using theoretical modeling suggests a higher rate,

another test indicated around a few 1000 amps,
SC test.jpg


Mike
 
Last edited:
Everyone is free to use or not use a fuse of any type.
I went with the class T, it's $20 to replace, and cost me $50 with the holder. So far, I haven't used any replacement fuses.

 
Everyone is free to use or not use a fuse of any type.
I went with the class T, it's $20 to replace, and cost me $50 with the holder. So far, I haven't used any replacement fuses.

I would too, without hesitation, if I could buy class T for that price.

The best price I can get for a class T fuse and holder is nearly £100 (~$140).
 
I would too, without hesitation, if I could buy class T for that price.

The best price I can get for a class T fuse and holder is nearly £100 (~$140).
I wish I could help.

You might actually try contacting the vendor, he was very helpful when I ordered. The replacement fuses were on backorder, and he kept me informed and updated. He might actually ship to Europe for an additional fee, I don't know. It wouldn't hurt to ask.
 
Just re-read this topic from the top. I might be missing something obvious.
I've been looking at MRBF (10KA @ 14v) as the second best option to a Class T.
But looking at the figures posted by @mikefitz, ANL is 6KA at 32v (and 80v for some). Which should mean 16KA (or even up to 40KA) at 12v. Way higher than MRBF.
Is it that simple?
Obviously Class T is the gold standard. I'm not totally against the idea of spending the money for a class T. In fact I already have one on order, but with long lead times here in the UK I've got another week or so to reconsider.



A different, related question- fuse sizing. My biggest load is an EPever 3000w inverter. The manual says to use a fuse rated for twice the max current (=500A) but they don't specify what type of fuse. If I use a 500A class T, looking at the delay curve, it's never going to blow in anything other than a dead short. I'm not sure that's a great idea? The retailer said to use a 400A Mega Fuse. I have ordered a 400A Class T thinking I was erring a little on the safe side, but maybe I should swap that for a smaller one. 250A? That lets me use normal size bus bars etc.

And then there's inrush currents... I don't fully understand what that's about, but I've read enough to know that they happen regardless of whether a load is connected, and that they vary from one inverter to another. And I worry that a 250A class T could be blown despite no actual fault, which could get a bit spendy.

I feel like I'm over thinking all this. But going back to my original question, I could have a massive class T followed immediately by a smaller ANL. That gives me a weak link that won't break the bank if it blows, as well as solid protection against dead shorts. Seems overly complex, but at UK prices I can buy ten ANL fuses and a holder for the price of a single spare Class T.




Obviously I don't want to go bigger than necessary as that forces me to upgrade everything else. I'm using a Blue Sea HD isolator (600A) so that's not going to be a problem. I guess everything is driver by the biggest load, which is the 3000w inverter. It's an EPever, and the manual says to use a fuse rated at double the max draw, so that would be 500A. I've been going round in circles looking at delay curves- they don't specify what type of fuse to use, which makes a big difference. A class T can run at 200% for about a minute, whereas the inverter should only draw 500A for up to 10s during surges.
I've already ordered a 400A class T, it's not shipped yet and I'm thinking of dropping that down to 250A. That makes it a lot easier to find bus bars etc which
 
Back
Top