Re: (ET) What size fuse

[Robert Adsett <subscriptions aeolusdevelopment com>]

At 12:52 PM 1/2/06 -0500, Bob Murcek wrote:
> Would it be good to also incorporate both "last-resort" and running fuses 
> into the layering?  For instance use fast but very high current fuses for 
> the A's, and slow-blowing but closer to expected current fuses for the 
> C's?    I guess this is sorta what they do in cars with the blades and 
> circuit breakers for running but a fusible link for disasters.

I think it would be better just to size the fuse correctly to begin 
with.  If the fuse is sized low enough then the high current case takes 
care of itself.  It probably won't hurt to add another layer but it may not 
make any difference either.  A fast blow fuse still has the same shape of 
curve as its slower blow cousin so it could end up being protected by the 
lower value fuse or vice versa.  I suspect that if you draw out the time 
curves you will find little to no overlap. I.E. all the protection will 
come from one fuse.

Hmm, let me do a quick calc and see how fast these fuses will blow under 
overload conditions.

Assumptions 36V nominal battery voltage, 10-100 mOhm resistance including 
connector and internal resistance.  That gives a fault current of between 
3600A and 360A.  I think you would have to work quite hard to get below 10 
mOhm.  I normally budget about 1mOhm for a bolted connection.  You can get 
better but I'm not sure you can do it consistently w/o risking physical 
damage to the parts.

A T105 has a 3mOhm Internal resistance.  Add in another 6mOhms for battery 
interconnections and you end up with 24mOhm giving 1500A short.  That's 
probably a high estimate.  Especially since I've neglected the drop due to 
the fuse connections and the fuse itself.  The flooded NiCds that started 
this probably have a much lower internal resistance so they may be able to 
get to the 10mOhm level (or even lower since they are paralled).

CNL/ANL fuse are only 32V rated so they are out.

CNN/ANN are 48V rated and from the wording used it appears they are 
referring to 48V battery in which case the actual rating is higher

Now a 100A fuse will take about 0.1 of sec to blow at 360A  and blow almost 
instantly at 1500A or 3600A.  However a 300A fuse will take minutes, 
perhaps hours to blow at 360A, ~50ms at 1500A and still take 20-30ms at 
3600.  The 5 second point for a 100A fuse is around 200A and for a 300A 
fuse it's around 700A.

700A at 36V would imply a 50mOhm short which is not that hard to 
envision.  That's a long time to be dumping over 25kW into a short.

I don't think it should be necessary to make the fuse that large though, a 
300A fuse implies you need on the order of 10kW of sustained power.   I 
think if you need a 300A fuse in that situation you need to consider 
electronic protection, fuses aren't going to be sensitive enough.

BTW the other fuses I looked at (certainly not an exhaustive search) had 
longer fusing times.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,   be 
they legal, genetic, or physical.  If you don't believe me, try to chew a 
radio signal. "  -- Kelvin Throop, III
http://www.aeolusdevelopment.com/