Calculating AIC for fuse selection.

[FilterGuy]

Folks,
As we get bigger capacity LiFePO4 cells and then put them in parallel, the internal resistance is going from very small to crazy small.... That means the fault current it getting crazy high. Has anyone seen a good analysis of how to calculate the AIC (Amp Interrupt Capacity) needed for fuses in and around the batteries?

As an example, the internal AC impedance of the Eve 280s is listed as .25mohm. (I did not see a DC Resistance but for this calculation we are talking about a sudden spike in current so AC impedance may be appropriate). Put 8 in series for a 24V system and you get 8*.25 = 2mOhm, but if it is a 2P8S it is 1mOhm. You will have resistance in your bus bars so lets multiply that by 2 to get a pack resistance of 2 mOhm as a first approximation.

At 24v that means the short circuit current is 24/.004=6000Amps. A good class T fuse should handle that just fine. But what about other fuses down stream of the Main fuse?



Does the 10A fuse in the diagram above need an AIC of 6000? What about the 250A fuse?

(If this is covered in another thread, please point me to it)

 

[John Frum]

When you add a bms I guess you could multiply by 2 again.
Which puts this fuse in spec but without a whole lot of margin considering I was guessing and you were estimating.

MRBF Terminal Fuse - 300A - Blue Sea Systems

Space-saving ignition protected fuse for 30 to 300 Amp loads. Must use with Terminal Fuse Block.

www.bluesea.com

 

[FilterGuy]

When you add a bms I guess you could multiply by 2 again

Yes for FET Based BMSs..... but if you are using the BMS to turn loads/charges on and off, you don't get that multiplier.

Either way... you get the point. How do you estimate the max fault current. Trying to come up with the resistance of everything in the circuit would be difficult at best..... and error prone at worst.

Where I really scratch my head is the small fuses. Does the small 10A fuse in the diagram above need to be able to break 6000A (or even 3000A) Fault current? Do they even make a 10A fuse that could do that?

 

[John Frum]

Yes for FET Based BMSs..... but if you are using the BMS to turn loads/charges on and off, you don't get that multiplier.

Either way... you get the point. How do you estimate the max fault current. Trying to come up with the resistance of everything in the circuit would be difficult at best..... and error prone at worst.

Where I really scratch my head is the small fuses. Does the small 10A fuse in the diagram above need to be able to break 6000A (or even 3000A) Fault current? Do they even make a 10A fuse that could do that?

I get your point and don't forget the shunt.
We don't fuse the individual cells(yet).
I think the physical security of a box mitigates most of the risk inside of the battery assembly including the shunt and/or bms.

 

[Hedges]

For AC wiring, the approach is for things to hold together long enough for the main protection to interrupt.
Branch circuit breakers are rated 10kA interrupting and the main 22kA.
Prospective fault current is determined by utility transformer DC resistance. Wire resistance at various places reduces what some circuits see.

There are "current limiting" fuses such as Class T which can limit the time a 200 kA fault current flows such that a main breaker it is protecting experiences the equivalent energy it would have seen from a 20 kA fault.

I looked over the limiting curves for those fuses, and at lower faults like 20 kA they do NOT provide proportionately lower (e.g. 2kA) stress on balance of components; those still experience 20 kA.

So I think low rated fuse may blow apart, and if current flows long enough through plasma the higher amperage Class T will stop the current flow.

So yeah, the box is probably what protects you.

 

[John Frum]

For AC wiring, the approach is for things to hold together long enough for the main protection to interrupt.
Branch circuit breakers are rated 10kA interrupting and the main 22kA.
Prospective fault current is determined by utility transformer DC resistance. Wire resistance at various places reduces what some circuits see.

There are "current limiting" fuses such as Class T which can limit the time a 200 kA fault current flows such that a main breaker it is protecting experiences the equivalent energy it would have seen from a 20 kA fault.

I looked over the limiting curves for those fuses, and at lower faults like 20 kA they do NOT provide proportionately lower (e.g. 2kA) stress on balance of components; those still experience 20 kA.

