# Diode substitution - Is "Reverse Current" something I need to consider?

Discussion in 'General Electronics Chat' started by bteddy, Apr 12, 2010.

1. ### bteddy Thread Starter New Member

Apr 12, 2010
3
0
Hello, great site.

I need to replace a diode that is not available anymore. I have its data sheet and have found close substitutes but nothing exact.

The specs of the original are:

Vrm - 400V
Vf - 1.3V
If - 1A
Ifsm - 25A
Ir - 10ua
trr - 50ns

My problem is the reverse current.
All the close matches I have found are 5ua. From what I have read, the "ideal diode" would have a reverse current of zero, the lower the reverse current the better.

This makes sense, If I am applying 350VAC at .5A continually, the voltage and current do not change when in reverse bias to the diode. Meaning the reverse current is a "Rating" of how much current the diode will let flow through it, not a "limit" of how much reverse current it can handle. So, I should be able to replace it with a diode with a lower reverse current rating.

My Question:
Is my logic in the above statement correct?
If not, why? Please provide details or point me relevant references.

Thank You

2. ### beenthere Retired Moderator

Apr 20, 2004
15,815
283
With those easy specs, a 1N4005 will do quite nicely as a general purpose rectifier.

Unless you have some situation where reverse current is critical, it's not something to worry about.

3. ### bteddy Thread Starter New Member

Apr 12, 2010
3
0
I wish it was that easy. The original is a fast recovery diode, trr = 50ns. This is a critical specification. I can use one faster but not slower (<= 50ns).

4. ### beenthere Retired Moderator

Apr 20, 2004
15,815
283
That does make a difference. Take a look at Digi-Key part #UF1004DICT.

5. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
You want fast find a Schottky diode. They come in decent currents, but they only drop around 0.2V or so. This might be a negative for your application, and they aren't your normal diode in that their basic theory of operation is a bit different. They approach the ideal a bit closer than conventional diodes, but blow faster if you exceed the specs.

6. ### AlexR Well-Known Member

Jan 16, 2008
735
54
The problem with Schottky diodes is that have low Vmax, in the order of 40V-60V and they they have really bad reverses current characteristics Ir=.5mA-1mA is not uncommon.

Lower Ir is generally better. Some applications can tolerate relatively high reverse reverse current but I can't think of a situation where you would go out of your way to have extra reverse current.

If you are running the diode at 350Volt then using a 400Volt diode is sailing a bit close to the wind. Probably that is why the original diode failed in the first place. I would go for a diode rated at 600Volt or greater.

Something like a BYV26C should fit the bill and should not be too difficult to get.

7. ### bteddy Thread Starter New Member

Apr 12, 2010
3
0
The information is great, but I need an answer to the question:

With all else the same, can I replace a diode with a Ir (reverse current) of 10ua with a diode that has an Ir of 5ua?

8. ### AlexR Well-Known Member

Jan 16, 2008
735
54
YES!

I thought that's what para 2 in my post effectively said.

9. ### rjenkins AAC Fanatic!

Nov 6, 2005
1,015
69
The reverse current is temperature dependent and could be anywhere between 0 - 50uA depending on the actual diode and temperature etc.

A new one with a specification of 5uA is simply in the same range, but never so bad.

The UF4004 is a very close match to the original - the Fairchild data sheet gives reverse leakage at 10uA (at 25' C), and 50nS recovery.

The 4005 - 4007 are higher voltage but the reverse recovery time is slightly longer at 75nS.

Datasheet:
http://www.fairchildsemi.com/ds/UF%2FUF4004.pdf