Best sounding SS rectifiers?

Any concensus on the best sounding SS rectifier?

I've been using Vishay HexFRED with excellent result. I read that Cree SiC Schottky is excellent too, but I haven't tried it. I also have some Diotec soft recovery rectifiers and they are good sounding too.

Has anyone done comparisons on these top notch rectifiers? Which one is the best?
 
I think this thread is worth resurrecting.
It probably also makes a difference in what circuit they are being used.
It might be difficult to generalise about this.

I just tried 11DQ06 in my DAC power supply and they are a big improvement over 1n4007. No complaints at all (so far)

Also tried Fairchild stealth FFP08S60STU in my DAC Valve output stage power supply and I don't like it!
The bass has no attack. and the midrange seems to be too prominent. (gave them 2 days to run in :( )

I intend to try out a few different fast and ultra fast diodes in the valve (high voltage) supply and let people know what I think..
 
Last edited:
^ some claim they can and some claim they can't. Define "well designed" power supply. Does that mean it has RC filtering, snubbers, decoupling caps, LC filters, it's all on the same PCB or not, ground planes, 40X the capacitance it should need per current, etc. etc. etc?
 
I didnt think there would be any sonic difference, but low and behold.... there was!
I was a bit surprised about this.

FWIW, the 11DQ06 only costs about 15 cents. I wouldnt call it fancy, rare, or exotic.

To elaborate, when I installed 11DQ06 in my DAC I noticed a considerable increase in clarity across all frequencies. I also noticed smoother (grain free) HF and mids.

The power supply in the DAC is fairly standard. Transformer> rectifiers > Caps > voltage regulators > rest of circuit.

I know a lot of guys here are into designing specific power supplies for specific uses, and in comparison, I guess this power supply is fairly standard.
Probably a PS using L C R filters will be less susceptable to mains noise and diode switching noise.

For me the 11DQ06 is a big audible upgrade to a "standard" power supply, and costs very little.

The Fairchild Stealth diode made a very big negative impact on the sound.
Music lost its "sparkle" and "magic" - eveything became too even sounding. No part of the music "jumped out" and the bass disappeared. yes the highs were smooth, but everything else sounded "wrong".

Really hoping some other guys can share their positive or negative experience with the "sound" of diodes....
 
With projects using say regulated +-15 t0 20V
I prefer the SF11 or SF12. They have a maximum reverse recovery time of 35nS which is better than the UF400X series.
They also have a lower forward voltage drop than the 1N400x series at low to medium current.
It is also worthwhile trying a 470pZF across the outers of the secondary windings when using a 30VA toroidal. The 470 PF is close to the value posted in a chart in DIYAudio some time back.
 

Attachments

  • cap-on-the-ac-side-of-a-rectifier.jpg
    cap-on-the-ac-side-of-a-rectifier.jpg
    541.1 KB · Views: 2,535
^ Who cares about reverse recovery time with 100/120Hz rectified mains frequency? Also, nobody cares about forward drop voltage differences with +-15 to 20V (30 to 40V) rails on an audio amp.

(just my opinion, apparently someone does care).

Let's get down to science on this topic. "IF" there is a sonic improvement it will be had with soft recovery diodes and among them, the differences aren't much to consider, but that's only if the rest of the PSU and amp can't reject the switching noise which should easily be mitigated with an RC snubber subcircuit if a decoupling cap alone isn't enough.

If your supply is regulated, there should not be any sonic difference from a diode change unless there is noise the supply isn't designed to handle... hint: inductors

That's not to suggest a diode change can't possible effect sound, only that it's not a magic bullet but rather depending on what the weaknesses are in the design so it would be fair to say that other/different changes instead could make as much difference if not more, or of course we could idealize and say change the diodes, change the other PSU fault too, keep going until the pursuit of perfection leads to satisfaction... or frustration, or just enough time, work and expense that you declare the project done ;).

