Need to find replacement diodes

Status
Not open for further replies.

It's a 20V shottky barrier rectifier good for 3.0 Amps.. I doubt you'll find something suitable at RS. You should use a shottky with the same rating.
If you can find any shottky's at RS, you could parallel them to get the current rating. I.e. 3, 1amp in parallel. If all you can find is small signal types like 200mA, I wouldn't parallel that many.

Why does online take too long to arrive? if you have $$$, you can get it in a day? If you can't wait a day, you'll have to go beg borrow or steal from someone / something else... if you can any old switchmode power supplies laying around, you're bound to find multi-amp shottky diodes in there.. pull a couple off the board...
 
Well if it takes a day I guess its not that bad. I already found the part on Jameco...the only thing is that Jameco doesn't have the transformer I'm looking for in the circuit. Will have to look around on digi-key.....
 
Paralleling diodes is a bad idea, as has been discussed here many times. Mismatches will cause one diode to hog most of the current. As it warms up, its forward voltage will drop, causing it to hog more current. This is known as thermal runaway. When this diode dies, the next one suffers the same fate, until they are all dead.
 
Ron H said:
This is known as thermal runaway. When this diode dies, the next one suffers the same fate, until they are all dead.
Click to expand...

But isn't diode usually dies a short circuit, sort of sacrifices oneself for the rest of the group. :roll:
 

yeah generally true...small series resistors might help.. funny thing is, some app notes on switchmode converters recommend paralleling one with the body diode of the switching device. This is for recovery time reasons I know, but the same exact issue exists and design get away with it. The issue seems much more real for transistors than it does diodes and I don't know any good reason why..
 
I think they generally recommend a diode that is meant to handle all the current (lower Vf). The body diode is then out of the picture.
 
Maybe so. I gotta confess I don't have a lot of first hand experience with diode failure modes.
Depending on the application, I can see a few other parts dying (electrolytic caps, voltage regulators, etc) as a result unless the fuse is really fast.
 

The effect is not so bad with Schottkys though. Std silicon diodes have a much sharper I/V curve over their specified range and never share well. Schottkys are 0.3v to ~1.0v over their specified range. It's like they already have sharing resistors flattening out their I/V curve but there's still the negative temp coefficient. For all the current to go through the warmest diode it might need to be at 0.9v, the others would start conducting at 0.3v so it's not possible to have all the current in one diode.

Figuring out just how much sharing there is, and how much is required to avoid stressing a diode, is still an exercise.
 

They still significantly share though (I've measured it in an SMPS) before... At least for the one particular case. My understanding was they recommend the external diode for a better recovery time characteristic and not because they needed a larger current handling diode. Also, at the very low forward voltages it doesnt take much parasitic resistance to make for adequate "ballasting" resistors in the paralleling case.

One other thought about thermal runaway, a careful thermal design can avoid the scenario alltogether. A good thermal design can place an upper bound on how far that temperature coefficient can cause a current increase. If you can remove enough heat before the next switching/conducting cycle, you will avoid the accumulating / positive feedback effect.

I agree that it is not a good idea to do if you can do the job with a single device. The original poster conveyed a sense of ugency (he indicated he couldnt wait a day?!) and to get something up and going, he can probably get a way with paralleling signal level shotttkys if he found them. I admit the recovery time in that solution would be terrible due to lead inductance
if he wasnt careful.
 
its ok guys, i may be able to order the part off jameco and digikey...the paralleling stuff sounds kinda like more work. thanks for the help. now could someone tell me where i maybe able to get the coiltronics transformer? the ctx4-20 in the schematic is not in digikey.
 
What has been said about paralleling diodes without matching them is true.
I just so happens that I once matched a bunch of 1N5820 diodes for a circuit that needed 15A. So I carefully matched five of them passing 3A thru each one and measuring the forward voltage drop. When I had 5 with the same voltage drop + or - a few mV , I connected them in parallel.
The rectifier worked without problem for about 3 years when I took it out of service.
 
Status
Not open for further replies.

Similar threads

O
  • Question
Help finding a replacement relay for sealer
Replies
12
Views
1K
Diver300
D
Stress testing unknown diodes
Replies
13
Views
1K
MrAl
M
H
  • Solved
Newbie: Help finding replacement diode and transistor
2
Replies
20
Views
4K
shokjok
S
P
  • Question
11ES4 replacement
Replies
2
Views
1K
Paulok
P
A Cool Find – '50s/'60s Weller Soldering Gun
Replies
16
Views
2K
Doug
D
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…