advertisement


Rectifier specification

I

IDM

pfm Member
Hi,

I am in the process of designing a powersupply for my CD player. It is based around centre tapped transformer feeding 2 rectifier diodes and a 10,000uf smothing capacitor.

Whilst I know the current demands of the CD player are modest, it suddenly struck me that the biggest stress the rectifier might see is in-rush current to the smoothing cap. Is there a formula for calculating the in-rush current, is this an issue that might lead to rectifier failure?

Cheers,

Ian
 
Rectifiers have a specification for the non repetitive peak current to deal with exactly this problem. For example, if you look at http://www.fairchildsemi.com/ds/1N/1N4001.pdf, you will see that even this small, cheap diode is good for 30A in a single half cycle. Now, 30A flowing for 3ms gives 9V of voltage rise on 10,000uF, so that is probably more than enough.

To actually estimate how big the current surge might be, you need to know a bit about the transformer; the regulation and the rated current and voltage. For arguments sake, lets suppose you have a 20V nominal CT winding, rated at 3A, with say 15% regulation (this numbers are picked out of thin air, but not atypical). Then the no load voltage is 15% higher, or 23V. 3A of current drops this back to 20V, so the total resistance is (23-20)/3 = 1Ohm.

You are only using half the winding at a time, so you will get 11.5Vrms (no-load), with a resistance of 0.5 Ohm. The peak voltage will be about 16V (= sqrt(2)x11.5). The ESR of the cap is much smaller than winding resistance (maybe 30milliOhms), and can usually be neglected for a rough estimate. Knock 0.7V off for the diode drop (0.4V if using a schottky diode), gives you say 15V and 0.5 Ohm, just giving a worst case peak current of 30A - if you turn it on exactly at the peak of the wave.

The upshot is that even a dinky diode can handle quite big inrush values.
 
Many thanks for the explanation. I am planning to use 11DQ09 schottky diodes which having now checked the data sheet indicates a max peak one cycle non-repetitive surge current of 85A. The PSU has three seperate transformers, 2 of 12-0-12, and 1 of 18-0-18. I therefore guess from your explanation I am unlikely to have a problem with the diodes failing.

Cheers,

Ian
 
I would be amazed if there could be a problem, but it is worth doing the sums, putting the transformer regulation figures in.
 
Yes, it does.

It also sorts out the much more serious and annoying problem of arcing switches, mains fuses blowing and MCB tripping when you start up a big transformer. In addition to the capacitor inrush current, there can be a much bigger magnetisation current surge, especially for toroidal transformers. This causes the problems.

Not all power amp circuits are compatible with soft-start; some class A designs give real trouble, as they draw so much current during soft-start that the power supply never gets anywhere near full voltage, and you get a massive current surge at the end of the sot-start period. But there is no problem for most sane circuits.
 
You must log in or register to reply here.


advertisement


Back
Top