0.02dB difference. 😏
Probably you heard the difference of two capacitor structure (construction, material, quality difference etc.) ... or did you want to hear the difference. 😛
Bela sometime less is more
BTW I hear very well the difference between electrolytic or dry capacitor😉
BTW I hear very well the difference between electrolytic or dry capacitor😉
Perhaps (as coupling capacitor).
Some manufacturer make paralleled capacitors within one envelope, some use different thickness -or even material- isolation for different values.
For same value, but different voltage almost all of them use different material-thickness-structure variations.
In my experience, in the same series the higher voltage one's "tone" different, sounds better -for me-.
That's why I mostly use 600-1200V coupling capacitors even for "low" voltage equipments too.
Some manufacturer make paralleled capacitors within one envelope, some use different thickness -or even material- isolation for different values.
For same value, but different voltage almost all of them use different material-thickness-structure variations.
In my experience, in the same series the higher voltage one's "tone" different, sounds better -for me-.
That's why I mostly use 600-1200V coupling capacitors even for "low" voltage equipments too.
i have my second 26 stage up and running in filament bias. This one has 10 ohm filament resistors instead of 7.5 ohm. Otherwise very similar. With exactly 1.5V on the filaments, the cathode voltage turns out to be 9.4V not 10V as expected from a 10 ohm resistor. The resistor is a nice NOS Ayrton Perry type at 30W. On my other 26 stage the same resistor is bypassed to reduce it to 7.5 ohms, but the wire wound 10W bypass isn't quite so good. You'll note that the Coleman reg is V4 - I've collected a few V4s over the years I've been using them. For a new build you would use V9 and take 3V off the supply to the reg, so a 12+12V 50VA transformer would work. Use dropper resistors in the supply to get the right voltage. For this 10 ohm version the dropper resistor was 0.68, 5W.
It sounds very slightly cleaner than the other one, so all in all it's been a constructive build. Good luck finding NOS Ayrton Perry resistors at 20W to 30W. Yes, they do make an audible difference.
It sounds very slightly cleaner than the other one, so all in all it's been a constructive build. Good luck finding NOS Ayrton Perry resistors at 20W to 30W. Yes, they do make an audible difference.
Last edited:
I've been happy with this 26 stage into my PSE 4P1L outputs, so I'm now getting down to some detailed work on parts. I'd be very interested to know what parts you guys are using on your DHT stages. I've also posted a more accurate schematic.
Coupling caps: I used to use 0.1uF FT-3 200V caps but the larger 600V ones sound better.
Anode resistors: I've been using 2W NOS black Panclimatic resistors, or alternatively 2 of 1W in parallel or series. I've been collecting these for a while. I want to try some 2W Audio Note Tantalums from HiFi Collective. Not cheap but they get a lot of love. I prefer Panclimatic to Vishay bulk foil.
Cathode resistors in filament bias: You need 20W to 30W here. I mount them above the top plate because they get hot. I've tried a few types, and here are my findings:
Welwyn Zenith NOS resistors - my preferred ones so far. Dark green with 2 leads on each end, one long and one short. I don't have any more stock, alas
Other NOS large vitreous enamel resistors - e.g. Ohmite. Can also be good
Stackable NOS Dale dark grey resistors: Easy to mount. Good but NOS vitreous enamel are better
Russian military light grey ceramic resistors: Different sound to the Dale ones. OK but the vitreous enamel ones are better
Welwyn W series: Maximum is 12W or 14W so need to be used in parallel. OK but the vitreous enamel are better.
Aluminium clad power resistors: I didn't manage to get those sounding good
Thick film: didn't like those at all.
Pics below. Do please share your own choices.
Coupling caps: I used to use 0.1uF FT-3 200V caps but the larger 600V ones sound better.
Anode resistors: I've been using 2W NOS black Panclimatic resistors, or alternatively 2 of 1W in parallel or series. I've been collecting these for a while. I want to try some 2W Audio Note Tantalums from HiFi Collective. Not cheap but they get a lot of love. I prefer Panclimatic to Vishay bulk foil.
Cathode resistors in filament bias: You need 20W to 30W here. I mount them above the top plate because they get hot. I've tried a few types, and here are my findings:
Welwyn Zenith NOS resistors - my preferred ones so far. Dark green with 2 leads on each end, one long and one short. I don't have any more stock, alas
Other NOS large vitreous enamel resistors - e.g. Ohmite. Can also be good
Stackable NOS Dale dark grey resistors: Easy to mount. Good but NOS vitreous enamel are better
Russian military light grey ceramic resistors: Different sound to the Dale ones. OK but the vitreous enamel ones are better
Welwyn W series: Maximum is 12W or 14W so need to be used in parallel. OK but the vitreous enamel are better.
Aluminium clad power resistors: I didn't manage to get those sounding good
Thick film: didn't like those at all.
Pics below. Do please share your own choices.
Last edited:
@andyjevans - which thick film resistors did you try?
I'm currently working on a DHT line stage using the EML 20A-mesh tubes, but it will likely be a couple months before I have anything far enough along to listen to. I am planning to use filament bias with Rod Coleman regulators, but I'll be using a 250H Lundahl choke plate load with a parafeed Lundahl 8:1 output transformer since I want balanced outputs and only modest gain.
I'm currently working on a DHT line stage using the EML 20A-mesh tubes, but it will likely be a couple months before I have anything far enough along to listen to. I am planning to use filament bias with Rod Coleman regulators, but I'll be using a 250H Lundahl choke plate load with a parafeed Lundahl 8:1 output transformer since I want balanced outputs and only modest gain.
Yes - transformers are a good way to do a line stage. I used to use LL1660s for that.
But my 26 stage is in fact the driver stage for my 2-stage PSE 4P1L amp - it's not a line stage. I prefer the sound of good anode resistors where possible, and for my use they work fine with a 0.1uF coupling cap.
I can't remember which thick films - they were something like TO220 shape.
But my 26 stage is in fact the driver stage for my 2-stage PSE 4P1L amp - it's not a line stage. I prefer the sound of good anode resistors where possible, and for my use they work fine with a 0.1uF coupling cap.
I can't remember which thick films - they were something like TO220 shape.
For anode load resistor enameled Welwyn or Vishay are good but I like more Mills MRA12.
Cathode resistors filament bias depends the current & voltage, for 26 nothing sounds so good like Mundorf M-Resist Supreme but are 20W if you use only one , 2nd Mills MRA-12 you can parallel 3 resistor of 30R to get 10R 36W, also Vishay aluminium housed non-inductive styles (type N) with Ayrton-Perry winding for lowest reactive components when you need more watts & only one https://www.mouser.es/ProductDetail/Vishay-Dale/NH25010R00FE01?qs=Cpc6SZ3%2B4uMAUbC%2BqlusCw== .
Cathode resistors filament bias depends the current & voltage, for 26 nothing sounds so good like Mundorf M-Resist Supreme but are 20W if you use only one , 2nd Mills MRA-12 you can parallel 3 resistor of 30R to get 10R 36W, also Vishay aluminium housed non-inductive styles (type N) with Ayrton-Perry winding for lowest reactive components when you need more watts & only one https://www.mouser.es/ProductDetail/Vishay-Dale/NH25010R00FE01?qs=Cpc6SZ3%2B4uMAUbC%2BqlusCw== .
I am planning to use three Mundorf M-Resist Supreme resistors in parallel. Each will dissipate about 2.6W and they will be mounted to a fairly large heatsink, so they should be well within their power handling capability. We'll see how they work. I can switch to something else if necessary, but I like how easy it is to heatsink these.
For the EML-20A, I only need 5.6V across the cathode resistors for the operating point I have chosen (10mA), so power dissipation is lower than with the 26.
For the EML-20A, I only need 5.6V across the cathode resistors for the operating point I have chosen (10mA), so power dissipation is lower than with the 26.