So I think low rated fuse may blow apart, and if current flows long enough through plasma the higher amperage Class T will stop the current flow.

So yeah, the box is probably what protects you.

I blew my first fuse with my 8s 280 amp hour pack the other day.
It was a 10 amp fuse and it popped pretty loudly accompanied by a flash.
I've been a fan of mrbf fuses for a while but I went with a class t as close to the main positive terminal as possible.
I'm glad they don't make a 6mm carrier or the decision would have been harder.

 

[FilterGuy]

I get your point and don't forget the shunt.

Yup, that will intentionally add ~.1 -.2 mOhm. That reduces the current a little.

I think the physical security of a box mitigates most of the risk outside of the battery assembly including the shunt and/or bms.

True, but a lot of us don't have the stuff all inside a fireproof box.

So I think low rated fuse may blow apart, and if current flows long enough through plasma the higher amperage Class T will stop the current flow.

I was thinking about that.... That counts on the resistance of the plasma remaining *very* low in order to blow the Class T. I have no idea if that is the case or not. (I have never put an ohm meter on a plasma arc )

Where I really scratch my head is the small fuses. Does the small 10A fuse in the diagram above need to be able to break 6000A (or even 3000A) Fault current? Do they even make a 10A fuse that could do that?

BTW: I am 90% sure the answer is 'no'. Otherwise there would have been problems with the tiny automotive fuses years ago even with LA batteries. However, I am looking for the logic for why that would be true.

Here are a couple reasons why the small fuse may not need a huge AIC rating:
* The main fuse adds resistance
* There are more connections between the fuse and the battery so the resistance would be higher.

Both of the above points are true.... but is that enough proof? I would like to be somewhat definitive on selecting safety devices.

 

[FilterGuy]

I blew my first fuse with my 8s 280 amp hour pack the other day.
It was a 10 amp fuse and it popped pretty loudly accompanied by a flash.

Could you tell me more? What kind of fuse was the 10A? did it damage the holder? Was there a large fuse upstream from it?

The fact that you heard a pop ans saw a flash tells me there was a lot of current when it blew. I have poped a few automotive fuses over the years and don't ever recall hearing or seeing anything.....Lead Acid is much higher resistance.

Sometimes, experience can tell you what works even if you can't prove it.

 

[FilterGuy]

As an aside, Many years ago I was about 10 feet from a power pole when one of the big fuses they have on the transformers blew. I was not facing it so I did not see anything but the sound was like an bomb going off right over my head. I about soiled my shorts.

 

[John Frum]

Could you tell me more? What kind of fuse was the 10A? did it damage the holder? Was there a large fuse upstream from it?

10 amp bluesea ATC fuse in a cheap amazon fuse block.
Carrier was fine and the fuse legs looked fine I think.
I think its in the trash now.
I was setting the voltage on one of these https://www.amazon.ca/gp/product/B078Q1624B
I blew up 2 of them
First one was before I got my batteries so it received it's power from my 25amp@25volts ac2dc charger. That time was a typical blown fuse. I had to check to make sure the fuse blew.
Did the same thing with the replacment buck converter and although I wasn't looking at the fuse I think the pop was significantly louder and my eye registered the flash even though I wasn't looking at the fuse block.
I have a class t upstream and an mrbf fuse.
Neither of those blew.

 

[FilterGuy]

10 amp bluesea ATC fuse in a cheap amazon fuse block.
Carrier was fine and the fuse legs looked fine I think.
I think its in the trash now.
I was setting the voltage on one of these https://www.amazon.ca/gp/product/B078Q1624B
I blew up 2 of them
First one was before I got my batteries so it received it's power from my 25 amp ac2dc charger. That time was a typical blown fuse. I had to check to make sure the fuse blew.
Did the same thing with the replacment buck converter and although I wasn't looking at the fuse I think the pop was significantly louder and my eye registered the flash even though I wasn't looking at the fuse block.
I have a class t upstream and an mrbf fuse.
Neither of those blew.

