Tube rolling??

[Zoidburg]

Hi all,

I am currently using the following valves in my amp and am considering experimenting with others, is there any benefit in "rolling"? what can I expect (if anything?) The amp came with these valves in it and they were all fairly low hours use according to the previous owner. Over the last 18 months or so I use they have moderate use (probably between 4 to 10 hours a week roughly) but the amp is never run hard (volume never goes about 9 am generally). I have no idea how long valves should last either?

Quad of Svetlana Winged C EL34 ,
pair of Shuguang CV 181-Z (6SN7) http://www.audioaffair.co.uk/Shuguan...r/product_3431
remaining valves are a pair of Mullard 12at7 .

Apparently I can also use 6550 valves as well (but I'm not sure what these replace....). Does any 1 type of valve have more of an impact on sound than another for instance?:scratch:

I believe that these are pretty decent valves, but can anyone recommend anything worth consideration and if so where to purchase from?

Thanks in advance :D

cheers

Ben

[Wakefield Turntables]

I'm afraid the only way of telling a difference is to bite the bullet and have a little experimapentation. I'm extremely sceptical with regards all the BS that surrounds hifi gear. I decided to tube roll my croft micro basic last night. I had a box of old valves mainly mullard and brimar ECC82's and decided to experiment. You only need to replace two valves in the micro basic, basically left and right channel. My amp came with JJ ECC82 which to my ears make a very very nice sound, however, plonk in the Mullards and that sound became even nicer. I can't put my finger on what I heard but I just knew that I preferred the mullards to the JJ's. I think one the secrets to tube rolling is having an extremely intimate knowledge of your systems "sound" you have to be able to accurately detail what you hear only then can you see how the valves impact on your system. I hope this makes some sense. ;)

[Ali Tait]

Different valves will give different sounds. It's worth trying as the sound can be tuned to suit your taste. Not much point generalising on how various valves will sound, as they will sound different depending on the amp they are used in. Just decide on a budget and buy a few to play about with. A good point about doing this is, once you have settled on a preference, you can sell the rest and recoup most of your cash.

The 6550's would go in place of the EL34's.

[Ali Tait]

Said I wouldn't generalise, but 6550's would probably give a more muscular sound than the EL34, which I've found to have a more relaxed presentation with a great midrange.

I had a PP EL34 amp for a while, I liked a set of cryoed Shuguang EL34, even above a set of NOS RFT EL34 which I bought to try.

[User211]

Rolling valves produces quite readily measurable aberrations in frequency response. Your mission is to find out which set of aberrations you prefer best!

[jandl100]

Careful changing the big power valves (the EL34 in your amp) as the amp may need rebiasing if you do, unless it's one that auto-biases in which case it looks after itself.

The small input valves (6SN7 and 12AT7) can be changed out without any such hassles and that's where I suggest you start.
Yes, significant changes can be heard, not just in freq response but in levels of transparency and detail as well as the overall nature of the sonic presentation.

I'd start by rolling the 6SN7 ... hmm, I just looked at the link you gave for yours and it doesn't present anything useful for me.
Is this them? http://www.thetubestore.com/Tubes/6SN7-Tube-Types/Shuguang-Treasure-CV181-Z
If so, blimey, they look special! :stalks:
It would be interesting to compare them with others, I have no idea how they sound but they ain't cheap!

My suggestion would be to ask if anyone here has some 6SN7 they could loan you to try out.
I have some cheap Russian 6SN7 that always sound good to me and that you would be welcome to try.

If you wanted to try some other 12AT7 I'd suggest JJ - cheaper than Mullard and a bit less 'sophisticated' sounding, but could well be clearer and more transparent and dynamic.
http://www.hotroxuk.com/jj-valve-ecc81.html
I'm a fan of JJ valves. :)

[Zoidburg]

Thanks for all the comments folks, sounds like it's really a case of trial and error eh.

I know my am amp is auto bias so I hope that means hassle free trials / valve swops.

Jerry those are the valves I currently have installed (they came with the amp) and yes they are a bit pricey by the looks of it, certainly I'd have to think long and hard about that level of expenditure..........decisions decisions......maybe il look at changing the el34s first. I will also pm you regarding your amazingly kind loan offer, Thanks dude!!