Overall I would have to think that if time is being spent on which diodes to use, that's time better spent on something else instead. Whatever diodes you pick, if you change them and the result is the exact same sound, a fair % of people will still declare it was an improvement because ego just causes people to think like this. I don't mean that it's impossible for there to be an improvement, rather that if you look at random changes people make, overall you'll find them declaring more improvements than not even though they were following time tested designs that were deemed good enough that nothing needed improved.
 
Last edited:
^ Who cares about reverse recovery time with 100/120Hz rectified mains frequency? Also, nobody cares about forward drop voltage differences with +-15 to 20V (30 to 40V) rails on an audio amp.
(just my opinion, apparently someone does care).

Well, actually, its more about the little HF voltage spike that the diode makes when it stops conducting.

Let's get down to science on this topic. "IF" there is a sonic improvement it will be had with soft recovery diodes and among them, the differences aren't much to consider, but that's only if the rest of the PSU and amp can't reject the switching noise which should easily be mitigated with an RC snubber subcircuit if a decoupling cap alone isn't enough.

Schottky made a big audible improvement equal to any improvement I have heard before from cap swaps etc.

If your supply is regulated, there should not be any sonic difference from a diode change unless there is noise the supply isn't designed to handle... hint: inductors

for those modifying a pre existing PCB, fitting inductors may not be an option, and this is where the schottky comes into its own. IMO.

That's not to suggest a diode change can't possible effect sound, only that it's not a magic bullet but rather depending on what the weaknesses are in the design so it would be fair to say that other/different changes instead could make as much difference if not more, or of course we could idealize and say change the diodes, change the other PSU fault too, keep going until the pursuit of perfection leads to satisfaction... or frustration, or just enough time, work and expense that you declare the project done ;).

I never said it was a magic bullet. But every little thing helps.
The reason I tried different diodes was becuse I had already fitted my DAC with my favourite good sounding quality parts, and it was/is sounding pretty darn good, and was wondering if I could get it sounding any better. So one of the things I had not yet tried was different diodes. I was pleasantly surprised how much positive difference they make.

!, It would probably be equally good for your own knowledge and interest to try some schottky diodes and listen for yourself rather than arguing theory. You are entirely correct with all that you have written, but even going on the theory, you should know that the little turn off spike generated by a diode will create unwanted noise in the rest of the circuit.
Sure a snubber and inductor *may* attenuate it, but wouldn't it be better to eliminate the little spike with a schottky rather than treating the spike with snubbers etc?

I for one am very interested in the subjective opinion of others on this topic.
 
Last edited:
Any concensus on the best sounding SS rectifier?

I've been using Vishay HexFRED with excellent result. I read that Cree SiC Schottky is excellent too, but I haven't tried it. I also have some Diotec soft recovery rectifiers and they are good sounding too.

Has anyone done comparisons on these top notch rectifiers? Which one is the best?

I usually listens to music when my stereo is on - IF the rectifier is overloaded and it makes a lot of mechanical noise/sound, it is time to consider a bigger rectifier! :D
 
Well, actually, its more about the little HF voltage spike that the diode makes when it stops conducting.

Exactly... but either way you have that, if anything a higher frequency pulse requires better filtering.

for those modifying a pre existing PCB, fitting inductors may not be an option, and this is where the schottky comes into its own. IMO.

but if we're talking ideals, wouldn't one pick the best pre existing PCB and/or make their own?

I never said it was a magic bullet. But every little thing helps.

This can be taken three ways, one way a builder looks at what helps most, then the next best improvement and so on until they run out of options, time, space, or construction budget. Another way it can be taken is a builder can chase someone else's ideal component combination only to find that they either don't hear a difference or that what someone else subjectively feels sounds better, sounds worse.

The third way is psychological, if time, money and thought are put into a change and it at least sounds as good and someone else claims it sounds better, ego can cause the conclusion that a change made really does sound better whether the rest of an amp is identical (has same strengths and weaknesses) or not.

The reason I tried different diodes was becuse I had already fitted my DAC with my favourite good sounding quality parts, and it was/is sounding pretty darn good, and was wondering if I could get it sounding any better. So one of the things I had not yet tried was different diodes. I was pleasantly surprised how much positive difference they make.

There needs to be a distinction made about why it sounds better. With regulation after the rectification stage, diode changes should not make a difference unless the regulation stage isn't working optimally to remove noise... and yet, we come back to the subjective impression of what sounds good. Certain types of noise in the power supply seem to be considered having a positive effect on the sound by some people.

!, It would probably be equally good for your own knowledge and interest to try some schottky diodes and listen for yourself rather than arguing theory. You are entirely correct with all that you have written, but even going on the theory, you should know that the little turn off spike generated by a diode will create unwanted noise in the rest of the circuit.
Sure a snubber and inductor *may* attenuate it, but wouldn't it be better to eliminate the little spike with a schottky rather than treating the spike with snubbers etc?

Schottkys are fast for sure, but that doesn't eliminate the spike it just moves it to a higher frequency meaning the regulation stage is less apt to filter it any moreso if there isn't sufficient inductance.
 
Exactly... but either way you have that, if anything a higher frequency pulse requires better filtering.
if generalizing, wouldn't the higher frequency use cheaper and smaller parts to filter it out, and move the noise well above the audible frequencies?


but if we're talking ideals, wouldn't one pick the best pre existing PCB and/or make their own?
That would be the best way to go, but not if you are a person who likes to modify pre-existing equipment.



This can be taken three ways, one way a builder looks at what helps most, then the next best improvement and so on until they run out of options, time, space, or construction budget. Another way it can be taken is a builder can chase someone else's ideal component combination only to find that they either don't hear a difference or that what someone else subjectively feels sounds better, sounds worse.

The third way is psychological, if time, money and thought are put into a change and it at least sounds as good and someone else claims it sounds better, ego can cause the conclusion that a change made really does sound better whether the rest of an amp is identical (has same strengths and weaknesses) or not.

I guess the psychological aspects your statement really comes down to the individual. I swap parts over quite a lot. Sometimes they sound better, sometimes just different, and sometimes they sound worse to me in my system. I agree that the sound of different parts wont satisfy all people all of the time. This is why experimentation is good. I still think it would be good for you to try schottky diodes next time you have the cover off something and in the mood for tinkering.....

There needs to be a distinction made about why it sounds better. With regulation after the rectification stage, diode changes should not make a difference unless the regulation stage isn't working optimally to remove noise... and yet, we come back to the subjective impression of what sounds good. Certain types of noise in the power supply seem to be considered having a positive effect on the sound by some people.

There is lots of stuff that people think "shouldn't" make a difference but does make a difference. Whether "most" people prefer the different sound is another story.
 
if generalizing, wouldn't the higher frequency use cheaper and smaller parts to filter it out, and move the noise well above the audible frequencies?
At first thought yes, one would only need smaller value capacitors to filter out higher frequency signals, but often a quantitative change in some parameter (in this case going to a higher frequency) makes a QUALITATIVE change in effect. Higher frequencies are more easily radiated (as in approaching RF or "Radio Frequency") from the diode/transformer wire directly into low-level, relatively high impedance audio inputs where they can cause nonlinearities or outright rectification in a transistor's base-emitter junction and become easily audible. Also, capacitive and inductive coupling more easily conduct higher frequencies (below perhaps 1 MHz this is probably the dominant cause).

And yes, a "no-charge-storage" Shottky or other high-speed rectifier can solve the problem, but I like the idea of using capacitor or RC snubbing across every 1N400x rectifier diode as opposed to using a single "high-end" diode that costs more than the other components combined.

I guess the psychological aspects your statement really comes down to the individual. I swap parts over quite a lot. Sometimes they sound better, sometimes just different, and sometimes they sound worse to me in my system. I agree that the sound of different parts wont satisfy all people all of the time. This is why experimentation is good. I still think it would be good for you to try schottky diodes next time you have the cover off something and in the mood for tinkering.....
I think this is only half of science (or engineering, or whatever people are trying to do to make things better) - detecting a difference when changing components is one thing (indeed very important by itself), but determining WHY there's a difference is another and contributes much more to overall knowledge.