Film cap across bridge rectifier and new "soft" diodes

[001]

Film cap across bridge rectifier (or diodes) was common in XX century

The idea is RF and on-off noise  suprestion, isn`t it? Is it still necessary with new "soft" diodes?

[The Doktor]

What is a "soft" diode? Are you referring to an active rectifier that uses an FET?

[001]

What is a "soft" diode? Are you referring to an active rectifier that uses an FET?

No!

[T3sl4co1l]

It's always necessary given a worse enough circuit.

So, it depends.

Is this for mains rectification?  Barely matters (even with all the inductance in a transformer).

BTW, an R+C is the better solution, sized to dampen the stray inductance.  Often C was used, giving suboptimal damping (most often what happens is, it still rings, but with a smaller step, lower amplitude, and much lower frequency, probably good enough not to be a problem).

Tim

[coppercone2]

I don't really understand this. Is a soft diode slower then a fast diode?

So we have moved to using more lossy diodes to prevent EMI problems? Is this something really old like comparing gallenium rectifiers to silicon ones? Are you preventing RFI rectification by shunting the diodes at high frequencies to prevent even higher frequencies from being spawned/favorably chaining PSD?

Don't you want that stuff turned into high frequencies as much as possible so its easier filtered? Or is it more likely to cause problems if it gets into the circuit in the first place?

Does this have anything to do with transformer shunting with capacitors like I have seen once? Perhaps to bypass its mixing nature?

[schmitt trigger]

I do remember, in the 1970s and 1980s, seeing those caps in parallel to mains rectifier diodes in some (not all) products.
The reason given was to prevent RF issues, whatever that meant. No further details were provided.

I also believe that an RC solution would dampen any spurious oscillations. But my personal opinion is that blindly applying a circuit without actually understanding what one is planning to achieve, or actually measuring its effect, is a recipe for problems.

[T3sl4co1l]

I don't really understand this. Is a soft diode slower then a fast diode?

Slower voltage rise.  Yes, in a sense.  The headline figure that a diode is rated by, is only reverse recovery (time taken from current passing through zero, until reverse peak).  That can still be very fast.

So it is a little bit disingenuous, because recovery isn't when the switching losses occur, it's just after, while current is falling and voltage is rising.

Softness is defined as the ratio of the two periods (recovery and fall time).

It helps EMI at the expense of some power (as is often the case).  It helps, but it can still be defeated, particularly for very short forward-bias conditions (<100ns pulse?) and fast reverse recovery (>1kA/us?).  That pushes closer to the drift step recovery regime, where recovery can happen rather suddenly (softness ~= 0).  Add an inductive loop (which doesn't need to be very much at this dI/dt) and you can get large voltage spikes with quite high frequency content.

What's happening after reverse recovery is, the breakdown (avalanche) voltage is rising.  If the circuit is too "bouncy", it can happen that the voltage swings up and down during this phase, clamping to ever-higher peak voltages until the reactive energy is gone, or the breakdown voltage goes up high enough that it stops clamping and the system rings down on its own.  In short, keep the reactive energy small, and the impedance low, around diodes.

So we have moved to using more lossy diodes to prevent EMI problems? Is this something really old like comparing gallenium rectifiers to silicon ones? Are you preventing RFI rectification by shunting the diodes at high frequencies to prevent even higher frequencies from being spawned/favorably chaining PSD?

Gallifreyan rectifiers?...

Just a process optimization.  t_rr is much lower than it used to be, and softness is really only an issue on high speed diodes.  (Slow diodes are usually quite soft anyway, and with t_rr in the microseconds, who cares?)  So the losses are still lower than for old types, and the EMI can be lower at the same time, too.

Don't you want that stuff turned into high frequencies as much as possible so its easier filtered? Or is it more likely to cause problems if it gets into the circuit in the first place?

Ease of filtering assumes you don't have a bunch of metal loops open to space, radiating all the high frequencies they can (>100MHz?).  In that case, you're best off not generating those frequencies in the first place.

Tim

[David Hess]

I don't really understand this. Is a soft diode slower then a fast diode?

The "soft" recovery characteristic has to do with the di/dt of the "snap" when the diode turns off.  Snap or step-recovery diodes are designed to make this as fast as possible by having the minority carriers which hold charge in the junction all run out at the same time.  Soft recovery rectifiers are designed to make turn off after reverse recovery more gradual.

Standard recovery rectifiers, especially higher voltage ones, are sometimes produced using the same PIN structure that snap-recovery and varactor diodes use so it should not be surprising that they can behave like them.  (1)

So we have moved to using more lossy diodes to prevent EMI problems?

That is sort of the case but the networks used to control EMI also lower efficiency.

Don't you want that stuff turned into high frequencies as much as possible so its easier filtered? Or is it more likely to cause problems if it gets into the circuit in the first place?

If the frequency components are too high, then they more easily couple through the parasitic elements of the filters and circuit.  Placing small capacitors directly across the diodes removes the inductance produced by the loop area between the rectifier and bulk input capacitor.  No amount of filtering will help once a large loop between the rectifiers and filter is spewing magnetic flux everywhere.  (2)

Does this have anything to do with transformer shunting with capacitors like I have seen once? Perhaps to bypass its mixing nature?

It is likely as that is one way to control the EMI.  The three steps I usually take in order are:

1. Use better diodes.  (1)
2. Add capacitors directly across the diodes.  I might do this in addition to using better diodes in a critical application.
3. Add ferrite beads in series with the diodes.  I have seen this done but I am not sure why capacitors were not used instead.  Maybe it is more effective or lower loss?

I do remember, in the 1970s and 1980s, seeing those caps in parallel to mains rectifier diodes in some (not all) products.
The reason given was to prevent RF issues, whatever that meant. No further details were provided.

Audio gear is particularly susceptible to this problem which results in a 120 Hz buzz but it shows up in test equipment also.  (4) When ADCs are involved, it can manifest as excessive flicker. (3)

I also believe that an RC solution would dampen any spurious oscillations. But my personal opinion is that blindly applying a circuit without actually understanding what one is planning to achieve, or actually measuring its effect, is a recipe for problems.

Ideally a series RC snubber would be used but leaving the resistor out if it is not needed is less expensive.  The capacitor values are usually 10s to 100s of picofarads for 1 amp diodes.  For expediency, I typically just find the smallest capacitor value which was sufficient and then double it.

(1) Standard recovery rectifiers seem to be the worst but it varies between part number, manufacturer, and perhaps even lot.  Some 1N4001s do it and some do not.  So a better diode might simply be from a different manufacturer and you could always qualify your standard recovery rectifiers for having a soft recovery.  But if you are not going to do that, then using fast recovery rectifiers which have less stored charge for line frequency rectification is usually the least expensive option.  If you want to use diodes without worrying about it, then add the capacitors.  A rough rule of thumb seems to be 4 to 10 times the diode's junction capacitance.

(2) Can of worms?  The can is open.  The worms are everywhere.

(3) This gets back to precision ADCs where once real flicker noise and other noise sources are accounted for, quantization noise should result in 1 count of flicker maximum and it is possible for 2 counts to indicate a real problem even at 16 bits.  That may not seem like much but some ADCs and circuit designs even going back to the 70s and 80s are potentially that good.  Some early 8-bit DSOs had noise levels this low.

(4) I see a business opportunity here selling special diodes, capacitors, and inductors for rectifiers used in audio.

[coppercone2]

What are examples of fast and slow diodes? When you say process optimization, does that mean a traditional diode like a 1n4007 used for a rectifier has changed in specification ?

I kinda associate what you guys are talking about with diodes using in switch mode power supplies, but for traditional applications (rectification of mains before the switching) or for rectification after a line frequency transformer, what are we looking at now?


if they are changing alot, it sounds like they should have a new part number. It sounds all a bit crazy.

