Valve Pre-amp & phono thoughts for future project

[diceman]

I am looking at starting to consider design aspects of a long-term project now a few things have been completed in the system. My experience currently extends no further than cross-over building, uprating components in existing circuits and building a kit dac & SUT in to a box. Whilst I am reading valve design books I think I would prefer to build a known design of sorts as staryting from scratch as a newbee will probably only end in a lot of trial and error. I would however like it to be a bit modular to allow swapping out of a PSU or Board for a new alternative if required.

I would like this to be of good quality and probably of simple design using very good quality components. I don't massive gain for line or current for long lengths of cable so I am considering a single triode (or pentode wired as triode) used as an active line stage.

I am essentially thinking of at present building a valve pre with phono stage incorporating MC SUTs (I may add a DAC later as well). I plan on building this in to 2 boxes; box 1 containing seperate power supplies and regulation for each section and the second box containing all the signal handling components.

A couple of questions for the experienced if I may
1) Use of copper for chassis - what is the benefits of this? I believe the copper is non ferrous hence non magnetic and reduces eddy currents from transformer floating around the case and presumably either impacting on signal carrying components or other transformers? If I use a 2 box design with PSU seperate is there any benefit? Does copper still provide good RFI? Is it worth considering running any heater AC wires in copper microbore pipe internally or is this a really dumb idea and showing my inexperience?
2) Power supply selection - heater circuits, do these need to be DC for line level? For phono stage it seems many designs use DC so I assume it is required or much better practice for such low level signal? I have started playing with the PSU designer software and probably erring toward a valve regulated supply or 2 (1 for line and 1 for phono).
3) Any suggestions on circuit deisgn or links to known good sounding circuits or kits?
4) Pre-amp output transformers or capacitor coupled?
5) Any suggestions on transformer manufacturer, I am currently looking at Lundahl and Sowter purely as thgat is all I have read about in DIY circles.
6) I am concerned about hum as anyone who has had a go at building something normally has massive hum issues with tube amps, is this mainly due to AC heater wires not being run correctly typiclaly? Any tips?

I am hoping to construct modules on boards but using tag strip hardwire connections so that a module can be removed a la PCB but with hardwired connections for ease of swapping components and sound quality.

All suggestions and discussion welcome and any input is greatly appreciated. I may have something working in 2 years or so!

[i_should_coco]

Do you have Morgan Jones's books? I would highly recommend them. He does a very nice phono design in "Valve Amplifiers". The second book "Building Valve Amplifiers" deals with construction techniques.

[diceman]

I have one of them which I think is just "valve amplifiers"

Hard going and I am on my 3rd read now. TBH I think I need to actualy design a basic line stage as first phase to run through the calcs fully and implement them just as a test case.

I will buy the second one.

Many thanks for the input & suggestion of the phono stage.

[toprepairman]

Agree with the M Jones books, generally very good, but there are other sources so look around.
The two box route is definitly the best way to go.
DC for the filaments is practically compulsory for phono stages, avoid 6 volts, far better to go 12 or even 24 Volts (filaments in series). The supply does not have to be regulated, just well smoothed.
Capacitor coupled from cathode follower is just fine, no need to spend lots of money on expensive transformers which always compromise performance.
As for the power Tx, there are plenty of suppliers out there, it depends on your budget.
Regards
Henry

[i_should_coco]

It's worth persevering. The line-stage idea is good and wold give some practical experience, it should not be too much effort or cost.

The second book deals with construction techniques, shielding, wiring, etc.

Other phono stage designs are the ones that used to be on Steve bench's site (archived here: http://diyaudioprojects.com/mirror/m...com/sbench101/). I built the "RIAA 2" for a friend and it turned out nice.

[diceman]

The second book has been ordered. I also picked up "designing power supplies for valve amplifiers" by Merlin Blencowe.

Thanks for the suggestions. I will have a look around for simple line stage schematics to get an idea of what to do and what not to do and build up on a bit of wood for testing :-)

[i_should_coco]

Ironically, if you're that way inclined, the Kondo Audio Note M7 is about one of the simplest line stage you could imagine!

Anyway, keep us posted.

[diceman]

I think as far as pre-amps are concerned a nice simple circuit with top quality components is probably the best I have heard within limits of operation of the system.

never heard an M7 I think unless it was on the end of speakers that are the anti-thesis of how I like them. And yes I appreciate that different people like different things but for my ears panel speakers and plenty of power is required for them so I am more a push-pull man than single ended although my 845 SE does some things that are rather amazing. Toms SJS was rather special which opened my eyes to very simple circuits which I believe is of a similar ilk to a Kondo/audionote high-end pre. Simon Shilton of SJS has some basic line stage designs published to ascertain what you like in different valves that seemed a good place to start.

