Given the same voltage, same current rating: are shottkey diodes quieter compared to hexfreds? Why do many people use schottkey instead of high speed switching diodes?
Thanks in advance.
Thanks in advance.
There is no one "best" or "quieter" - it all depends on the circuit. The hexfreds are meant for fast SMPS applications while Schottky's tend to haver lower Vf and are typically faster than standard but care should be used when using them in faster switching frequency SMPS applications. Rectifiers for a 50 or 60 HZ application should be standard recovery parts, possibly with parallel caps for applications that are either very noise sensitive of do not have much secondary filtering.
Hal
Hal
Thanks. Ok, so in a tube filament supply in pi configuration with 1) 12v/2amp and 2) 20v/6A, will it matter? And what size snubber cap would I use? Can I rely on an internet calculator?
For a filament supply, they are not needed at all - and even if they were, there should be plenty parasitic capacitance and inductance in the wiring itself. However, 1 to 10nF wouldn't hurt anything even if it doesn't help. There is actually a lot of math involved in calculating the approximate values for both rectifier snubbers as well as Zobel networks - and a lot of variables go into the calculations as well. It is something that can be found online but even doing the calculations only gives you a ballpark place to start since the parasitic effects play such a prominent roll - and in the end a good oscilloscope is needed to fine tune the response to the application.
Hal
Hal
To clarify: adding a 4.7nF cap in parallel with the rectifier diodes in a 50 or 60 HZ rectifier will almost certainly not hurt anything and may well have a slight technical benefit. But do not add them across anything in a high frequency rectifier if not strictly following datasheet calculations and recommendations.
Hal
Hal
Well, you are typically either rectifying after a transformer secondary (50/60HZ) or as part of a SMPS or related circuit. Use for the former, go by the datasheet or reference design for the latter. For that matter, in my personal view no one that is not a complete masochist should try designing a SMPS from scratch - just go out and buy a COTS supply and let someone else have the headaches 😵
Personally, I just almost always use a linear supply and I do use snubber caps on rectifiers - for which I usually use bog standard 1N400x discrete parts as they are cheaper and more flexible than integrated bridges. Followed by very low noise regulators and filters since I work in the world of micro and nano volts. And again, I am not saying not to use them if nothing else because of the 1 time in 100 they might make a legitimate difference - cheap and easy insurance if nothing else. The main reason I would not use the high speed rectifiers is that the resulting harmonics are much higher frequency than standard parts and those higher frequencies are what are filtered the lease by any following circuitry. They, like any other RFI/EMI can not only affect circuitry directly but can easily alias down into impactful bands. For whatever reason people in audio seem to want to play around and replace existing op-amps and transistors with alternate parts. This is mostly harmless except that, for some silly reason, they think that "faster" means better. Really doesn't. What it does mean is much, much more likely to oscillate. And high frequency artifacts couple into the circuit from the power supply (and elsewhere) just love to trigger these oscillations in these modified (or just badly designed to begin with) circuits.
Hal
Personally, I just almost always use a linear supply and I do use snubber caps on rectifiers - for which I usually use bog standard 1N400x discrete parts as they are cheaper and more flexible than integrated bridges. Followed by very low noise regulators and filters since I work in the world of micro and nano volts. And again, I am not saying not to use them if nothing else because of the 1 time in 100 they might make a legitimate difference - cheap and easy insurance if nothing else. The main reason I would not use the high speed rectifiers is that the resulting harmonics are much higher frequency than standard parts and those higher frequencies are what are filtered the lease by any following circuitry. They, like any other RFI/EMI can not only affect circuitry directly but can easily alias down into impactful bands. For whatever reason people in audio seem to want to play around and replace existing op-amps and transistors with alternate parts. This is mostly harmless except that, for some silly reason, they think that "faster" means better. Really doesn't. What it does mean is much, much more likely to oscillate. And high frequency artifacts couple into the circuit from the power supply (and elsewhere) just love to trigger these oscillations in these modified (or just badly designed to begin with) circuits.
Hal
In a tube amplifier power supply that includes a power transformer, the major source of power supply HF noise is the oscillatory ringing of the L-C resonant circuit comprising the secondary's leakage inductance, and rectifier + transformer capacitance. The resonant circuit is the "bell" and the diode reverse recovery is the "hammer" which strikes the bell / stimulates oscillatory ringing. To eliminate ringing, increase the damping ratio "zeta" above 1.0. My preferred method for doing this is to add a C+RC snubber circuit across the transformer secondary, whose resistance has been tuned using the Quasimodo test jig and an oscilloscope. And if you're really, really lucky, somebody else has already done this on the exact same transformer as yours, AND they have posted their optimized snubber component values here on the Forum thread named Quasimodo Results (only). Then you can copy their snubber design, since their transformer is identical to yours.
Last edited: