Single vs. Multiple Class-T Fuses

[cs1234]

I have a 100 amp DC breaker on each of my 10 280aH batteries. I have a 125 amp class. T fuse at the midpoint between cell 8 & 9 of each battery. Then I have a main 250A DC breaker on each Sol Ark 12K.

What 100 amp DC breaker are you using?

 

[cods4]

Is this cells or batteries?
Perhaps 2000 amps is empty and 4000 amps is full?

Its a single cell.

Obviously it would only decrease slightly for any number of cells in series (someone did the calculations for that earlier in the thread).

And if you are paralleling up cells then it would increase. I.e for a 2p16s battery made of LF280K cells it would be ~4000-8000A.

It seems fairly well understood from what I've read that temperature also affects these values by quite a lot, probably a lot more so than SOC.

I'd suggest that it would be 4000A at around 40-55°C, and 2000A at < 0°C.

SOC would probably only have an effect at quite low values.

 

[John Frum]

Its a single cell.

Obviously it would only decrease slightly for any number of cells in series (someone did the calculations for that earlier in the thread).

That was me.

It seems fairly well understood from what I've read that temperature also affects these values by quite a lot, probably a lot more so than SOC.

My mistake, I was assuming standard test conditions of 25C.

SOC would probably only have an effect at quite low values.

I consider my numbers a 0 order estimate.
I suspect the voltage sag has to be significant into a dead short and I have not accounted for that.

 

[John Frum]

My ~1500 watt tea kettle boils a liter of ~25C water in ~200 seconds.
At 4000 amps a big blue prismatic cell would be something like a ~13000 watt sealed tea kettle.
Does that seem roughly accurate?

 

[cods4]

I suspect the voltage sag has to be significant into a dead short and I have not accounted for that.

Yeh I think that's the case. The chemical reaction in the cell can't keep up with the short circuit situation and the voltage drops right off.

 

[cods4]

My ~1500 watt tea kettle boils a liter of ~25C water in ~200 seconds.
At 4000 amps a big blue prismatic cell would be something like a ~13000 watt sealed tea kettle.
Does that seem roughly accurate?

Not exactly since we expect the voltage will not be staying at 3.2V.
Instead we can calculate the power from the cell using P=I²R.
If we assume the internal resistance will remain the same. Then P=4000²x0.0002 = 3200W

And the voltage of the cell can be calculated with V=IR. So the cell voltage will be as low as 0.8V.


These are all pretty rough calculations since we don't know exactly how EVE measured the short circuit current.

 

[John Frum]

In light of new info from @cods4 it seems mrbf and anl fuses have sufficient breaking capacity for simple 24 volt lfp batteries.

 

[John Frum]

On the other hand...
I wonder if the dead short ampacity is higher than 4000 amps for the first milliseconds.
Possibly long enough to form a plasma filament across the air barrier in the blown fuse.
Once the plasma filament is established it may sustain at lower current and voltage.

 

[cs1234]

On the other hand...
I wonder if the dead short ampacity is higher than 4000 amps for the first milliseconds.
Possibly long enough to form a plasma filament across the air barrier in the blown fuse.
Once the plasma filament is established it may sustain at lower current and voltage.

So it will fix itself then.

 

[John Frum]

So it will fix itself then.

You would still have 16 ~3000 watt electrolyte kettles.

 

[cs1234]

You would still have 16 ~3000 watt electrolyte kettles.

When life gives you lemons!

 

[John Frum]

When life gives you lemons!

Its got electrolytes!

 

[RV10flyer]

What 100 amp DC breaker are you using?

CHYTI 100A/600V DC/10,000A

 

[Mithril]

On the other hand...
I wonder if the dead short ampacity is higher than 4000 amps for the first milliseconds.
Possibly long enough to form a plasma filament across the air barrier in the blown fuse.
Once the plasma filament is established it may sustain at lower current and voltage.

Someone should convince Electroboom or styropyro to do an experiment and see the actual volts and amps required to sustain an arc for various fuses

 

[Mithril]

Ok, so how do people feel about the need for T-class fuses "beyond" the main system fuse(s). Assume batteries and battery banks are all fused correctly (enough AIC, correct size to protect wires). Is there any actual NEED for T-class level AIC on branches? In theory if a 50A fuse blows and there's a 5000A arc the main T-class is ALSO going to blow and it has sufficient AIC. I think the only concern is if you used a fuse with such a low AIC that it could form an arc that would NOT kick the main fuse(s).

 

[John Frum]

Ok, so how do people feel about the need for T-class fuses "beyond" the main system fuse(s). Assume batteries and battery banks are all fused correctly (enough AIC, correct size to protect wires). Is there any actual NEED for T-class level AIC on branches? In theory if a 50A fuse blows and there's a 5000A arc the main T-class is ALSO going to blow and it has sufficient AIC. I think the only concern is if you used a fuse with such a low AIC that it could form an arc that would NOT kick the main fuse(s).