[User211]

not just in freq response but in levels of transparency and detail as well as the overall nature of the sonic presentation.

I'm not sure the frequency response aberations aren't responsible. For instance, a 6A3 output triode in the Lampy circuit has a pretty jagged frequency response especially towards higher frequencies compared to a VT-52. It sounds subjectively more "spacey" and much more of a "bad boy" when playing rock than its VT-52 counterpart.

The 6A3 also sounds quite a bit brighter - but that isn't because of an upward trend in its overall frequency response. Rather, it appears to be because of the notably higher standard deviation figures at higher frequencies.

[jandl100]

Yes, it can be difficult, maybe impossible, to deconvolve freq response changes from rez or soundstaging differences.

I think they are different things though.
I may be wrong. :)

[User211]

I fink it is da electwongs. They are moody buggers and they behave differently depending on the environment. Just like humans.

[jandl100]

:hmm: So you're saying that all valve differences are down to freq response changes? :hmm:

Nah, you're wrong! :ner:

I've heard different valves have very different control in the bass, for example, nowt to do with freq response at all. :)
Is that called 'crunch' in the DJ/pro world? :scratch:

[lurcher]

I'm not sure the frequency response aberations aren't responsible. For instance, a 6A3 output triode in the Lampy circuit has a pretty jagged frequency response especially towards higher frequencies compared to a VT-52. It sounds subjectively more "spacey" and much more of a "bad boy" when playing rock than its VT-52 counterpart.

The 6A3 also sounds quite a bit brighter - but that isn't because of an upward trend in its overall frequency response. Rather, it appears to be because of the notably higher standard deviation figures at higher frequencies.

Given that description I can't help wonder if what you are measuring is the level of microphony the valve is showing. In my experience IF the valve type you are replacing is equivalent you will see a measurable different distortion specta, but the frequency response should be constant.

[User211]

Given that description I can't help wonder if what you are measuring is the level of microphony the valve is showing. In my experience IF the valve type you are replacing is equivalent you will see a measurable different distortion specta, but the frequency response should be constant.

Nick - I think we have covered this one before. I'm just using a mic at the listening position at normal listening levels (well, quite high at 85DB) and pink noise. You might be right some of it might be microphony - even all of it. But the results are readily repeatable with each valve type. Given the plethora of output triodes a Lampy will take, you also see variations in output level e.g. VT-52 is a couple of DB below 6A3 all other things be equal during the tests.

The change in sound is also readily evident at much lower volume levels though. And the DAC is on a VERY solid table.

Try the same thing yourself with your DAC and see what happens. You have to compare with very little smoothing to see the obvious repeatable differences though.

I discovered this ages ago and proved it, posting the plots on the Wam measuring 211 output valves. Serge was crazy baffled.

I would expect the changes in imaging/soundstage to be more due to phase effects, but I haven't attempted to prove that.

These are all big old triodes though... more subject to microphony - but I think I had some plots that show the same thing happening to a lesser extent with Russian E88CC equivs and some Teslas but I know not what I have done with them.

Whether it is microphony or not the end result is what matters - the fact you can easily hear it esp. with a massive set of ribbons.

[lurcher]

Whether it is microphony or not the end result is what matters - the fact you can easily hear it esp. with a massive set of ribbons.

Well, I am with you all the way until the last part, IMHO, the reason it makes a difference is what matters to me, when you understand that then the difference in the end result can be understood or even predicted. A DHT on the output of a line level system is going to be way more microphonic than the same valve on the output stage of a power amplifier. I know, I found that experimenting with DHT preamps. But its not the valve change that is causing the frequency response, its the sensitivity to vibration and the gain that follows it that makes the change. Two 6a3's may be electrically identical but one is microphonic, the other not. If (as you do) you suggest other users of your DAC try a particular valve you like in their system, they may find entirely different results as the one you have and the one they have, while electrically the same differs in the level of microphony. And of course it will also differ depending where the DAC is in relation to the speakers and room modes.

I would expect you would see different gain from different types of valve, I assume its a non feedback circuit so the mu of the valve will alter the output level.