[i_should_coco]

Yeah, the SJS designs are similar in philosophy to the M7/Kondo stuff. I suggested it as it's basically an incredibly simple design and the schematic can easily be found. As you note, power amps are a different kettle of fish and have to be matched to the speakers. Even I'm not mad enough to suggest a 5W SET to drive panels,

[diceman]

And I am not mad enough to suggest panels to you :-)

To be fair if I had a partner, neighbours and lounge accomodating enough then speedysteves horns or simiilar would be on the agenda first! Adam's old lowthers were pretty good as well and are probably (apart from AG with bass horns at munich) the only horns I have heard that have a flat enough freq response and timing to consider me swaying from panels.

I am, however, unfortunately in a position where box speakers will have to be reinstated in the lounge for a while to avoid damage by prying baby fingers/arms/heads.

[Valvebloke]

I just wrote a typical 'considered' (i.e. long-winded, rambling) response only to have IE delete it when I tried to post it. So here's a terse grumpy summary :

COPPER CHASSIS: Looks pretty for a while but you have to keep polishing it or find a decent protective lacquer. You can't buy it so you'd have to make it - how good are your metalwork skills ?

MICROBORE PIPE FOR HEATER WIRING: OTT for low-current preamp valves. Should work fine for DC heaters but you might struggle to twist it if you want to use AC, and what about the insulation ? Probably a real pain to attach it neatly to the valve base pins.

TRANSFORMERS: Cons include cost, performance limitations (unless you spend quite a bit of money), tendency to leak signal out and noise in, requirement for quite a bit of space. There are some pros though: ground isolation lets you eliminate earth-loop hum, they're also a nice way of interfacing to balanced interconnects which are what the professionals use, they can also give you a very low output impedance so the pre can drive practically anything. I ran a World Designs Pre3 for a while http://www.wduk.worldomain.net/acatalog/Modular%20Preamp%20Pt2.pdf which had output transformers and was a load of fun.

In the mid 90's Maplin published a design for a valve pre called the Newton (articles at the bottom of this page http://livinginthepast-audioweb.co.u...ium/index.html). The design wasn't flawless but the articles were very detailed and gave a nice insight into how a preamp design might be put together.

Whatever you choose to have a go at I'd encourage you to stick with it. There is nothing like building your own kit if you want to learn about hifi technology. And it can be very satisfying and fun too !

VB

[diceman]

Thanks for the comments Mr valvebloke.
Idid read the 90s articles on the newton amps at the time and have considered the world audio pre amps if they were not so expensive nowadays.

I did mean to say about running the regular heater wires within the microbore not using microbore as the conductor :-) As I say it is an odd idea and maybe just plain wrong?

[sjs]

Simon Shilton of SJS has some basic line stage designs published to ascertain what you like in different valves that seemed a good place to start.

By the looks of it you already found the article I wrote for Common Ground all those years ago. The idea was a simple test bed to promote experiments and get a view on which valves you may like, and what component choices do to the sound. Having said that the results can sound excellent, Tom's pre was basically the Figure 2 approach.

If I were writing it today, I would probably add some further options, including: choke load on the anode or CCS on the anode.

At present I still favour the pot before the valve approach, and I am mainly using a similar single tube linestage, and the Russian 6H30 and 6N6P are excellent for this use.

As for a phono stage, I think there is another article about this, and I have built a number of 417A designs similar to the Steve Bench circuit Pete indicated. For a first build it may be safer to stick with 6072A as the input and second tube as they are much less fussy, and dont tend to oscillate wildly at MHz frequencies . If that goes well, change the design to use high transconductance valves such as 417A/5842, EC8010 (a la Morgan Jones), or triode strapped D3a (there are some nice designs out ont'web), just make sure the PSU can cope with a step up from 3mA to 15mA per tube !

As for your questions:
1) it just sounds best!
2) DC filament supplies for the phono are probably a must, with the line stage you "may" be able to get away with ac heating, or you may not.....
For HT, personally the only regulated ht I every liked to listen to music with was shunt regulated designs, I don't use series regs, even using tubes..... not nice. The old AI L2 and P2 shunt PSU sounded good using a 5687, a 6080 version of it sounded HUGE, and Gary Dews built a shunt reg with a 300B..... now that was loverly. As a sand based alternative, the guys over at diyAudio go nuts for the Salas Shunt regulated ht supply, so it may be worth including in your list of options to test. Generally I just like choke input supplies with valve rectifiers and polyprop caps.........
3) you seem to have options.....
4) caps - affordable and probably a known quantity... reasonably cost effective to go cap rolling, unless you get an itch for Dueland.....
personally I tried lots of different transformer outputs, but never really could live with them. Fabulous walk in midrange and ease, but it all went a bit milky and soft for me, but ymmv
5) Llundahl or AE Europe (Netherlands)
6) Hum.... only if I dont know the tune
normally just down to poor psu design, or iron work too close to the circuit/tubes, but your 2 boxes should sort that.

Tag board, try the ptfe board from ampmaker, I think this would be a very good solution http://www.ampmaker.com/store/PTFE-board-257x67mm.html

and finally, best of luck!
Simon

[Valvebloke]

I did mean to say about running the regular heater wires within the microbore not using microbore as the conductor :-) As I say it is an odd idea and maybe just plain wrong?