Also this kind of made me think, if you have 50Hz systems and 400Hz aircraft systems, and you are doing a traditional design, should you maybe pay more attention to the diodes?

[001]

I see some curcumedons in the audio net https://www.diyaudio.com/forums/pass-labs/8826-fast-vs-slow-rectifiers.html

What is actual situation now?

[T3sl4co1l]

(4) I see a business opportunity here selling special diodes, capacitors, and inductors for rectifiers used in audio.

Funny you say that...

I see some curcumedons in the audio net https://www.diyaudio.com/forums/pass-labs/8826-fast-vs-slow-rectifiers.html

What is actual situation now?

...You'll see a lot of those customers at dA.  It's a pretty diverse group, ranging from a few intensely technical folks, to the feng-shui virgin-blessed crystals under the speaker cable types.

So, as far as what the diodes actually do, as a reference, I wouldn't read very far into their opinions.

What are examples of fast and slow diodes? When you say process optimization, does that mean a traditional diode like a 1n4007 used for a rectifier has changed in specification ?

No, the specification hasn't changed, that's fixed by JEDEC.

JEDEC specs are notoriously thin, however.  Pretty much any diode can be sold as a 1N4007, assuming there's value in doing so.  Better diodes cost more, so the answer is usually no, unless, say, they fail one spec but still meet another one, like the 1N4007.

That doesn't mean they're worse, but it does mean there's that much more variability between parts, in the parameters that aren't specified.

Diodes in general have greatly improved since those early days (when 1N4007 was first written), with soft recovery and some avalanche capability being typical.  Sloppy diffusion processes gave way to precise epitaxy, ion implant and such.  Impurities and defect levels have continued to improve.

I kinda associate what you guys are talking about with diodes using in switch mode power supplies, but for traditional applications (rectification of mains before the switching) or for rectification after a line frequency transformer, what are we looking at now?

Both, but yes, fast diodes have more options, more improvements.

Also this kind of made me think, if you have 50Hz systems and 400Hz aircraft systems, and you are doing a traditional design, should you maybe pay more attention to the diodes?

You might as well save the Vf (maybe 5-20% improvement) and use a slow type.

Tim

[David Hess]

What are examples of fast and slow diodes? When you say process optimization, does that mean a traditional diode like a 1n4007 used for a rectifier has changed in specification?

It just means that the 1N4007 specifications do not cover its recovery behavior and some standard recovery diodes have snap-off behavior which is a problem in sensitive circuits.  An example of a fast replacement for the 1N4001 series is the 1N4933 series which is the slowest of the common fast recovery rectifiers.

if they are changing alot, it sounds like they should have a new part number. It sounds all a bit crazy.

They are just changing in a way not covered by the specifications.  If the snap-off behavior is a problem but you want to continue to use these diodes, then qualify them by manufacturer and verify that each lot meets your requirements.  If this is a problem, then either use a diode which never displays this behavior or design the circuit to mitigate it.

I kinda associate what you guys are talking about with diodes using in switch mode power supplies, but for traditional applications (rectification of mains before the switching) or for rectification after a line frequency transformer, what are we looking at now?

This is about line frequency rectification.

Also this kind of made me think, if you have 50Hz systems and 400Hz aircraft systems, and you are doing a traditional design, should you maybe pay more attention to the diodes?

Or if it matters, include snubbers across or in series with the diodes to prevent the problem.

[coppercone2]

is there some manufacturer of diodes that is known to keep a more consistent product out, or does additional documentation?

This is kind of a compliance problem if suddenly your product line that was working fine 20 years ago starts doing something funny because a process improved.

textbook chapter with some graphs about rectifier snubbing
https://www.desmith.net/NMdS/Data/Linear%20Audio%20-%20Rectifier%20snubbing%20-%20background%20and%20Best%20Practices.pdf


Also, whats the model of a great/good diode for audio rectifier power supplies, to be used with preamplifiers and a high power variant for power amplifiers?

