fuse and breaker testing

[John Frum]

It may not be practical to test very large fuses but it should be possible to test smaller ones
By smaller I would say < 60 amps
Not a perfect test but not useless either

environment
    Fire safe space(think mythbusters on sweeps week)
    plexiglass sheild at a safe distance
    protective clothing and eyewear
    properly rated fire extinguisher

agm_battery + -> quick_blow fuse_250_amp -> 4/0 cable -> disconnect_switch_300 amp -> bus_bar_250_amp -> fuse_or_breaker_under_test

agm_battery - <- 4/0 cable < -------------------------------------------------------- bus_bar_250_amp <- fuse_or_breaker_under_test

baseline:
    open disconnect switch
    connect the - and + bus_bars together with a copper_bar
    close the disconnect switch
    move to a safe distance behind a plexiglass screen

expected results
    The quick blow fuse should fail open within 5 seconds
    test setup should not show any indications of stress other than the blown fuse

If we observed the expected results in the baseline proceed to test

test:
    open the disconnnect switch
    If necesary replace the mrbf fuse and recharge battery
    connect the - and + bus_bars via the fuse or breaker under test
    close the disconnect switch
    move to a safe distance behind a plexiglass screen

expected results(pass)
    the fuse or breaker under test should open within 5 seconds

observed results
    ?

 

[BiduleOhm]

5 sec for a dead short? I hope the fuse is quicker than that to break the current in those conditions, like a few dozens of ms.

You probably want a slow blow for the MRBF fuse so it doesn't blow before the DUT

 

[John Frum]

You probably want a slow blow for the MRBF fuse so it doesn't blow before the DUT

Dang, I wonder if this test setup is going to give reliable results even with a slow blow fuse.

 

[John Frum]

I wonder if a single lifepo4 cell without bms would be safer than an agm battery.
A single fortune cell for example.
Would the voltage difference impact the test?

 

[BiduleOhm]

It would limit the current to a lower value.

But then it all depends on what you want to test about the fuse: ultimate current it can break? how long it takes to blow with x % overcurrent? ...?

 

[John Frum]

It would limit the current to a lower value.

But then it all depends on what you want to test about the fuse: ultimate current it can break? how long it takes to blow with x % overcurrent? ...?

Its more what can be tested than what I want to test.
I guess if the baseline test works as expected then it should be good.

Its got me thinking, In a real world lithium system there are likely/typically 3 layers of over current protection...
BMS, main fuse, and branch fuse.
I suspect its not completely deterministic which will trip first especially between the BMS and the main fuse.

 

[John Frum]

I guess a normally open relay/contacter could be put in line with the disconnect switch.
Then the tester can close the circuit remotely and safely.

 

[John Frum]

I wonder how big a buck converter one can got off the shelf?
I would guess UL has a test rig and protocol for testing fuses/breakers.

 

[BiduleOhm]

Its more what can be tested than what I want to test.
I guess if the baseline test works as expected then it should be good.

Its got me thinking, In a real world lithium system there are likely/typically 3 layers of over current protection...
BMS, main fuse, and branch fuse.
I suspect its not completely deterministic which will trip first especially between the BMS and the main fuse.

A lot of things can be tested but you'll need equipment (like an oscilloscope) for a lot of them.

It depends on the fault type; there's a lot of differences between a mild over current and a dead short.

I guess a normally open relay/contacter could be put in line with the disconnect switch.
Then the tester can close the circuit remotely and safely.

Well, you need a relay who is capable of switching a few kA to a few dozens of kA so... yeah... I'd just make my own big switch with copper bars and some string to actuate it from a distance; simple, reliable and cheap.

I wonder how big a buck converter one can got off the shelf?

Pretty big ones... The biggest I can find on TME is a 10 kW Mean Well PSU: https://www.tme.eu/fr/en/details/rst-10000-48/built-in-power-supplies/mean-well/ but we're talking more than 2 k€ without taxes and shipping... NB: that's a PSU, not a buck converter (but technology is almost the same between the two anyway so that gives you a pretty good idea).

Actually that would make one hell of a charger... Want to charge your 200 Ah 16s pack at 1 C rate? well, you can ? and with the 91 % efficiency you have a ~1 kW of heat for free ?

I would guess UL has a test rig and protocol for testing fuses/breakers.

Didn't look but if I had to guess I'd say they use a capacitor bank because you can easily control the voltage and current but it's not cheap so not really a good solution for us.