Thanks:
To me, your story definitly says the AIC is something to consider even for the small fuses. You had the resistance of *two* Large fuses up stream and it was still a significant event!

 

[John Frum]

Thanks:
To me, your story definitly says the AIC is something to consider even for the small fuses. You had the resistance of *two* Large fuses up stream and it was still a significant event!

I didn't think about it much at the time.
I'm thinking about it now though.

 

[cinergi]

The way I've been thinking about this is the Class T needs to protect from a dead short across the output of the fuse and direct contact with the other battery terminal (nothing else in between). Downstream fuses need to protect the wires. It would be ideal to step down from 300 amps to 3 amps in stages BUT if the 3 amp fuse is protecting a small wire, then a short somewhere along that wire isn't going to be subject to the same available fault current at the battery terminals. If you short something directly AT the 3 amp fuse WITH a conductor that's large enough, there's a high chance that fuse will not properly interrupt -- and therein lies the question -- would the Class T blow? While researching materials to help me answer this question, I came across this video -- I think it's really important to understand what happens in a DC arc fault. The initial spark while connecting is small and the contacts must be very close together. Look how far away the contacts can be when interrupting the circuit! That's what people who say "12vdc can't travel past .004 inches" (or whatever) are dead wrong about the failure scenario:

What I don't know is the resistance of the circuit that's being completed by the plasma arc. If it's not that high, then yes - the class T fuse would blow.

 

[HRTKD]

Where I really scratch my head is the small fuses. Does the small 10A fuse in the diagram above need to be able to break 6000A (or even 3000A) Fault current? Do they even make a 10A fuse that could do that?

I may have stepped in it earlier today when my answer to someone with a (somewhat) similar question was that there were two wires, so two fuses. The fuse protects the wire for that segment. If the device cannot be expected to exceed 10 amps, then a 10 amp fuse is appropriate. However, if I'm understanding the "rules" right, we would be looking at no more amps on the wire than 80% of the fuse rating.

 

[Hedges]

Where I really scratch my head is the small fuses. Does the small 10A fuse in the diagram above need to be able to break 6000A (or even 3000A) Fault current? Do they even make a 10A fuse that could do that?

Hopefully not, but yes they do:

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

Also 11A, for my DMM:

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

 

[Bob B]

OK .... I give up ... What is AIC? I tried to google it, but I don't think it is all about blood sugar levels.

 

[John Frum]

OK .... I give up ... What is AIC? I tried to google it, but I don't think it is all about blood sugar levels.

DC Circuit Protection - Blue Sea Systems

Engineering high quality marine electrical components for safety, reliability and performance

www.bluesea.com

 

[Bob B]

Thanks smoothjoey .... so now I have another 3 digit anachronym to have to remember.

I can see where AIC would be a consideration with a contact device like a switch or circuit breaker .... but I have a hard time seeing how it would come into play in the case of a fuse .... seems to me that the fuse is going to open no matter what .... has anyone ever seen a fuse fuse?

 

[Hedges]

Thanks smoothjoey .... so now I have another 3 digit anachronym to have to remember.

I can see where AIC would be a consideration with a contact device like a switch or circuit breaker .... but I have a hard time seeing how it would come into play in the case of a fuse .... seems to me that the fuse is going to open no matter what .... has anyone ever seen a fuse fuse?

Ferraz Shawmut / Mersen Sales | UL Power Fuses

The better your overcurrent protection, the easier it is to minimize downtime and repair bills. Ferraz Shawmut's Amp-Trap 2000 current-limiting fuses open in less than one-quarter cycle to provide the Type 2 protection that prevents costly equipment damage.

ferrazshawmutsales.com

 

[HRTKD]

Thanks smoothjoey .... so now I have another 3 digit anachronym to have to remember.

I can see where AIC would be a consideration with a contact device like a switch or circuit breaker .... but I have a hard time seeing how it would come into play in the case of a fuse .... seems to me that the fuse is going to open no matter what .... has anyone ever seen a fuse fuse?

An AC breaker can't handle that arc. A DC breaker can. At least that's my understanding.