Two batteries (LFP). Diagonal vs. busbar method considerations

[luna]

Have you quantified the cost difference of an ANL versus a Class T fuse as the Main fuse on your battery bank?
Remember, "selecting a better fuse with a higher AIC is perfectly fine" and may save you some time and energy down the road.

 

[Frosty]

Have you quantified the cost difference of an ANL versus a Class T fuse as the Main fuse on your battery bank?
Remember, "selecting a better fuse with a higher AIC is perfectly fine" and may save you some time and energy down the road.

I have a Class T fuse for my main fuse right now. I will have TWO Class T fuses as my "main" fuses in future. IT'S NOT ABOUT MONEY. If it were, I would not have a boat. I hate being cheap, and fortunately I don't have to be (small boat makes it easier to splurge).

Just stop.

 

[Delta Riverat]

I don't think you would say that if you were me.

For one thing, the MRBF's would protrude up into a locker where random things get stowed. Not cool.

For another, their AIC rating is knife-edge close to the number of amps my batteries could put out in a short circuit situation.

Combine those two things and Class T fuses are an easy choice. They fit in the compartment where the batteries live, and they have ample AIC rating for my batteries.

Your boat space is likely different from mine. If I had your boat I might make a different decision.

Yeah, my batts are in battery boxes with covers.

I have seen the video with a MRBF short, the fuse exploded as I thought it would, case ruptured and little metal balls everywhere. But no arcing, just a blow out. An arc would make me nervous but also really hard at 12 volts.

Yeah if you have stuff in that locker that would be hazardous if that happens I wouldn't use them either.

 

[luna]

Listen carefully.
I would also have 3 fuses, I said that earlier. I am not your adversary.
The difference with what you are advocating versus what ABYC says, is the cost difference between your scheme, an ANL fuse and a Class T fuse as the Main Fuse on your system.

ABYC, nor your Insurer (being that you are in Portland) could care less what you have as an AIC Overcurrent Protection Device as due to your battery bank size, none are required.

Do what you want.
Is the premium (likely $50) to use a ClassT versus an ANL that important?

 

[Frosty]

Yeah if you have stuff in that locker that would be hazardous if that happens I wouldn't use them either.

If it weren't for the height I would consider them, even though their ratings are right on the edge. Interesting to hear your description of how they failed.

But obviously things have to fit in the space (Boat Tetris!). I have room to spread out sideways, but not upward. Thanks for recognizing that

 

[Delta Riverat]

It was a you tube video, you can probably google it. Yeah, fuse exploded, plastic case blown, melted element balls outside. But not that many. If you have seen one of these fuses the element is pretty small.

In event of a modest overload I'm sure it would just melt w/o the explosive drama. As I said I was most concerned about the arcing and there was none. Batts are in ER, in boxes, diesel engines, most flammable substance there is diesel. If gasoline in ER, no way would I use them.

 

[Frosty]

Yeah, fuse exploded, plastic case blown, melted element balls outside. But not that many. If you have seen one of these fuses the element is pretty small.

Just out of curiosity, besides the "mess" from the event.... would the fuse still have acted as a fuse? In other words, would it have prevented current from passing through? Or not.

If gasoline in ER, no way would I use them.

Yeah, that would be a whole different ball game.

 

[Delta Riverat]

Oh yeah, it did it's job, no current was going through.

 

[DDW]

OK, you want to go there. It has long been argued that a class T fuse is 20,000 @ 125V and perhaps at 12V it is 50,000. There are fuses not called class T that state a higher AIC. What miracle makes Mersen class T different except they have claimed at 50,000 @ 160VDC. Class T fuses never before claimed more than 20,000, so why isn't Mersen saying they are class U? Because all class T may well be 50,000 @ 12V.
I was hoping ABYC would have made this clear with the 2025 revisions.

It doesn't need to be magic. What is required is it quench the arc at the rated voltage and amperage. Clearly the technology exists to do it, as there are fuses and breakers that do. There is likely a reason that the Mersen cost somewhat more than the Eaton (except for the clearance ones on Amazon or Ebay). I think the 20K is a minimum spec for Class T, not the maximum. They claim UL listing to 248, which covers AIC up to 300K. A company like Mersen is unlikely to fake a UL listing.

