Fuse confusion

Thread Starter

joachym

Joined Feb 21, 2022
20
Dear all,

I have a power supply which I want to protect from overcurrent. As per specifics, it should withstand 2A continuously, 2.5A short time (short time not defined).

I decided to use fuses (multiple voltage outputs) and my idea was that:
fast 1A fuse blows at 1.1A in 2ms... so I chose slow fuse...

To my shock, it let through 1.8A for tens of seconds!?! :(

I tried to investigate and realized that the fuses blow according to certain Amp x Time curve - which may take tens of seconds and even hours just over the NOMINAL current... (What is the purpose then?)

To shorten the time (my idea was hundreds of milisecons) I would have to use a fuse with lower ratings. But considering the accuracy of reading from the graph and the fuse manufacture precision itself it would be practically "random"... :(

Is there a way to protect/open the circuit more precisely (2A good, 2.2A not good, switching of within... let's say 1 second)?

I am looking for easy, cheap, and small solution - if any.

Thanks for your advice!
j.
 

BobTPH

Joined Jun 5, 2013
8,995
Current limiting circuits can switch it off quickly. If you want that kind if protection, get a current limiting supply.

Another option is a crowbar circuit which detects over current and creates a near short to blow a fuse quickly.

Bob
 

Ya’akov

Joined Jan 27, 2019
9,170
I concur with @BobTPH. If you want fast action for overcurrent in a narrow range a crowbar is the right thing. Current limiting would probably be even better though since it will prevent overcurrent at the outset. It might not apply to your use case, though, if what you are trying to protect cannot handle the current at the limit for an extended time.

You ask what is that purpose of how fuses are made? Apparently, not your application (absent a crowbar) but in those applications where they are properly used those current/time curves are selected to protect devices from damage by overheating due to excessive current.

Current itself isn't a problem, it is the dissipation of heat caused by the current that does the damage. So, any given device is going to be able to handle a certain amount of current-induced heating in various parts for a certain period of time, possibly quite a long one.

Since current peaks are a normal thing and the heating requires current/over time, sudden short peaks of current that would blow a short acting fuse chosen for the nominal operating current will cause constant blown fuses even though no protection was required.

The fuse is chosen as a proxy for the device and is "blown" by overheating, just like the damage that would occur to the device, just some time before that damage can occur.

So if you want practical protection against overcurrent instantly a crowbar circuit will do that but a fuse without one will not.
 

dl324

Joined Mar 30, 2015
16,935
I tried to investigate and realized that the fuses blow according to certain Amp x Time curve - which may take tens of seconds and even hours just over the NOMINAL current... (What is the purpose then?)
To prevent fires.
Is there a way to protect/open the circuit more precisely (2A good, 2.2A not good, switching of within... let's say 1 second)?
You can add your own current limiting circuit. As I mentioned in your other thread on this topic, it would be best to put the current sense on the unregulated side of the power supply. If you put it on the output, the sense resistor will impact voltage regulation. As I recall, you didn't have a schematic.
 

Thread Starter

joachym

Joined Feb 21, 2022
20
As I recall, you didn't have a schematic.
Hi Dennis,
I don't remember chatting with you previously and I asked for the first time about fuses. Also don't remember being asked for schema... Do you have the right person?
Thank you anyway.
 
Top