SMPS for tube amp : is this anything worth a try ?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
If you want it for a guitar power amp then it is probably OK. This one, like several others I have seem provides 6.3V for the heaters at up to several amps and this is the output that is regulated. The HT output is not regulated so it will vary with the load on the heaters and its own load.

Cheers

Ian
 
A few things to watch out for:

1) If you don't load an SMPS enough, it can enter into a "hiccup" mode where it switches on and off at a relatively low frequency that will certainly introduce unwanted noise. So you'll need to test it and see if it loads it enough to keep it out of hiccup mode. If not, you can add a load resistor.

2) Not all SMPS are equal and buying one on AliExpress from an unknown manufacturer does not instill confidence. If the parts aren't that great or the layout is less than perfect, you could easily get noise that could necessitate additional filtering circuitry. A capacitance multiplier has a good chance of cleaning up most noise but with such a high voltage / wattage, it is not trivial to do. Fortunately tube amps usually already have quite a bit RC filtering so you might not have a problem at all or you might get away with just a choke perhaps.

3) I am weary about using SMPS with varying loads like power amps. I'm not familiar with the Ampex 350 but if there are any speaker or line outputs that can draw significant variation in current, you might get inexplicable noise and/or distortion. SMPS are designed for largely constant load.

4) The voltage output may not be regulated. Whatever the voltage specified, it is very possibly, if not likely, that the voltage will depend significantly on the load. Specifically, if the load is less, the voltage will be higher. Under full load, it will be lower.

So I suspect you will need to simply experiment to see what works. The first thing I would do hook it up to a simulated load and test it by itself. Rig up something to look at the frequency spectrum of the output. Meaning use a capacitor to feed the AC signal into a line or mic input and view the spectrum in your DAW or interface software or get some software that can show you a spectrum. It should be completely void of any noise. A good SMPS is actually superior to linear supplies in this respect in that they can be made to be completely free of mains noise which is difficult if not impractical with a linear supply.
 
squarewave said:
A few things to watch out for:

1) If you don't load an SMPS enough, it can enter into a "hiccup" mode where it switches on and off at a relatively low frequency that will certainly introduce unwanted noise. So you'll need to test it and see if it loads it enough to keep it out of hiccup mode. If not, you can add a load resistor.

2) Not all SMPS are equal and buying one on AliExpress from an unknown manufacturer does not instill confidence. If the parts aren't that great or the layout is less than perfect, you could easily get noise that could necessitate additional filtering circuitry. A capacitance multiplier has a good chance of cleaning up most noise but with such a high voltage / wattage, it is not trivial to do. Fortunately tube amps usually already have quite a bit RC filtering so you might not have a problem at all or you might get away with just a choke perhaps.

3) I am weary about using SMPS with varying loads like power amps. I'm not familiar with the Ampex 350 but if there are any speaker or line outputs that can draw significant variation in current, you might get inexplicable noise and/or distortion. SMPS are designed for largely constant load.

4) The voltage output may not be regulated. Whatever the voltage specified, it is very possibly, if not likely, that the voltage will depend significantly on the load. Specifically, if the load is less, the voltage will be higher. Under full load, it will be lower.

So I suspect you will need to simply experiment to see what works. The first thing I would do hook it up to a simulated load and test it by itself. Rig up something to look at the frequency spectrum of the output. Meaning use a capacitor to feed the AC signal into a line or mic input and view the spectrum in your DAW or interface software or get some software that can show you a spectrum. It should be completely void of any noise. A good SMPS is actually superior to linear supplies in this respect in that they can be made to be completely free of mains noise which is difficult if not impractical with a linear supply.

Good info !!
I also found this , but the price is steep. Maybe I build a linear one for half the price . Dunno
https://www.ebay.com/itm/High-power-LLC-switching-power-supply-KT88-300B-tube-amp-power-supply-L14-10-/143152265539
 
andre tchmil said:
Good info !!
I also found this , but the price is steep. Maybe I build a linear one for half the price . Dunno
The price isn't steep it's ridiculous. That's just some audiophoolery. It should be $17 not $170.
 
don't have any experience with that PSU but I wanted to say that recently the new Manley products are made with SMPS so maybe SMPS for tube is more in reach than ever before?
 
outoftune said:
don't have any experience with that PSU but I wanted to say that recently the new Manley products are made with SMPS so maybe SMPS for tube is more in reach than ever before?

I am sure it is possible with care. I use SMPS exclusively for heaters these days and recently I have replaced the linear 48V phantom supply with a switcher plus some post filtering recommended by Abbey. HT is the last bastion to crack. How hard it is depends a lot on how much current you require. I have found a KED SMPS from Meanwell which outputs a steady 300V at up to 500mA. I have been planning for some time to use it with a linear post regulator to provide a few hundred mA of 250V HT. It is part of the ongoing Mark 3 tube mixer design.

Cheers

Ian
 
Back
Top