How To Measure Diode Switching Noise?

[CJ]

I want to scope out any differences between these Freds (Fast Recovery Rectifiers) and these plain ol 1N4004's.

Is there anyway to do this wothout $1,000,000,000 worth of test equipment?
I have a prety good scope, pwr supplies, etc.
Thanks!

Note werid packaging on Fred leads:

 

[Svart]

Not weird at all, I've been using them like that for years!

Check out IRF's appnotes on the FREDs, their diagrams are pretty close to the actual measurements. Expect a greatly reduced snap during recovery and less radiated RF.

I'm using the MUR series of FREDs, cheap, reliable and fast/soft.

 

[CJ]

Hey, Thanks!

Here is the APP note.

http://www.irf.com/technical-info/appnotes/an-989.pdf

 

[mikka]

'Scuse my *******.... I've been wondering about the benefits of these.... :roll:

.... can you HEAR the difference.........by the time the juice gets filtered I would have thought the benefits to be virtually nil.

Comments appreciated...

 

[mcs]

[quote author="mikka"].... can you HEAR the difference.........by the time the juice gets filtered I would have thought the benefits to be virtually nil. [/quote]
The problem (or so I've read) is the radiated noise from the diodes - no filtering will remove that. But a mumetal shield and ferrite beads on all the wires probably would...

Best regards,

Mikkel C. Simonsen

 

[mwkeene]

A faster diode will improve switching supply efficiency, and there will also be less noise generated when the diode goes into reverse recovery...that goes for linear supplies too. It also makes me feel better not to see a 100nS 100ma shot of current in a 5 watt current source on the scope.
-Mike

 

[Samuel Groner]

...can you hear the difference... By the time the juice gets filtered I would have thought the benefits to be virtually nil.

You might check what I've posted here (last page):
www.groupdiy.com/index.php?topic=18673

But a mumetal shield and ferrite beads on all the wires probably would...

I was able to improve the EIN of channel 8 by 1.5 dB by removing the ferrite beads of the design mentioned above. A ferrite bead might help reducing HF noise if the noise is picked up at a different place (e.g. by a mic cable), but it will make the circuit more sensitive to noise existing inside the box. That's unfortunately the case with about any magnetic element--including transformers.

Shielding either the diodes or the sensitive circuit is surely effective, though I'd be interested if mumetal is the right thing to use. I guess that conductivity at high frequencies might be important and that there's better stuff out there for that.

Placing snubbers across the diodes would cure things as well.

Is there any way to do this without $1,000,000,000 worth of test equipment?

Perhaps a 1M ohm resistor used as pick-up element in close proximity to the bridge rectifier could shed some interesting light on the topic?

Samuel

 

[mikka]

Fook mare...... thanks guys.....appreciate your input ... :thumb:

 

[jdbakker]

[quote author="Samuel Groner"]

Is there any way to do this without $1,000,000,000 worth of test equipment?

Perhaps a 1M ohm resistor used as pick-up element in close proximity to the bridge rectifier could shed some interesting light on the topic?[/quote]
Have a look at Appendix J of Linear Tech Application note 70. It describes a simple EMI sniffer probe; they have a few examples on its use including sensing diode recovery noise.

JDB.
[EDIT: Yet Another Stoopid Typo]

 

[CJ]

Cool article, Thanks!

 

[Samuel Groner]

Cool article!

It is! :thumb:

Samuel