• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

#26 pre amp

Here's mine - it has a separate power supply. Glow tubes on the left for the input tubes. I use a modular chassis with several interchangeable top plates, with rackmount parts as you can see. All my chassis are 2U and usual rack width. Lundahl transformers underneath.

Andy

Nice looking amp Andy!! And that's clever ;)

Someday I would like to try out 10Y, heard a lot of good things about it...:)

Currently both this amp and the preamp are on cathode bias, battery bias is something I want next...
 
Currently both this amp and the preamp are on cathode bias, battery bias is something I want next...>>>

If you try filament bias, Rod's boards will be substantially the same as for cathode bias but you should check with him on that. The difference will be that you will need a higher voltage transformer, caps and maybe a larger heatsink. Worth thinking about!

Andy
 
Hi Andy,

neglecting copper losses in the transformer this is calculated as follows:

rp of the tube divided by the square of the turns ratio.
That is 7000 Ohms / SQR(4.5) This is roughly 350 Ohms

Best regards

Thomas

350 ohms would agree reasoably well with hi-impedance headphones. Does a pair of 26's have enough grunt to run 600-ohm cans like the old style AKG k240's?

Thx!
 
Hi Pieter,

26 tube with 8:1 transformer; that means 2 VRMS in for about 2 VRMS out.
2 VRMS with a load of 300 ohms is about 13 mW.
Should be plenty enough to drive a HD600.

You would loose some gain since the output impedance would be somewhat above 100 Ohm. Loaded with 300 Ohm you will loose some signal. Also that would mean to load the tube with less than 3 times it's rp.

But the intention was to use it for both linestage with some gain and headphone amp.

What could be done is to provide some switching at the output between 9:1 for headphone and 4.5:1 for linestage

Best regards

Thomas
 
Hi!

Does a pair of 26's have enough grunt to run 600-ohm cans like the old style AKG k240's?

Not optimal. That would load the tube too much. The point of transformer coupling (among others) is to give the tube a high impedance load to achieve a load line close to horizontal. I like to load a tube with 5 times it's rp or more.

In your case that would be less than 2 times the rp

better would be a 9:1 output trans like Lundahl LL1689

Thomas
 
What could be done is to provide some switching at the output between 9:1 for headphone and 4.5:1 for linestage

Best regards

Thomas

Very doable; actually I made some line output transformers for this application with very good results. Even 1 VRMS into the 300 ohms HD600 will be very very loud. Nice tricks can also be done to make the transformer as a volume control as well (very good for load and noise!).
 
If the preamp is merely for line level it's being a separate unit is already questionable, adding another tube to get more drive would be a nice example of over-engineering IMHO.
Way more sense would make to drive an integrated line level input amplifier with an inductive volume control. These controls are very well able to drive long interconnects; something like this www.tribute-audio.nl/products/signal/inductive volume control.htm
 
Hi Pieter,

but how do you ensure that the inductive volume control is adequately driven? I found inductive volukme controls to sound best when driven from a low impedance source. Preferably a transformer coupled one.

This is the reason why a stand alone linestage makes sense IMHO to provide a high impedance to all possible sources and to act as a buffer to drive an inductive volume control nicely

Thomas
 
Hi Thomas,

In the meantime a bit OT..., but when I developed my autoformer volume control, I tested it with source impedances up to 3k, which is pretty "safe" I think. That asks for quite some inductance. My autoformer has an inductance of about 150H, that is an AC impedance at 20 Hz of almost 20k, going up with frequency of course. At the same time, using advanced core material, coil DC resistance could be kept very low at some 30 Hz. Bandwidth depends on volume position but with a 2k source impedance it nevers falls under 200 kHz. Already some 8 years ago I visited Dave Slagle and we did some comparisons, showing that the quality was very much on par, using pretty different materials. In the meantime there have been some improvements, and Dave's will not be the same now I guess, but these autoformers are excellent options for volume control.
 
Hi Pieter,

I have not heard yours yet, but I heard other TVCs which claimed to work well with highish output impedances. I never liked them.

As you read in this thread some DIYers don't really care a lot for low Zout even linestages with 7k Zout are not uncommon. I've seen Phonostages with even more.

To be universally usable a linestage should have very high input impedance and low Zout. That's what I'm doing with my commercial linestages since I don't know what customers use them with.

Different with my own stuff where I have everything under control.

Thomas
 
Hi Thomas,
There is a difference between transformer volume controls (TVC's) and autoformer volume controls.
Feedback from listeners always confirm the autoformer type to be more transparent and dynamic.
I do care very much for correct impedances between elements of a chain. When output impedances are (too) high I always find the sound stage to suffer; with correct impedances the sound stage is broad and deep. Therefore I wound for a high inductance with the autoformer; interface with normal source equipment seems OK, but yes not everything can be controlled. When something is wrong it points to something weird with the source.
 
Very doable; actually I made some line output transformers for this application with very good results. Even 1 VRMS into the 300 ohms HD600 will be very very loud. Nice tricks can also be done to make the transformer as a volume control as well (very good for load and noise!).



Are we talking here of using the same 9:1 Lundahl OT, with headphone and line out hooked up differently at the seconday?

IIRC, the 1669 is wired something like 9 : 9 / 1 : 1 : 1 : 1, tho I don't really know what that means.