High AIC fuses are only required on the battery(feeder) circuits.
The branch circuits can get away with lesser fuses.
The branch fuses can break on over-current, but are not guaranteed to break a dead short.
In a dead short on a branch the BMSs will likely cascade trip if the branch fuse sustains an arc.
If any of the BMSs fail closed then the high AIC feeder fuses are the last line of defense.

 

[John Frum]

Also the further away from the battery, the higher the path resistance.

 

[Mithril]

High AIC fuses are only required on the battery(feeder) circuits.
The branch circuits can get away with lesser fuses.
The branch fuses can break on over-current, but are not guaranteed to break a dead short.
In a dead short on a branch the BMSs will likely cascade trip if the branch fuse sustains an arc.
If any of the BMSs fail closed then the high AIC feeder fuses are the last line of defense.

Ok that aligns with my thinking, and how I've already designed my system. I do think I'm going to replace certain fuses with higher AIC for a bit of "belt and suspenders" but if 2000-4000AIC works for the non T-class fuses that does lower the cost.

Re: Path resistance; Very god point! It's sometimes easy to forget the *reason* for the high possible fault current is due to the insanely low resistance of the cells/batteries.

 

[John Frum]

Ok that aligns with my thinking, and how I've already designed my system. I do think I'm going to replace certain fuses with higher AIC for a bit of "belt and suspenders" but if 2000-4000AIC works for the non T-class fuses that does lower the cost.

Re: Path resistance; Very god point! It's sometimes easy to forget the *reason* for the high possible fault current is due to the insanely low resistance of the cells/batteries.

I think its safe to say that the Victron lynx system is the most popular distribution system for experience people.
I just found these.

https://www.littelfuse.com/media?resourcetype=datasheets&itemid=3511fb09-2c05-43b4-a093-f9f22b9f3cdd&filename=mega120v-sf56-datasheet-v16.2

AIC of 2500 amp at 120VDC
From all the fuses that show their breaking capacity at different voltages, I infer that the AIC might be as high as ~5000 amps at 60VDC.

 

[cods4]

I think MCBs or MCCBs are the best option for branch circuits (assuming you mean to feed inverters or charge controllers etc). It's important to be able to isolate the device from the batteries if needed. I.e when working on them, replacing faulty units etc.

I'm going to be using 3 of these for my 3 inverters. (I'm currently using 3 MCBs for my current lead acid battery, but need something that can handle a constant 100A per breaker when i get my LFP batteries.)

BENY BDM Series: DC Circuit Protection for BESS | BENY New Energy

BENY New Energy's BDM series, including BDM-125 and BDM-250, offer robust DC circuit protection for BESS, with a 500V rating and up to 250A current capacity.

www.beny.com

 

[upnorthandpersonal]

Someone should convince Electroboom or styropyro to do an experiment and see the actual volts and amps required to sustain an arc for various fuses

As I mentioned somewhere, something like this is on my todo list. I'll borrow some high current probes from my lab and do some dead short tests on a 48V pack. It takes some time to set this up, and I will do this outside, so once we hit spring. Not doing that in -20C. I also have some Class T and BS88 fuses (and a few BMS) that I can sacrifice.

 

[AHTrimble]

I think its safe to say that the Victron lynx system is the most popular distribution system for experience people.
I just found these.

https://www.littelfuse.com/media?resourcetype=datasheets&itemid=3511fb09-2c05-43b4-a093-f9f22b9f3cdd&filename=mega120v-sf56-datasheet-v16.2

AIC of 2500 amp at 120VDC
From all the fuses that show their breaking capacity at different voltages, I infer that the AIC might be as high as ~5000 amps at 60VDC.

Since I am using a Lynx PowerIn as my busbar for my 200ah batteries, I could easily add the bolts for these fuses at a very reasonable price.
I found them here for a $6.27. [https://www.waytekwire.com/item/47255/Littelfuse-0888250-U-2M8-High-Performance-MEGA/]
What is your opinion on using the 250a version on each battery?

 

[John Frum]

Since I am using a Lynx PowerIn as my busbar for my 200ah batteries, I could easily add the bolts for these fuses at a very reasonable price.
I found them here for a $6.27. [https://www.waytekwire.com/item/47255/Littelfuse-0888250-U-2M8-High-Performance-MEGA/]
What is your opinion on using the 250a version on each battery?

The lynx power in has to be modified to accept fuses.
Also, batteries should be fused as close to the positive terminal.
This minimizes the chance of an un-protected short.

 

[Danke]

The lynx power in has to be modified to accept fuses.
Also, batteries should be fused as close to the positive terminal.
This minimizes the chance of an un-protected short.

As he said, “ I could easily add the bolts for these fuses”

 

[John Frum]

The lynx power in has to be modified to accept fuses.
Also, batteries should be fused as close to the positive terminal.
This minimizes the chance of an un-protected short.

Also I'm guessing that the breaking capacity is ~5000 amps at 60VDC.