Anyone interested in a DIY valve line stage?

[Arkless Electronics]

Here it is

[IMG][/IMG]

Obviously this shows just the meat of the thing and does not show the power supply detail or selector switch, phono sockets etc but it shows how nice and simple it is... but elegantly I think Now't new or "no one's done this before", just, I hope, sound engineering of the K.I.S.S type.

The valve is a dual triode so there is just the one valve.

[Arkless Electronics]

Get away......you're lovin' all the attention

Oh luvvies, dahlings

[Arkless Electronics]

Specifications:

Gain 10dB (x3)
Output Impedance 520R
THD <0.01% @ 2V RMS out into typical 47K load. <0.1% into 10K load
Frequency response <-1dB 5Hz-100KHz into 47K

Not half bad for a super simple one triode per channel pre amp I think you will agree

I should point out that this is not intended to be "ooh listen to that euphony and warm roses in me contact lenses valveyness" More "straight wire with gain but with a hint of valve, sounds more like SS but without any SS grain/grit/greyness". If folks want more of the former I will show a few things to drag the poor thing more in that direction later

[tubehunter]

..

[Arkless Electronics]

Plenty of feedback

not much drive

how about a cathode follower




and a constant current

eg:

Yours has roughly the same feedback, no constant current source, an input impedance of only 10K and actually has far less drive potential in practice than my suggestion It would work well though yes but I've no reason to believe it would be any better than my simpler offering and suspect that in practice it would be slightly less good.

1/ The main principle here is K.I.S.S. Yours is twice as complicated for a start and needs 2 valves.
2/ I'm doing this on some sort of strange whim which I haven't really analysed myself .. maybe to put something back for the site in appreciation of Marco's assistance in erm certain things It is NOT intended to be a "lets all uber analyse the circuitry and then suggest our own alternatives and design a horse by committee that turns out more of a camel that's no longer suitable for the DIY beginner"

3/ As someone who does this for a living I'm not going to design something unique, a new topology, patentable and then put it in the public domain! It is good sound non ground breaking engineering and several high end amps have used a similar topology in the past

[Arkless Electronics]

Here it is with power supply and both channels.

[IMG][/IMG]


Nice and simple May be subject detail changes... but that's about it. Note that there is only one physical valve as it is a dual triode. U1 and U2 are actually left and right halves of the same valve. Same with vol control R6 and R12 if a dual gang control is used. They can of course be separate left and right vol controls or even a switched attenuator for even better results.

[Firebottle]

That's about as good as it is going to get for a KISS design line stage

You could omit C1 and C3 if you are paranoid about having electrolytics in circuit, and expect a nads less gain.

Good one Jez

[Arkless Electronics]

That's about as good as it is going to get for a KISS design line stage

You could omit C1 and C3 if you are paranoid about having electrolytics in circuit, and expect a nads less gain.

Good one Jez

Thanks Alan I think it will sound really nice too. The original did and this has a few detail improvements.

I spent a bit of thought on exactly that re the cathode caps but decided to keep the output impedance low and drive as good as possible. Yes it would work fine and is an option though.
A valve diode in line from the output from the SS rectifier would obviate the need for the two Zener diodes... a valve rectifier instead of the SS would do this just as well also. Both add to cost and complexity though...

[Arkless Electronics]

Oh and it's ideal for those wishing to experiment with "boutique" components as so few are needed! Seven per channel are all that's "in the signal path"...

[Arkless Electronics]

Some feedback from all the people who said "I'm interested " would be useful around this point in order to get to grips with how people would want a practical implementation to be, what help is needed to "make it so" and any other questions people have really