Do you have reason to believe they do not meet their published spec? It seems like a reputable company. The fuses aren't even made in China (but rather, Mexico).

 

[SteveK]

@DDW Over the past two years there has been rumors about the class T fuse being more than 20kA @12V, 50kA @12V was the possibility, but no one tested a class T at 12V to confirm the supposition. I checked Mersen and could not find such testing.
ABYC has based the new standard on a class T 20kA, not a 50kA which would change most of the 2025 revision and would be welcome for LFP banks.

IMO all class T fuses should be rated the same, should it not?
We have become accustomed to 20kA @ 125VDC, now Mersen is 50kA @ 160VDC. What are they @12/24/36/48V?
Call me skeptical, because I am.

 

[luna]

@DDW Over the past two years there has been rumors about the class T fuse being more than 20kA @12V, 50kA @12V was the possibility, but no one tested a class T at 12V to confirm the supposition. I checked Mersen and could not find such testing.
ABYC has based the new standard on a class T 20kA, not a 50kA which would change most of the 2025 revision and would be welcome for LFP banks.

IMO all class T fuses should be rated the same, should it not?
We have become accustomed to 20kA @ 125VDC, now Mersen is 50kA @ 160VDC. What are they @12/24/36/48V?
Call me skeptical, because I am.

I don't think any Class T fuse manufacturer has an AIC rating at 12, 24 or 48 Volts. It could be that the existing rating at 125 or 160 V fills most of the fuse market's need and what is left unfilled is not big enough to justify further testing as other products already exist to satisfy this small portion of the market.

I would be astonished if ABYC based their Standards on a Class T 20kA fuse.
The wording in the Standard does not seem to support this claim as it does not mention any particular type of overcurrent protection at all, fuse or CB or?.
Instead you are given a formula to calculate the system's short circuit current and then it is up to you to select what component meets the entire Standard. That is the way Standards work.

Why do you feel that all Class T fuses should have identical ratings. There are a lot of different sizes and designs (not all are bolt in) of Class T fuses. A Mersen A6T300 has an IR of 100kA (5 times that of your 20kA) at 300 VDC. Should this fuse not exist? Wouldn't your opinion tend to stifle innovation?

 

[DDW]

I'd agree that a 20KAIC 160V fuse most likely has an AIC much higher than that at 12V. Like Luna, I don't think testing is done there because there is no market. There is a market for high AIC rated fuses at high voltages, because of the millions of installs of solar arrays operating in that area. If you are skeptical, you could contact Mersen and ask to see the UL report. UL certifications are often self (or 3rd party lab) done, so there is the opportunity to lie about it, but UL also does frequent audits, and this is a big hammer: your market for fuses will disappear overnight if you lose UL certification, and the liability is huge.

In my case I am putting these fuses in to satisfy a future survey, not because I perceive any extraordinary risk. The data sheet from Mersen is all that is required to satisfy the risk of survey.

 

[SteveK]

@DDW The mersen fuse claims 50kA @ 160vac.
Are you expecting 50kA @ 12v.
With the ABYC suggestion of 5000 AIC for every 100 Ah battery you could fuse up to 1000Ah bank.

 

[DDW]

Here is the data sheet for the Mersen fuses. There is no claim for 12V, but the A3T series are rated at 50kA @ 160VDC. The VAC rating has no meaning in a DC circuit. I have a 900AH battery bank, according to the ABYC I need 45KAIC. So 50K is enough. For a larger bank you could use Mersen's A6T series, rated at 100K AIC @ 300V. In both cases the actual interrupt capability will be substantially higher at 12V than 160V or 300V, but without that showing on the data sheet, it would carry little weight with the surveyor.

For the individual battery fuses I can't use an MRBF type, as each battery is 300AH so an AIC of 15K is called for. But ordinary Eaton Class T fuses at an AIC of 20K are sufficient. They are a little cheaper than the Mersen, around $56 vs $88 and I need 3 plus spares.

 

[luna]

The VAC rating has no meaning in a DC circuit.

Ooops, and I know that.
It seems that I just took the VAC value and used it for the VDC value, by mistake in my previous post. The Post has now been corrected. Sorry about that. The premise however remains.

 

[Marco Flamingo]

Class T is available in 10k, 20k, and 50k (and maybe others). As one would guess, size matters and governs the AIC. The easiest way to snuff a plasma arc is increasing distance. The 10k Class T I've seen are quite a bit smaller than the more common (for our discussions) 20k. The tiny little glass-tube fuses have huge AIC ratings because when the little wire melts there is an inch between contacts. Can't do that with MRBF.

Will Prowse tests the the smaller Class T (for basic amperage) in a video where he also tests a variety of MRBF and ANL fuses. The video leaves one with the impression that if the fuses don't blow very close to their rated amperage, what makes one believe they will blow close to the rated AIC? His advise is to buy recognizable name brands, especially for the MRBFs, but does no testing to suggest that this claim is anything other than a possible improvement.

I got the impression from some of the threads that there is a concern for LFP batteries shorting between themselves when in a bank. The theory is that an MRBF might stop that (if one has batteries smaller than mine as the AIC of a MRBF isn't enough). Assume a 320AH per LFP three-battery bank of battery A, B, and C. Battery C has an internal fault (not sure what that would be). Battery A and B send 14k amps into battery C (because both of A and B's discharge overcurrent cutoffs simultaneously fail as does battery C's charging overcurrent cutoff at the same time as its temperature cutoff fails). Sending 14k amps to where exactly? If through C to ground, then put a Class T or a 50k AIC MCCB on the ground cable. When I try to send 100 amps to a single LFP battery, it won't accept that much. Doesn't get near my set overcharge shutdown. What has happened inside of battery C that it can now accept 14k without a ground?

 

[SteveK]

Here is the data sheet for the Mersen fuses. There is no claim for 12V, but the A3T series are rated at 50kA @ 160VDC. The VAC rating has no meaning in a DC circuit. I have a 900AH battery bank, according to the ABYC I need 45KAIC. So 50K is enough. For a larger bank you could use Mersen's A6T series, rated at 100K AIC @ 300V. In both cases the actual interrupt capability will be substantially higher at 12V than 160V or 300V, but without that showing on the data sheet, it would carry little weight with the surveyor.

For the individual battery fuses I can't use an MRBF type, as each battery is 300AH so an AIC of 15K is called for. But ordinary Eaton Class T fuses at an AIC of 20K are sufficient. They are a little cheaper than the Mersen, around $56 vs $88 and I need 3 plus spares.

And I hope the 50kA @ 160vdc is correct.
There are several classes of fuses. Most class T claim 20kA @125

 

[Frosty]

I have a 900AH battery bank, according to the ABYC I need 45KAIC. ...

For the individual battery fuses I can't use an MRBF type, as each battery is 300AH so an AIC of 15K is called for. But ordinary Eaton Class T fuses at an AIC of 20K are sufficient. They are a little cheaper than the Mersen, around $56 vs $88 and I need 3 plus spares.

@DDW
If you have a moment, I'm still trying to work out a puzzle from the "other" thread. (It might not be a puzzle if I could read the ABYC standard in its entirety, vs. excerpts). I think your example above is the key.

So is the 45k AIC number for the "main" fuse arrived at by adding up the short circuit amps minus 1 amp from each battery? So they are thinking that theoretically each battery could pass 14,999 amps and still get through the inter-battery 20k fuse, hence you need a 44,997 amp main fuse (rounded to 45k). Is that the calculation?

I don't have a way to read the complete standard right now so am only getting it in bits and pieces here and there.

