Author Topic: paralleling fuses  (Read 9755 times)

0 Members and 1 Guest are viewing this topic.

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
paralleling fuses
« on: January 26, 2015, 12:40:27 pm »
is it wise to put fuses in parallel to make up the total capacity required particularly where inrush currents are involved ?
 

Offline bobcat

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: paralleling fuses
« Reply #1 on: January 26, 2015, 12:44:38 pm »
No, very bad idea. The circuit must open all at once. If dealing with high inrush, use a slow blow fuse.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: paralleling fuses
« Reply #2 on: January 26, 2015, 12:49:25 pm »
The customer is complaining that everything worked fine for a while but now that have blown the fuses. They sent their diagram over and it has 4 fuses and I beleive the inruse current is 75% of the total fuse capacity so it only take for one weak fuse to go one time and next time the other 3 go, or another 1 goes and then the next time the last two go. Unless current is shared very equally with exactly the same length wiring I'm assuming that 1 in 4 can quickly get over stressed and blow and then they go like cristmas lights
 

Offline Wilksey

  • Super Contributor
  • ***
  • Posts: 1329
Re: paralleling fuses
« Reply #3 on: January 26, 2015, 12:55:30 pm »
No, 1 fuse only!
 

Offline mux

  • Regular Contributor
  • *
  • Posts: 119
Re: paralleling fuses
« Reply #4 on: January 26, 2015, 12:56:02 pm »
It is possible to parallel fuses and it is fundamentally safe (because a fault condition that exceeds the combined fuse rating will necessarily blow all fuses), but the carrying capacity of the fuses will not scale linearly. Meaning:

Let's say we're parallelling 4 fuses rated at 1A carrying, 5A guaranteed blowing capacity (both half-sine 10ms).
- These fuses will be guaranteed to blow at 5*4=20A
- These fuses will NOT be guaranteed to carry 4*1=4A. This will be somewhere between 1 and 4A (i.e. the carrying capacity of one fuse).

Statistically, you can take sqrt(n) as the guaranteed carrying capacity, i.e. about 2A. So all inrush and non-fault conditions should be guaranteed by design to be less than that, otherwise you will end up in a situation like you're in now, where possible one or two out of four fuses have blown.
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: paralleling fuses
« Reply #5 on: January 26, 2015, 12:59:14 pm »
Well unfortunately I can only but speculate but my datasheets tells me that the load may cause an inrush current of 80A and they have 4x30A fuses. I'm guessing that at those currents every extra scrap of wire length makes a difference and eventually the weaker of the 4 will blow.
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21686
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: paralleling fuses
« Reply #6 on: January 26, 2015, 03:54:59 pm »
What are we talking?  Surge into a film cap?  Electrolytics?  Motor starting??

As mentioned, paralleling is possible but usually ill advised (since you'll get... what we have here, basically).  Same advice applies to other things that share current grudgingly, like IGBTs connected in parallel (which need to be thermally matched as well; although to be fair, the fuses should be too, as the burnout current has a tempco).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline grumpydoc

  • Super Contributor
  • ***
  • Posts: 2905
  • Country: gb
Re: paralleling fuses
« Reply #7 on: January 26, 2015, 04:04:08 pm »
Quote
Well unfortunately I can only but speculate but my datasheets tells me that the load may cause an inrush current of 80A and they have 4x30A fuses. I'm guessing that at those currents every extra scrap of wire length makes a difference and eventually the weaker of the 4 will blow.
Why was that arrangement chose - I mean, it's not like 80A fuses don't exist.
 

Online electr_peter

  • Supporter
  • ****
  • Posts: 1302
  • Country: lt
Re: paralleling fuses
« Reply #8 on: January 26, 2015, 04:12:48 pm »
Do not connect fuses in parallel if you want to have reliable system with predictable behaviour. You are asking for trouble if you try this scheme.

Main problem is that if load requires 5A, putting, say, 1A+4A fuse combination would not guarantee 1:4 current sharing between fuses. This leads to premature blowing of overloaded fuse and subsequent failure of other fuse.
There are practical problems too - if one fuse blows, load can still work without you noticing. You should replace all fuses at once. If fuses are being replaced in a live circuit (by mistake), first put fuse will blow immediately as well.
 

Online ConKbot

  • Super Contributor
  • ***
  • Posts: 1385
Re: paralleling fuses
« Reply #9 on: January 26, 2015, 04:19:42 pm »
80A isnt a particularly huge amount as far as fuses are concerned?  As far as AC power stuff goes, time delay RK-5 fuses will do 500% for 10 seconds, and littlefuse FK3 maxi automotive fuses do 200% for 2 seconds.  Is this a mains AC application? DC? Nominal current vs inrush current, and duration of the inrush?  There should be a fuse somewhere that covers your needs. 
 

Offline SimonTopic starter

  • Global Moderator
  • *****
  • Posts: 17816
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: paralleling fuses
« Reply #10 on: January 26, 2015, 05:39:00 pm »
It's for 300V electric heaters, I have spoken to the customer and it emerges that there are 2 in parallel on the positive and 2 in parallel on the negative,

As it happens the fuses blowing are not the problem, the problems are elsewhere and they acknowledged that the parallel idea is not a good one.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf