Heat rejection of MV fuses

[jraef]

Does anyone have a usefull formula for estimating the heat losses in medium voltage fuses based on loading? We are doing heat runs on some equipment and the hottest parts are the fuses and clips. With a 1200kW 11kV load, we are seeing 22kW in total heat loss, of which only 4kW can be accounted for in all the other components combined. That means we are seeing 18kW in losses through the fuses and holders. That is 1-1/2% of the total load! I think there is a problem with the clips, but I don't know what to estimate from the fuses to help prove my point. The fuse mfgr. is slow to respond, so I was hoping someone here already knows.

"Venditori de oleum-vipera non vigere excordis populi"

 

[busbar]

 
Aaak! Something's amiss. 1200kW at 11kV 3ø is 63 amps/ø. 18kW "loss" or 6kW/fuse is a real cooker, with a ?V of 95 volts.

Can the gear be taken offline to take fuseholder resistances with a Kelvin ohmmeter [DLRO] and/or use a low-end IR thermometer on the gear?

 

[Skogsgurra]

What a nice problem j&b!

I have no idea at all, but I am following this with great interest. One question: how do we know that the total heat loss is 22 kW? Difference between almost equal measurements with about equal error margins?

 

[jraef]

Since we are building the gear, the losses are based upon our heat run (temperature rise) test in an environmental chamber and a known load. We did use a thermal scanner to see that the hottest spot was the fuse assembly, but the problem was separating the fuses from the clips.

Just got the fuse info fom the mfgr. (surprise surprise). The 15kV 200E fuses themselves are only rejecting 230W/fuse set, so 690W total! I was right, there is a serious problem with the clip assembly( or maybe the connection).

Thanks for the help anyway guys.



"Venditori de oleum-vipera non vigere excordis populi"

 

[jghrist]

I think busbar has the key. Measure the voltage across various current carrying parts/contacts. Knowing the current and the voltage, you can calculate the loss. 63A·95V = 6000W.

 

[busbar]

 
230W/63V = ~3.7V seems logical.

{For a low-VA current source for heatrise tests with low compliance voltage, a 600:5 C200 CT, a powerstat and some welding cable is something to experiment with. You can also test motor-overload relays and small circuit breakers or fuses.}