Best tubes for bass?

[ChopperCharles]

Since when are all tubes of the same designation supposed to sound alike? There are plenty of sites on the internet where people review the different sounds of different brand tubes.

For instance, I've been looking at a set of Gold Lion KT88's eventually, and there are a lot of reviews comparing them to other tubes on the 'net.

http://www.audiocircle.com/circles/index.php?topic=52094.msg465714;topicseen
http://www.audiokarma.org/forums/archive/index.php/t-121686.html


And hell, there's talk on guitar forums about how different 12AX7s make their guitar amps sound completely different.

http://thetubestore.com/12ax7review.html
http://reviews.ebay.com/12ax7-6l6-T...e-Tone-of-Your-Amp_W0QQugidZ10000000003445606


I mean, the whole point of "tube rolling" as it's called is to try out different tubes to see how they sound. Different brands of the same tube type can sound completely different. Manufacturing, design, I don't really understand why, but apparantly that's how it is.

The Electro Harmonix tubes DEFINITELY reduced the brightness in the amp. They didn't do much for the bottom end, but the over-driven treble disappeared.



Charles.

 

[Wavebourn]

Tubes don't sound; amplification stages that use tubes change the sound. If well designed they meant for tubes with certain parameters so when parameters are out of range sound changes VS what was intended by original developer.

 

[keithgreenhalgh]

Tubes don't sound; amplification stages that use tubes change the sound. If well designed they meant for tubes with certain parameters so when parameters are out of range sound changes VS what was intended by original developer.

Wavebourn is right! Each tube has sets of graphs that you can design with(eg plate load resistor) or even the size of capacitors(eg cathode bypass capitator)The resistors and capacitors will determine the operating parameters.(where on the graphs of the tube,that the tube will operate.) If you do your calculations wrong you may be operating in a non linear part of the tube(distortion).Since there is only one set of graphs for a tube(eg 12ax7)these are the graphs that engineers use to design circuits.

 

[jnb]

a non linear part of the tube(distortion).

Maybe that is what is meant by the sound of tubes. Surely you'd agree that a differently constructed one could have different kinks in the curves, even if the specs are the same.

 

[keithgreenhalgh]

I am currently reading info on 12ax7(google:12ax7 specks). People do seem to believe thats it make a difference.I will keep reading and comment later.

 

[Zoran]

for the good bass, and whole bandwidth in general
it is essential the quality of the output transformer...
from the vital importance is primary inductance
which is depending of each type of the tube.
main parameter is Ri internal resistance of the tube...
the higher Ri, Lp increasing...
that is why manufacturers, by the rule, missing to wrote the Lp,
total C, gap, and Rdc of the secondary...
next parameter is Rdc og the secondary,
should be as small as possibile
(I have on 2A3 OT less than 0.5 ohm...)
usually that is around 5 ohm which is not good at all...
That is because Rdc of the secondary is determing the output resistance of the device and it is little bit lower than Rdc of the secondary...
It is from the vital importance that we have bigger dumping factor
for driving open buffle properly, that means tight, low bass...
expetially horns.
the smaller Rdc of the secondary the bigger D. factor...
Unfortunatly, standard window space at the laminates of OTs
usually is not sufficant to accept higher diameter wire to meet the lower dc resistnce value...
another thing.
Value of Ck kathode condenser is important,
bigger values, for output tube circuits, will not improve bass when we dont have Lp high, but will blur the midle, so keep the values like 22 to 47uF
Next thing is that the lower internal resistance of the power supply is highly welcome...again main thing is dc resistance of the power transformer...
.
like a one member said before in thread,
tubes have a litle with bass sound quality,
another parts and design have.
.
cheers

 

[keithgreenhalgh]

Well charles,the tubes definately look different! the first is like the one that came with my yaqin 300b.I don't know what to say.They are very different in looks.The only thing I can think of is that someone holds a patent for the way the 12ax7 is made so the chinese tried to make one differently.I think patents normally expire after 30 years.Anyone else have any ideas?