For my own bank (2 batteries each capable of putting out 10,000 amps), I couldn't figure out why the "main" fuse needed to have a high AIC rating if each battery already had a 20,000 amp AIC-rated fuse. If I'm back-calculating the ABYC thinking correctly, I think I don't in my specific case, since 2 x 9,999 is only 19,998 and each battery is already proposed to have a 20,000 AIC rated fuse on it. So in effect I would already have two "main" fuses, which just happen to also be inter-battery fuses.

Put another way, I couldn't figure out why a downstream "main" fuse with a 20,000 amp AIC rating would be needed after a pair of 20,000 amp AIC-rated fuses on a pair of batteries, when the total short circuit rating of the entire bank is not over 20,000 amps.

Do you think I'm understanding the intent correctly? I wasn't able to work it out at all (not having read it) until I back-calculated from your figures.

Obviously, this situation is a little unique just because I only have two 10,000 amp short circuit batteries, and they would already each have a 20,000 amp Class T fuse on them (basically picture a Victron Lynx Class T Power In, or the drawing below). I'm not blaming ABYC here because they have to word a standard to cover all kinds of things. My proposed way would not work for most setups, because things would add up differently (higher).

In the drawing below, which is my proposed setup, I could not work out why a third fuse of the same value would be needed after the Lynx (in my specific case). Where I drew in the ANL. But another poster felt that it was absolutely necessary and would do something the other two Class T's would not already do. I'm only in the design stage so....

My motivation is not cost. The additional cost for another Class T is a drop in the bucket. It's two-fold:

1) Understand how things work electrically, and what is needed to do the job.

2) Space. Another Class T fuse is considerably more bulk on a boat of my size if it's not needed electrically.

(I have reasons for not using MRBF's between the batteries, which then would obviously want a Class T for the "main" fuse.)

Thank you for your time.

 

[SteveK]

Class T is available in 10k, 20k, and 50k (and maybe others). As one would guess, size matters and governs the AIC. The easiest way to snuff a plasma arc is increasing distance. The 10k Class T I've seen are quite a bit smaller than the more common (for our discussions) 20k. The tiny little glass-tube fuses have huge AIC ratings because when the little wire melts there is an inch between contacts. Can't do that with MRBF.

Will Prowse tests the the smaller Class T (for basic amperage) in a video where he also tests a variety of MRBF and ANL fuses. The video leaves one with the impression that if the fuses don't blow very close to their rated amperage, what makes one believe they will blow close to the rated AIC? His advise is to buy recognizable name brands, especially for the MRBFs, but does no testing to suggest that this claim is anything other than a possible improvement.

I got the impression from some of the threads that there is a concern for LFP batteries shorting between themselves when in a bank. The theory is that an MRBF might stop that (if one has batteries smaller than mine as the AIC of a MRBF isn't enough). Assume a 320AH per LFP three-battery bank of battery A, B, and C. Battery C has an internal fault (not sure what that would be). Battery A and B send 14k amps into battery C (because both of A and B's discharge overcurrent cutoffs simultaneously fail as does battery C's charging overcurrent cutoff at the same time as its temperature cutoff fails). Sending 14k amps to where exactly? If through C to ground, then put a Class T or a 50k AIC MCCB on the ground cable. When I try to send 100 amps to a single LFP battery, it won't accept that much. Doesn't get near my set overcharge shutdown. What has happened inside of battery C that it can now accept 14k without a ground?

Marco, I tend to accept your post and will review ABYC E11 to see how it is worded. I may have gotten used to a classT as only being 20kA. But your post makes it seem as class T just relates to arc prevention ability.

 

[Frosty]

@SteveK
I was thinking Class T was just a "shape." Like ANL, or blade fuses, or glass tube fuses, or Midi fuses, etc.

(Granted, there are different physical sizes of this shape, like where they jump from 200ah to 225 ah; but then there are mini blade fuses and normal blade fuses too so I guess some other "shapes" have physical size variances, while still being the same style.)

Or am I off base?

 

[SteveK]

@SteveK
I was thinking Class T was just a "shape." Like ANL, or blade fuses, or glass tube fuses, or Midi fuses, etc.

(Granted, there are different physical sizes of this shape, like where they jump from 200ah to 225 ah; but then there are mini blade fuses and normal blade fuses too so I guess some other "shapes" have physical size variances, while still being the same style.)

Or am I off basbasewell even Blue seas confuses me since class T holder and fuse is one size and shape.

yes, now we see different shapes of class T some of which are only rated for AC.

Frosty, now you could be class T 10kA off battery and 20kA on main class T.

Just like watts is the same AC or DC, there should be a formula to Calc what the AIC is at 12V

 

[DDW]

There is no way I can explain the thinking behind the ABYC recommendations, to me they make little sense in this case. The 45K is from their formula of 5000 x AH capacity. I don't know where they got that, but perhaps a generic LFP cell can do that under certain circumstances. Why a fuse followed by a fuse? Again I don't know, and it isn't clear it makes any sense. If there really is a short resulting in 45K A, all three individual battery fuses would blow and if we believe the 45K was shared between the three batteries that would be 15K per, below the AIC of the fuses. Also, internal to the battery case there may be two or more parallel strings. These are not, and cannot be, fuse isolated. Why is that different?

I'd really like to see the justification for all of this, seems like it is belt and suspenders and better safe than sorry (and we aren't paying for it after all...). I'd be more convinced if they said, we have the following 5 incident reports where one main fuse of 20K was insufficient and resulted in a fire. I doubt very much they have one such. In my case there are BMSs protecting short circuits and the cabling would not support 45K amps

As far as fuses not blowing near their rating, if you look at the manufacturers charts there is a range, and the time and amps to blow depends on a lot of things. A 100A isn't going to blow at 101A. Even the classification specs allow (or require) some over amperage. May blow at 150A after a minute, or 300A in a second. They have to heat up and melt. It takes time, and varies depending on other factors, including the cable connected to the holder. I'm using the BEP holders as they pack tightly, but even that affects the fuse behavior.

 

[Frosty]

There is no way I can explain the thinking behind the ABYC recommendations, to me they make little sense in this case. The 45K is from their formula of 5000 x AH capacity.

Okay, so I got that part right (meaning, the ABYC numbers). So then I am good, as proposed, because it just so happens that my two first fuses are 20,000K each, and I have 400ah total capacity. So either of them (even alone) would qualify as a "main fuse." I just will happen to have TWO main fuses (double protection!). Clearly there is no electrical need for yet another duplicate after that. This also matches up to the Victron Lynx Class T Power In drawings (which is really just two ganged Class T fuses of 20k each).

Of course if I had the 10,000k MRBF's as inter-battery fuses (or if I had a larger battery bank), then a larger "main fuse" would be called for. So that would apply in most cases (inter-battery fuses followed by a needed main AIC-rated fuse).

As far as fuses not blowing near their rating, if you look at the manufacturers charts there is a range, and the time and amps to blow depends on a lot of things. A 100A isn't going to blow at 101A. Even the classification specs allow (or require) some over amperage. May blow at 150A after a minute, or 300A in a second.

Right. But I'm assuming the 20k AIC rating isn't like that. Meaning, it's just 20k and never changes(?). I understand the normal (non-AIC) blow curves of the various fuse shapes (Blue Sea has nice graphs).

*******

The BEP installer stuff didn't really end up helping me for my specific DC distribution layout and switch desires, but it's amazing how compact it is. But no, I was the one hunched over the workbench making a bunch of little shims of different heights and shapes to get my Blue Sea and Victron bits and pieces to match, height-wise

 

[SteveK]

But... But, that is what I am doing 150 posts ago. Class T on each 400Ah pack combined thru 12all used as buss bar.

 

[DDW]

Assume a 320AH per LFP three-battery bank of battery A, B, and C. Battery C has an internal fault (not sure what that would be). Battery A and B send 14k amps into battery C (because both of A and B's discharge overcurrent cutoffs simultaneously fail as does battery C's charging overcurrent cutoff at the same time as its temperature cutoff fails). Sending 14k amps to where exactly? If through C to ground, then put a Class T or a 50k AIC MCCB on the ground cable. When I try to send 100 amps to a single LFP battery, it won't accept that much.