Aaah. Interesting. Now you're going to drag me right back to basic EM shielding theory (gulp !). It's late, so forgive me if I don't try and reconstruct the explanation from first principles. I might just end up talking bollocks. But the result, I believe, is that for relatively low-frequency high-current circuits such as valve heater wiring the main problem is the magnetic field that such circuits create. Copper shielding is an effective way of eliminating stray electric fields but it doesn't work so well against magnetism. So I fear that you won't get much improvement in the noise from your heater wiring by running the cables in copper tube. And in any case once they get near the valve, which is where much of the pickup can occur, they have to come out of the tube. So it ceases to help you where you need it the most. The best approach is just to keep the wiring as far away from the valve as you can, as close to the chassis as you can and twisted together as well as you can. Or, these days, to use DC.

VB

[pure sound]

Do consider the gain structure. A typical phono applies 40 dB of gain (x100) so a typical MM giving 5mV will put out 500mV. Your line stage gain then needs to be kept sensibly low particularly if you'll also be using CD as a source (2V rms) and typically sensitive power amps ie 500mV-1.5V input sensitivity depending on the design. Many active line stages have way too much gain for use with CD which then requires excessive attenuation somewhere to make it user friendly. How that is applied can also affect the end result profoundly.

[diceman]

Some excellent advice guys & guy Many thanks to all, even current & Ex manufacturer's chipping in - gotta luv the wam.

Interesting that the russian 6h30n tube is sounding quite nice, I have read a lot about this and has been lebelled the "super tube". As a single valve it looks quite interesting as has plenty of current compared to most.

The russian Reflektor & Voskhod tubes sound very nice in my book and are normally quite cheap (apart from the 6h30n it seems).

Gain is indeed a very important consideration and easy to overlook. I suspect what makes sense is to use very low/no gain as line stage and try and match phono stage output to typical CD of 2V rms.

This will all be thrown out the window if I use it with my Leak Stereo 20 though as opposed to any of my regular amps!

thanks again chaps.

A quick clarification on copper. Is there any benefit in using it for both PSU & Signal boxes or just one?

My metalworking skills are iffy / so-so. I do have access to a guillotene and folding machine for small jobs though although I am concerned that the copper would be badly marked from the process.

[sjs]

Interesting that the russian 6h30n tube is sounding quite nice, I have read a lot about this and has been lebelled the "super tube". As a single valve it looks quite interesting as has plenty of current compared to most.

The russian Reflektor & Voskhod tubes sound very nice in my book and are normally quite cheap (apart from the 6h30n it seems).

The 6H30 has become rare, costly and hyped in the past 10 years, but the 6N6P is almost the same, sounds almost the same, and hasn't yet hit the big time. Either buy matched pairs from a good dealer, or a bunch of NOS ones from Eastern Europe and sort them yourself.

Other interesting alternatives could be to go for an octal tube, Thomas Meyer likes the 6AH4, and Gary Dews uses the low mu part of the 6FD7/6DE7 family.

Gain structure is important, I tend to aim for an overall gain of around 6 to 8 dB, as is output impedance which should be determined by what power amp load you wish to drive. Many valve amps may have a 100K to 220K input impedance, so a valve plate impedance of 2 - 3K may be acceptable, however transistor amps can go as low as 10K which sets a real challenge to a cap coupled common cathode stage, partly because of the low plate impedance required, and partly the size of the cap! This is where an OTX is useful, as it avoids using a cathode follower.

I have also been playing with battery grid bias on my line stage recently, which enables the electrolytic bypass cap to disappear, which is a very nice game.

[pure sound]

I recently found a 'new to me' indirectly heated triode which has low gain & which ought to make quite a fun, if current hungry, & grunty cap coupled line stage. I've just received a few but haven't got around to building anything yet. It's a 7233. Might make a nice 1-2W power amp too.

[hullevad]

I think that the 6N6P is a wonderfull valve! I use it myself in a heavyly modified DIYHiFiSupply TRAM line amp. Not in production anymore! But You can find "The recipe" on Troels Gravesen website: http://www.troelsgravesen.dk/Diy_Lou...r_Projects.htm The best linestage for the money, Silk transformers highly recommended! Use only "super caps" they make quite a difference ("the best caps are no caps").

[anubisgrau]

i've had a 6N6P preamp based around modernized "euridice" WOT schematics and it's a killer. very easy to build, you just need some good iron (chokes and output transformers) and voila. this may be a good reading point to start:

http://basaudio.net/pubs/jfl_euridice_2005.pdf

and from there on, there's a plenty of variations around on DIY fora, just google "euridice preamp". the design has a huge following among DIY-ers.

a step forward is this - a differential parafeed line stage:

http://www.raleighaudio.com/MI%20articles.htm

it's a well elaborated DIY design with very little chance to make something wrong as you will find a plenty of different components tested and chosen for sound merits. it's based around lundahl iron but obviously you can put anything there...

for me the output transformer is the must in a line preamp, otherwise you will end (in most of the cases) with either a difficult, high output impedance that will discriminate almost anything but a 100k tube power amps or a cathode follower that will just not sound as good.