Valve pres with solid state powers , what works
- Thread starter DavidS
- Start date
Arkless Electronics
Trade: Amp design and repairs.
Currently using a couple of interesting preamps with some rather obscure valves in the line stage...
6EW7 and 8608
Both can be bought for very little and sound very impressive.
Untitled by The Biglebowski, on Flickr
Untitled by The Biglebowski, on Flickr
I'm thinking that at least the top one is a DIY one that started life as something else... the chassis clearly shows "E55L" next to the socket... this is utterly different from the 6EW7, the first is a pentode (and very similar to the 8608! but different pin out etc) and the latter a dual triode.
They are very interesting valves and vastly better than the usual suspects.
Unfortunately hard to obtain and pretty expensive.. which is the huge problem with many/all of the best valves.
The likes of ECC81/2/3 etc are toys compared to these!
The last gasp of valve technology was the frame grid valve (and the Nuvistor) and many types were developed for industrial, professional and military applications but also some for TV. They have much better characteristics than conventional valves. The only one that springs to mind as a commonly available valve used in audio is the ECC88/6DJ8/6922 and it's parameters make it my favourite triode. It is what I used in my DIY amp project mentioned above and is what makes my circuit an order of magnitude better in many regards compared to the Croft design in terms of measured performance.
There are many types around which are much better still than the ECC88 but unfortunately not common.
And therein lies the rub... There are valves around which as I said are an order of magnitude better than the usual suspects but maybe £30 - £100 (some a hell of a lot more!), often available only in small quantities, you usually can't just get them from Watford Valves/Hot Rox etc etc and a suspicion I have is that after all these years many of the best in terms of microphonics, noise, matching etc will have already been selected and sold off, leaving a bit of a lottery as to what you get.... not to bad if it were modern ECC83's @ £12 each but if they are rare and £90 - £160 each...
Russia made many superb frame grid valves for military use and at one time they were dirt cheap, like say £3 each, but as mainly DIY'ers (and a very few manufacturers) got wind of them their value went up and even these are now often say £25 - £50 these days. There's still a few bargains to be had though.
Darren
So gentle when he tries to understand.
C'mon Biggy. Spill the beans
Arkless Electronics
Trade: Amp design and repairs.
With posts such as the above and a PM a got I think I need to spell out that I am not suggesting that anyone should refrain from using valve pres with solid state power amps!!! Just that the worst bad matches should be avoided and that people should be aware of the likely increase in distortion with quite a few combinations... a distortion which some are looking for!
"I used to run a Croft 25 in to a Quad 909. At first a complete mismatch, lots of noise and no useable volume" That wouldn't happen and so there must have been a fault.
Yes it applies to "valve DACs" and to ANY AND ALL valve sources. More so in fact than with most pre amps as they are in effect always at full volume.
You will not have "obvious problems" under almost any circumstances. ie sound breaking up, full volume too quiet, hiss, funny noises etc
The distortion is not of the type that people may think of as distortion. It is very different to the sound you hear when a small amp or say a portable radio etc is turned up too loud and distorts or a cartridge mistracks etc.
It is the very distortion that many WANT as it helps give "valve warmth" etc.
It subtly changes instrumental timbre, how "woody" or "shrill" or "spot lit" or "dark" etc the sound of say a clarinet comes across.
Ultimately it will often limit the SQ, to me anyway, as it can help turn massed choirs and other things with loads going on "a bit mushy", "lacking separation between instruments" or even a bit harsh etc but can make a "girl and guitar" sound "better" to many people.
TheDecameron
Unicorns fart glitter.
Used an ARC Ref3 into Chapter Audio hypex based stereo power amps for a few years, worked very well indeed. Started with one amp as a stereo power amp then used the double balanced outputs of the ARC to drive one stereo amp per side in a vertical bi amp arrangement. Wouldn’t hesitate to use a similar combo again.
G
Guest432
Guest
On loan from Simon Shilton, they were headphone amps he modified into preamps using those valves.
L&N republican
pfm Member
Anyone tried EAR preamp(864) with naim amp such as 160 or maybe 140.My 890 amp generating a lot of heat so it will be nice to have nice cold amp for summer which can drive quad esl's
I have tried an endless range of solid state amps with my Croft 25R pre and a Conrad Johnson PV12A pre.
The Croft needed a bit of tweaking by Glenn Croft to make it compatible a wider range of power amps.
The PV12A needed no adjustment but I got it serviced and had a remote volume control installed.
Had very good results with many combos, The Croft excels of course in many ways but the Conrad Johnson blew me away with its soundstage and fullness. It really gave me the full valve smoothness and presence with no loss of power when combined with a Parasound A21, a Hegel power amp, and a Plinius power amp. In fact the only combo with the CJ PV12A that disappointed was ironically with a similar vintage CJ valve power amp.
So yes - needs a bit of trying but you can get great results
Incidentally, I might be selling the CJ PV12A now that I have moved to an solid state system....Will post in classifieds at some stage
The Croft needed a bit of tweaking by Glenn Croft to make it compatible a wider range of power amps.
The PV12A needed no adjustment but I got it serviced and had a remote volume control installed.
Had very good results with many combos, The Croft excels of course in many ways but the Conrad Johnson blew me away with its soundstage and fullness. It really gave me the full valve smoothness and presence with no loss of power when combined with a Parasound A21, a Hegel power amp, and a Plinius power amp. In fact the only combo with the CJ PV12A that disappointed was ironically with a similar vintage CJ valve power amp.
So yes - needs a bit of trying but you can get great results
Incidentally, I might be selling the CJ PV12A now that I have moved to an solid state system....Will post in classifieds at some stage
Tony L
Administrator
Personally I can’t see that not working. I used a Croft 25R happily with a Quad 303 (22k /0.5V) and it sounded fine, I’d expect you to get rather more usable volume range than I did. For me all the volume I’d ever want was within the first 5-10 minutes, but the Quad is far more sensitive and my speakers efficient. I’d expect the Rega to work well.
uncl_nigel
pfm Member
I used to get nice sounds from a Canary 601 valve pre into an Avondale stereo power amp.
I got to hear the combo again when I visited the friend I gave it to a couple of weeks before the lockdown and it still sounds good.
At another price point, I have also had some very nice sounds from an Aesthetix Calypso Eclipse (valve pre) into a Moon 760 solid state power amp (my Aesthetix Atlas power amp was in the States being upgraded at the time).
I got to hear the combo again when I visited the friend I gave it to a couple of weeks before the lockdown and it still sounds good.
At another price point, I have also had some very nice sounds from an Aesthetix Calypso Eclipse (valve pre) into a Moon 760 solid state power amp (my Aesthetix Atlas power amp was in the States being upgraded at the time).
Chris81
pfm Member
Arkless Electronics
Trade: Amp design and repairs.
I'd already covered it generically when I said that in the vast majority of cases where a valve pre had around that output impedance and a SS (or any!) power amp around 20K ish input impedance that there would be some rise in distortion in comparison to the optimum but that most wouldn't notice it, it is exactly the type of distortion that many like anyway, and that it is level dependent and so at "normal listening level" it is nothing to worry about.
Arkless Electronics
Trade: Amp design and repairs.
I've been working a bit on the valve line stage design I put in the DIY section and have ended up with a completely new version which I believe answers all these sort of issues
Or should I say answers them even better as the DIY section design already has only 500 Ohm output impedance.
Still just the one dual triode valve but how does around 100 Ohm output impedance sound? And THD around 0.005% ?
It occurred to me that we don't actually often, if at all, use a pre amp in the conditions it is generally measured, ie when giving >1V output and driving the power amp to full output. At normal listening levels we are only putting say 100mV ish into the power amp on average and this allowed me to make some changes to the very concept on which it worked to optimise it to likely actual conditions in use It will drive anything and sounds great
Or should I say answers them even better as the DIY section design already has only 500 Ohm output impedance.Still just the one dual triode valve but how does around 100 Ohm output impedance sound? And THD around 0.005% ?
It occurred to me that we don't actually often, if at all, use a pre amp in the conditions it is generally measured, ie when giving >1V output and driving the power amp to full output. At normal listening levels we are only putting say 100mV ish into the power amp on average and this allowed me to make some changes to the very concept on which it worked to optimise it to likely actual conditions in use
It will drive anything and sounds great
Chris81
pfm Member
I have read the post, but how do I know which technical parameter fits the best or when do I have to consider problems?
How do I know if I can use a good range of the volume control and not only a quarter? Currently I'm using a Audiolab MDAC as a pre with 2.25V output but I think i can't use the full power of the Rega monos when playing low volume recorded tracks loud.
At the very least, it has nothing to do with volts.
A pre' will have an input and an output impedance, a power will have an input impedance (plus an output impedance, though not of interest here).
VERY crude rule of thumb, the input impedance of anything should be a minimum of around ten times the output impedance of what is feeding into it.
Does anything sound OK? If it does, in all likelihood you are fine.
Croft amp's are odd/atypical for "valve" amp's - the pre's (Micro in various guises) have output impedances of less than 300R. The Croft power amp's are also not valve amp's, but amp's with a valve.
A pre' will have an input and an output impedance, a power will have an input impedance (plus an output impedance, though not of interest here).
VERY crude rule of thumb, the input impedance of anything should be a minimum of around ten times the output impedance of what is feeding into it.
Does anything sound OK? If it does, in all likelihood you are fine.
Croft amp's are odd/atypical for "valve" amp's - the pre's (Micro in various guises) have output impedances of less than 300R. The Croft power amp's are also not valve amp's, but amp's with a valve.
Arkless Electronics
Trade: Amp design and repairs.
I've already gone into considerable detail on all this so I'm not repeating it for the sake of one person I'm afraid.
You can generally use all the volume control. If you need to turn it up to 10 then do so. In many cases the worst technically is half volume (this does NOT mean "5" on a scale of 10 as vol controls are logarithmic! prob more like "7" ish). This especially applies to passives. At full volume they effectively disappear electrically and it's as though the DAC was plugged straight into the power amp with no vol control in the way... = "perfect".
Valve pres will not be overloaded by anything. They have high voltage power supplies and can deliver say 20V+ (100V is not unknown!) into the power amp if required! THD probably 3% by then but nothing will clip or overload.
If you actually need more gain (headbanging contest!?) then use the Croft pre if it has gain.
To reiterate, any "old wives tales" of "for best results you should find yourself using between around 3 and 5 on the vol control" are usually just that .... old wives tales! Yes there are exceptions that prove the rule but not that common.
Tony L
Administrator
I’d argue they were very common. In many cases with a typical 2V+ input from a loudly mastered digital source the power amp stage (even of integrated amps) goes into clipping long before the volume control gets anywhere remotely close to the end stop. In some cases before it even gets to 10 o’clock on the knob’s travel! As such not “an old wive’s tale” at all, but a useful way to explain input gain etc.
Arkless Electronics
Trade: Amp design and repairs.
It has everything to do with Volts!
The rule of thumb is a good one but it is indeed a VERY crude one.
Croft pres are NOT atypical! They are very typical as they use pretty much textbook standard circuitry. one of the first things I said is that some sites say the Croft 25 pres are <300R but they are actually around 3900R!!
As I also spelt out, most cathode followers stages, which of course have no gain, will be <1K and often <3-400R. ECC83 actually one of the worst valves for a cathode follower but still only 625R Zout.