The fear is that it will send 14K amps to the other parallel battery. To do that requires multiple very unlikely faults. The BMS would have to fry and fuse into a short, the battery electrode would have to fuse into a very low resistance short, all of the internal cabling (which tends to be small) would have to support 14K amps. Also the ABYC doesn't believe internal BMSs exist (the same spec is for batteries with and without), and certainly doesn't believe they work as designed (hence all the external fusing nonsense).

The job they have is difficult: try to define a recipe for battery installation that covers a whole galaxy of possibilities, in 2000 words or less. And their clients are builders and surveyors, many of whom are unknowledgeable about electricity, and who want a recipe that they can point to to deflect a lawsuit. I don't envy that job. The result is not optimal, or even applicable, in many situations.

 

[Marco Flamingo]

The fear is that it will send 14K amps to the other parallel battery.

Got it. But I'm still stuck in the mindset of electricity is like water through a hose. Especially DC current. It has to "go" somewhere. If, using my example, 1,000 amps equals a gallon of water, then the faulty battery is instantly (and constantly) choking down the 14 gallons produced by batteries A and B. I guess I could imagine the cells in battery C as being super-absorbent pads, but I'm still having difficulty with the physics. Having the fault go to ground I can understand.

 

[DDW]

A single battery isn't circuit (or shouldn't be) but connect it so another with a short and you have a circuit. Out through the positive, through the short, and back to the negative. Don't need anything else. The "water" will be transformed into heat - and that is what all the worry is about. A 200 AH battery is about 2600 Wh, or call it 9000 BTU. A lot of heat in a small battery.

 

[SteveK]

Marco, I tend to accept your post and will review ABYC E11 to see how it is worded. I may have gotten used to a classT as only being 20kA. But your post makes it seem as class T just relates to arc prevention ability.

I found this

11.10.1.3.2.3 20 kA at 125 VDC or higher, if a short circuit current rating exceeds 20 kА.

NOTES1. The manufacturers' AIC rating of overcurrent protection devices is specific to a particular voltage. Typically, the AIC rating decreases as the voltage increases. Some current limiting fuses are rated at125V and above.

What does the bolded mean to you?

IMO it means a 160V should be less kA than a 125V kA
AND 12V should be higher kA, which has been the theory
and why I previously said @ 12V it is proposed to be 50kA

Mersen claims 50kA @ 160V,
Blue Seas claims 20kA @ 125V
Shouldn't Blue seas be higher kA than Mersen?

 

[luna]

The Bolded sentence simply means that if I have a fuse that has an AIC of 20kA at 125 VDC and I was to raise the system voltage to 150 VDC, typically the AIC for that fuse would go down to something less than 20kA. How much is anybody's guess.

It does not mean that the Mersen A3T fuse which is rated at 160 VDC will have a lower AIC than the Eaton/Bussmann JJN fuse. (Blue Seas does not make these fuses) which is rated at 125 VDC.
They are different fuses made by different people even though they are the same size and Class.
In this instance Mersen builds a better mousetrap.

why I previously said @ 12V it is proposed to be 50kA

Who is the proponent of this proposal to reassign a higher AIC rating to a fuse? Eaton?

 

[SteveK]

The Bolded sentence simply means that if I have a fuse that has an AIC of 20kA at 125 VDC and I was to raise the system voltage to 150 VDC, typically the AIC for that fuse would go down to something less than 20kA. How much is anybody's guess.

It does not mean that the Mersen A3T fuse which is rated at 160 VDC will have a lower AIC than the Eaton/Bussmann JJN fuse. (Blue Seas does not make these fuses) which is rated at 125 VDC.
They are different fuses made by different people even though they are the same size and Class.
In this instance Mersen builds a better mousetrap.




Who is the proponent of this proposal to reassign a higher AIC rating to a fuse? Eaton?

Luna we agree on the bolded ABYC. Then, the reduction of voltage allows the AIC goes up. Do we agree on that?