Hi everyone, I’m designing a three-phase inverter that receives an input voltage range of [42 .. 52]V from a power supply and draws a maximum of 4A.

However, there is a inrush (when it is switched on) like this:

I DON'T need to reduce this inrush current in the current application; let's just focus on the fuse, please


I need to select a fuse to protect the circuit (to be placed in the green circle), bearing in mind that the 48 Vdc will be split between two DC-DC converters:

Please disregard the calculations; they may be incorrect and shouldn’t be used to select the fuse, as I personally measured the maximum 4A input using a current probe on the 48V supply of a PCB I’m currently using with the motor and the load… so please focus on the block diagram. Obviously, if you spot any obvious mistakes, please let me know, even though that isn’t the purpose of this question


While reading through a few application notes online, I found that:

1) Vfuse > 1.1 * Vi,max

2) Ifuse >= Ii,max / (Fsafe * Ftemp) .. dove Fsafe=0.75 to prevent premature ageing by operating it at a lower current; Ftemp depends on the thermal derating curve of the selected fuse

3) I2T calculated using the following formula found here:

in my case: I2T = 1/2 * (15A)^2 * 16ms = 1.8 A*s^2

Could you confirm whether these three points are correct?

I’ve chosen this fuse – let me know if it’s OK:




4) I know I could have considered fitting fuses here too, but perhaps we should look at this case once we’ve worked out how to size the one upstream…



Thanks.

 

I DON'T need to reduce this inrush current in the current application; let's just focus on the fuse, please

You're approaching this wrong.

The fuse should be sized for the application, not the inrush current. To deal with inrush current, you should use an NTC thermistor to limit it to an acceptable value for the fuse.

 

You're approaching this wrong.

The fuse should be sized for the application, not the inrush current. To deal with inrush current, you should use an NTC thermistor to limit it to an acceptable value for the fuse.

Hello, and thank you for your reply.

But wouldn’t a hypothetical peak inrush current risk blowing the fuse?

 

But wouldn’t a hypothetical peak inrush current risk blowing the fuse?

Not if you size the inrush limiter appropriately.

 

Not if you size the inrush limiter appropriately.

There is already another circuit board that limits the inrush current, and the oscilloscope waveform shows the result.

Assuming there is nothing else that can be done about the inrush current, given that peak, have I ‘chosen’ the wrong fuse?

 

I’m designing a three-phase inverter that receives an input voltage range of [42 .. 52]V from a power supply and draws a maximum of 4A

Assuming there is nothing else that can be done about the inrush current, given that peak, have I ‘chosen’ the wrong fuse?

In my opinion, yes, you've chosen the wrong fuse value. You mentioned a maximum current of 4A, but you're selecting a 7A fuse? That seems wrong.

Is the actual maximum current 4A? Or is that just what you happened to measure at a particular input voltage and load?

It looks like the inrush current drops off after around 12ms. Maybe a slow blow fuse (I haven't looked at fuse specs recently) would be sufficient.

 

In my opinion, yes, you've chosen the wrong fuse value. You mentioned a maximum current of 4A, but you're selecting a 7A fuse? That seems wrong.

Is the actual maximum current 4A? Or is that just what you happened to measure at a particular input voltage and load?

It looks like the inrush current drops off after around 12ms. Maybe a slow blow fuse (I haven't looked at fuse specs recently) would be sufficient.

I measured 4A, but I’m playing it safe by choosing one with a slightly higher current rating; I’m not sure if that’s a mistake. Furthermore, there shouldn’t be any components that would be damaged at that amperage

 

I measured 4A, but I’m playing it safe by choosing one with a slightly higher current rating; I’m not sure if that’s a mistake.

Guessing is usually a mistake. You should select the fuse for the maximum load current; I'm assuming that the rest of the circuitry was designed properly.

 

Look at time-delay (slo-blo) fuses as dl324 suggested.
They are intended to tolerate momentary high inrush currents without blowing.

The standard design standard for fuse rating is about 1.25 of the maximum load current, so a 5A fuse for your 4A load.

 

the inrush current is peaking at 15A and lasting some 10ms. typical slow blow or time delay fuse handle usually 150-200% of rated current for several seconds. the extremely large values such as 10x the rated current will trip in few ms. as already stated 5A time delay fuse will be appropriate