Actually diodes are quite robust, it is the filter caps that need protecting. Case in point: a 3A rectifier shunting a 600A lightning strike around a PFC circuit.

 

And, for something more concrete, one that we use at work, http://www.diodes.com/datasheets/ds21221.pdf is a 25A bridge that is rated for a 350A half cycle sine.

In other words you generally do not have to worry till you get well beyond 10X rated current. Generally your system resistance is enough to keep you under that limit.

 

Of course you are right but you still need to check the surge rating.

For a small 1.5A rectifier used in a little linear supply, this isn't normally a problem. A 1.5A rectifier can hapilly stand 25A and the internal resistance of the transformer will limit it to a safe level.

However, when you're running the circuit directly from the mains, it's a different story.

What's the peak current through the rectifier?

How many cycles does it take for it to charge?

If the resistance of the mains is about 0.5hm: and the peak voltage is 325V, the peak current would be 650A so you think your rectifier would be toast? Suppose your capacitor is 1000µF, then the surge would decay rapidly to within 63% within 500µs so the capacitor will be as good as fully charged within 5ms.

So as we've established a 25A rectifier will be fine but what about the 5A rectifier you initially suggested? Somehow I think it'll be toast.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

Wonder if there is any Planar Transformer of Pulse Technitrol match the switching transformer spec detail in attachment.

Attached Files

Any Pulse Technotrol Planar Transformer equilvalent.zip (194.3 KB, 18 views)

 

difficulty buying or finding equivalent for following:
PH4840, Si7370 and PH21NQ15
BCX55

from Fig 16, DC to DC converter circuit, of page 21.

Anyone can review and offer solutions.

Attached Files

	258908_1.pdf (255.1 KB, 8 views)
	lt1952.zip (162.5 KB, 11 views)

 

http://www.vishay.com/docs/88931/s5a.pdf
A 5A diode is good for 100A peak 8.3mS surge on top of the full load current.

That makes for an RMS of 70A for 8.3mS at 1.5V or 106W.

While not exact, it should be able to take a single pulse of 650A for 8.3*106/650A/1.8V=0.75mS.

Looking at the diode's I/V curve and projected to 650A gives approx 1.8V, the ratio of the wattages tells you how long it takes 650A to heat the crystal the same amount as the 70A RMS pulse rating does.

 

Even by your assumption (0.5hm: source ideal 1000µF capacitor) the current will be down to 300A in 0.5mS. A typical 400V 1000µF cap will add 0.25hm: to the 0.5hm: line knocking it down to 433A.

That brings us to 8.3*106/(433*1.8)=1.1mS ... you can add to this the exponential decay of the capacitor current to get an even better idea of the peak inrush current the diode will handle since this is a quick square pulse approximation.

 

What mains voltage are you using in you calculations? I used 230VAC.

You have point about the ESR of the capacitor reducing the current, then there's any inductance in the line which should also help.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

I think it is ok to assume that input current to a DC to DC converter (having 40Vdc input and output 13Vdc 10Amp) is less than 0.5Amp.

Appreciate comments, yes?

 

Hopefully you meant 240V and not 40V?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Actually I scaled from your 650A.

I would point out however that your 230V line can easily surge to 400V peak. That is only 10% variation (considered normal) and 10% surge, hence most universal PFC circuits running a 400V rail.