[001]

Also, whats the model of a great/good diode for audio rectifier power supplies, to be used with preamplifiers and a high power variant for power amplifiers?

IMHO
Modern fast diodes like HER108  produce swithching noise at common 50Hz rectifier too isnt`it?
Is capasitor still eliminate it?

The second question is
40 years ago I shunt EVERY diode with cap
New diodes own  capasitance became low
Is it mean that I must increase shunt cap value? 

[chris_leyson]

Some manufacturers give figures for trr for standard recovery diodes, I think Motorola used to do it. It's not too difficult to measure and Tim, T3sl4co1l posted a design for a simple trr tester. The topic also reminded me of one of Mr Carlson's videos. Interesting part starts at 12min.

[David Hess]

is there some manufacturer of diodes that is known to keep a more consistent product out, or does additional documentation?

Other than various soft recovery rectifiers, I am not aware of any.  If such a standard recovery part existed, it would probably cost more than adding snubbers or using a faster diode.

This is kind of a compliance problem if suddenly your product line that was working fine 20 years ago starts doing something funny because a process improved.

It sure is but that is what happens if you rely on unspecified behavior.

Also, whats the model of a great/good diode for audio rectifier power supplies, to be used with preamplifiers and a high power variant for power amplifiers?

I think fast recovery rectifiers like the 1N4933 series are the best compromise.  Schottky rectifiers should avoid the issue entirely.  Faster "soft" recovery rectifiers might still snap off too quickly but try them and find out.  The last time I tested this, every diode type that I had available except for some standard recovery parts eliminated the problem entirely.

Modern fast diodes like HER108  produce swithching noise at common 50Hz rectifier too isnt`it?
Is capasitor still eliminate it?

I have not noticed the problem when using any fast recovery rectifier.

The second question is
40 years ago I shunt EVERY diode with cap
New diodes own  capasitance became low
Is it mean that I must increase shunt cap value? 

The values are not that critical and diodes with equal current ratings tend to have about the same capacitance.

Some manufacturers give figures for trr for standard recovery diodes, I think Motorola used to do it. It's not too difficult to measure and Tim, T3sl4co1l posted a design for a simple trr tester. The topic also reminded me of one of Mr Carlson's videos. Interesting part starts at 12min.

Unfortunately, reverse recovery time is not the problem; it is the snap-off time or hard versus soft recovery which is.  Fast recovery rectifiers have less stored charge because of lower minority carrier lifetime and this seems to correlate with better results although the fastest parts do tend to recover very hard.

[coppercone2]

I have been thinking about snubbers and stuff, but is the RFI rectification through the diode bridge serious? In that case maybe even if you snub it you still want some HF bypass in parallel with the snubber. I don't know what the pass frequency should be or how to treat the filter impedance properly.

I think I might experiment with doing HF injections through a power amplifier into a diode bridge to see what happens to measure the output of a crude filter with a SA but I need to make some decent linear HV sine wave generator thats well behaved because I don't want to plug a expensive RF amp (100-500MHz @ 50W max) into something connected to the outlet. I need to get some APEX chips to make a low noise low power line simulator). Then when I get some data maybe the LISN can be used for something finally.

[T3sl4co1l]

Rectifier acts as a PIN diode, gating noise in/out of the PSU.

Therefore even if the PSU is clean and stable (making sharp tones at Fsw and harmonics), all of those tones have a 120Hz buzz modulated onto them.

Tim

[David Hess]

I first ran across this problem in one of my simple fixed voltage bench power supplies when using it for audio design work.  The impulse noise from the snap-off of the standard recovery rectifiers proved impossible to filter out and was actually audible in circuits not even using the power supply.  After filtering, I tried ceramic capacitors directly across the diodes and that worked fine.  Later I found that fast recovery rectifiers solved the issue also.

After that, finding small capacitors in parallel or ferrite beads in series with power line rectifiers in test equipment made a lot more sense.