Carlos' snubberized Gainclone Power supply

[peranders]

http://www.diyaudio.com/forums/attac...amp=1103139431

Above we see Carlosfm's power supply for Gainclones.

We see also that roibm have implemented these snubbers and they can be seen here.

Can anybody give a trustworthy explanation how they work and what they should eliminate?

Looking at the schematic above Carlos have a 100 nF in parallell of 1R+ 100nF. Does this make any sense? What would the 1 ohms resistor do? Why not 10 ohms or 2.2 ohms?

[tool49]

Hi peranders, the only place where I have seen a logical explanation to this is the following:

Also, please note the two symmetrical RC networks. They serve to get rid of the residual capacitor inductance, which should improve high frequency performance. In my experience, it always does, though to what extent remains open to debate, trial and error. No matter how good the capacitors may be, they always have some inductance left over; the better they are, the lower the value, and vice versa. Therefore, this is always good to have, even if its greatest effects will show up with the worst of capacitors. A side benefit is that the amp will tend to be even more stable with complex loads, although this is primarily something the amp design should deal with.

Incidentally, do not be daunted by the power rating of the resistors shown, depicted as 1 Ohm 17W here; the actual required rating of the resistor will depend on many factors, but with a 17W rating, you're not likely to go wrong with any amplifier up to some 150W/8 Ohms and corresponding increases of power with lower loads.

http://www.tnt-audio.com/clinica/ssps2_e.html

Which is probably where CarlosFM also took the information.

It does not say much, but it is worth a try for the price of 4 parts.
Hope this helps you!
Sébastien

[tiroth]

There is a thread somewhere here that shows spice simulations of diode/transformer ringing. I believe RC filter damps the ringing and the C only filter changes the frequency. The conclusion was to use both.

I think to make an ideal filter you probably need to optimize for a specific diode/transformer combination. But even non-optimized snubbers should generally help.

[ThorstenL]

Konnichiwa,

Originally posted by peranders
these snubbers

Can anybody give a trustworthy explanation how they work and what they should eliminate?

Yes, someone can.

Originally posted by peranders
Looking at the schematic above Carlos have a 100 nF in parallell of 1R+ 100nF. Does this make any sense?

Yes, to a degree. I'd probably like to see larger capaictor values.

Originally posted by peranders
What would the 1 ohms resistor do?

One should think it was obvious. Just draw out the full RLC equivalence ciruitry for the whole lot oc capacitors and analyse it....

Originally posted by peranders
Why not 10 ohms or 2.2 ohms?

1 Ohm is probably still a little high but a convenient value to get off the shelf.

Sayonara

[jackinnj]

Originally posted by peranders
http://www.diyaudio.com/forums/attac...amp=1103139431

Above we see Carlosfm's power supply for Gainclones.

We see also that roibm have implemented these snubbers and they can be seen here.

Can anybody give a trustworthy explanation how they work and what they should eliminate?

Looking at the schematic above Carlos have a 100 nF in parallell of 1R+ 100nF. Does this make any sense? What would the 1 ohms resistor do? Why not 10 ohms or 2.2 ohms?

you have a resonant circuit formed by the transformer secondary, the interwinding capacitance and the capacitance of the diodes. you can ex ante determine the ringing frequency from all the above, providing you have a calculator which takes square roots!

The RC network snubs the ringing.

Carlos' values are not optimal and I don't know what methodology he used to determine them. Too large a capacitor, too low a resistance wastes energy and can actually inject RF into the system. Purportedly, the effect of a badly designed snubber is audible to even those not gifted with golden ears -- (if you have ever made your own pulse-width modulated motor controller you know what I am talking about.)

The equations can be found here, the article is a follow-up to one Jim Hagerman wrote for Audio Amateur:
http://www.hagtech.com/pdf/snubber.pdf
or the more complete version, with more greek symbols and plenty of graphs:
http://www.cornell-dubilier.com/tech/design.pdf

[percy]

Cornell-Dubilier has a simplified version also.. www.cde.com/catalogs/5.029-5.031.pdf

You might want to read this one too - www.calex.com/pdf/3power_impedance.pdf

[janneman]

Percy,

The last link talks about the (in)stabilities in active regulated supplies. That is a totally different mechanism than in the unregulated, raw rectified case, and you can not compare this in any way as far as values are concerned.

Jan Didden

[Greg Erskine]

Originally posted by peranders
Looking at the schematic above Carlos have a 100 nF in parallell of 1R+ 100nF

I have recently tried 100uF + 1R in parallel with 10,000uF and it works as advertised by Carlos on my minimialistic IGC. Thanks Carlos for sharing this information. I used a 1 watt 1R and using 17-0-17 VDC it only gets slightly warm. I wonder if 1 watt rating is even necessary?

Simple mod highly recommended.

As usual with audio, YMMV.

[Upupa Epops]

Real heat dissipation of this resistor is practicaly insignificat in this connection. Slight warm what you fell is caused only by internal temperature inside machine.

[percy]

Originally posted by janneman
Percy,

The last link talks about the (in)stabilities in active regulated supplies. That is a totally different mechanism than in the unregulated, raw rectified case, and you can not compare this in any way as far as values are concerned.

Jan Didden

Thats a good observation, however I would think that the same concept (reducing the impedance of the power supply for better decoupling) applies equally to both regulated and unregulated supplies.
If not it should still be helpful to many who use a regulated supply.