BTW, you will see very little difference in my dac, the output stage is a long tailed pair, so any microphony in the valve will tend to be cancelled in the output transformer.

[Arkless Electronics]

Given that description I can't help wonder if what you are measuring is the level of microphony the valve is showing. In my experience IF the valve type you are replacing is equivalent you will see a measurable different distortion specta, but the frequency response should be constant.

Agreed. Biasing points can change as well of course which contributes to sometimes quite marked changes in distortion. There could be small changes in frequency response caused by the way the valve drives the next stage or output as eg in a cathode follower output impedance is directly related to transconductance.

My new hybrid phono stage will be sold with a warning NOT TO "TUBE ROLL" as it is carefully adjusted to match the characteristics of the supplied valves. Why do people call them tubes when we live in the UK?? :rolleyes:

[Arkless Electronics]

Different valves will give different sounds. It's worth trying as the sound can be tuned to suit your taste. Not much point generalising on how various valves will sound, as they will sound different depending on the amp they are used in. Just decide on a budget and buy a few to play about with. A good point about doing this is, once you have settled on a preference, you can sell the rest and recoup most of your cash.

The 6550's would go in place of the EL34's.

6550's are not a suitable replacement for the EL34 Ali. They are very different valves and in fact in the 6550, G3 is internally connected to the cathode whereas in the EL34 it isn't. 6CA7 is the yank equivalent of EL34 whereas the KT88 is near as dammit a replacement for the 6550 :)

[User211]

I just wonder if you are right, Nick.

VT-52 is the only one that shows a significant variance in output level.

Running the DAC and making it immune to vibration will, I am sure, not mean that valve rolling does not alter the perceived SQ. I will try it at some stage.

BTW I just describe how valves sound in my system. I frequently issue YMMV warnings, for obvious reasons. Marco accused me of telling people how valves will sound and it just ain't true, boo hoo.

[Ali Tait]

Just going on what Ben said in his post Jez. :-)

[User211]

:hmm: So you're saying that all valve differences are down to freq response changes? :hmm:

Nah, you're wrong! :ner:

I've heard different valves have very different control in the bass, for example, nowt to do with freq response at all. :)
Is that called 'crunch' in the DJ/pro world? :scratch:
No read what I have said.

[lurcher]

Running the DAC and making it immune to vibration will, I am sure, not mean that valve rolling does not alter the perceived SQ. I will try it at some stage.

I am sure you will still find differences, but I think the differences may be far less. But still worthwhile in most cases.

[User211]

I am sure you will still find differences, but I think the differences may be far less. But still worthwhile in most cases.
The reason I bought the DAC is because of its deliberate design to respond and change to valve rolling. It is like having many DACs in one.

The circuit I an sure runs many valves sub-optimally but that is the point, really.

Some microphonic feedback is probably an advantage as different valves respond with different levels of the aforementioned, in some ways behaving like a turntable.

As ever you are welcome to bring yours round sometime and compare.

[User211]

BTW you can flick the valves with quite some force with high gain and no music and hear nothing through the speaker's. Just thought I'd say - in other words microphony is low in practice.

[lurcher]

BTW you can flick the valves with quite some force with high gain and no music and hear nothing through the speaker's. Just thought I'd say - in other words microphony is low in practice.

That would indicate that its low. However it may be that the level of feedback is frequency dependent. Easy to test, try measuring with the dac in the room and outside it (assuming you have long cables).

[User211]

Obviously it is also valve dependant - but they all seem good - even a 1920s balloon.

I am wondering if I have long enough I/C and whether I can be arsed, but I have encountered a few microphonic valves and a good flick makes them ring like f£%&. They see a good range of frequencies with a good hard flick.

Point taken, though. And it isn"t rocket science. But you rarely see definite, simple, completely sussed answers, so maybe there is more than meets the eye.

[Arkless Electronics]

I guess there is a conflict here between choosing valves purely by ones idea of what sounds good and choosing the ones which, in the circuit they are used in, give the lowest measured distortion and noise....

[User211]

I guess there is a conflict here between choosing valves purely by ones idea of what sounds good and choosing the ones which, in the circuit they are used in, give the lowest measured distortion and noise....
Or perhaps even a desire for constant change.