 

[Miles Prower]

Anyone else have any ideas?

Type specs concern just the parameters, such as internal capacitances, rp, gm, etc. They don't concern how any particular VT is actually made. I have a 6KD6 that's put together in the usual manner: cathode, control, screen grids, beam formers, and plate. Another one consists of two pentodes in one bottle, all elements connected in parallel. Much different construction, but same type, and same characteristics.

I also have some 6BQ7As, some of which have the two heaters in parallel, others have them in series. You can make 'em any way you want, so long as they are within spec. As for what difference it makes sonically, who knows? I use gNFB to even those sorts of differences out.

 

[keithgreenhalgh]

Type specs concern just the parameters, such as internal capacitances, rp, gm, etc. They don't concern how any particular VT is actually made. I have a 6KD6 that's put together in the usual manner: cathode, control, screen grids, beam formers, and plate. Another one consists of two pentodes in one bottle, all elements connected in parallel. Much different construction, but same type, and same characteristics.

I also have some 6BQ7As, some of which have the two heaters in parallel, others have them in series. You can make 'em any way you want, so long as they are within spec. As for what difference it makes sonically, who knows? I use gNFB to even those sorts of differences out.

So it is commom to have tube with the same numbers look different? Interesting,and if they have the same specks how can they sound different? I wouldn't think that they should.

 

[Miles Prower]

^^^^

It's quite common. You can get all sorts of reviews concerning how different 12AX7s, 6SN7s, 6SL7s, etc. sound. Different claims for different arrangements, some of which go for some $BIG BUX.

As to why it should make a difference, I really don't know, other than the obvious: cheap, microphonic, sloppily specced VTs as opposed to better brands that aren't microphonic, adhere more closely to published specs, and are built to higher standards. Even back in "the day", tens of thousands of poor quality VTs poured off the assembly lines. I'm not so certain that it's not more audiophoolery than an actual phenomenon. Like I said before, differences between VTs are evened out by the judicious application of gNFB. And unlike a lot of audiphools, I don't consider gNFB to be the Spawn of Satan (same goes for SS in HS amps).

 

[Johan Potgieter]

Choppercharles,

Thanks, those photos are very clear. Cranking my head sideways to gaze into the Chinese (!), and from what I can detect, one cannot say that the triodes themselves are all that different. The Chinese tubes appear to have different mica spacers (more than two), with some construction to make things more sturdy. This might be for extra low microphony or such.

What Miles mentioned amazed me; I have not encountered that, also coming from the 50s. But we used mostly RCA, Sylvania and Mullard - a few Telefunken. Long and short twin triodes came later. But it is true that as long as the specs are the same ......

Regarding sound, I don't want to open a hornet's nest, but one must remember that the commentaries on "rolling" all refer to guitar amplifiers, where tubes are often worked into distorting regions for a particular sound, in contrast to hi-fi where tubes are definitely not supposed to sound - and does not except for some voodoo sometimes expressed. As the man said on the e-Bay commentary, it is quite subjective. I also mostly read there about differences in quality, microphony etc, governing choice, not so much sound

But I worry when I read about "special high-gain" 12AX7s or such. That makes a mockery of specs. A number is assigned to a tube so that a designer can know its specs. If you are making a few dozen of the same thing (never mind a production run of 100s), you cannot be expected to "listen" to every singe tube's effect. Often such were used in instrumentation, where sound did not apply. In the hey-day we rejected tubes deviating by more than +/- 10% from specs (and I must have used hundreds). I also see a 12AY7 compared to a 12AT7, and 6550s to KT88s. Those are different enough to work different (and then probably sound different) in the same circuits. Further, I once read a quite extensive "sound" test (guitars again), also giving the measured anode currents of some 12 different examples of the same number tube "rolled" (same circuit). That varied by more than 50% - no wonder they sounded different!

Coming back to the original question, and depending on how "more bass" is defined, it might be simpler to just include a bass augmenting filter somewhere than rely on this business of hunting for the right tube.

 

[keithgreenhalgh]

Choppercharles,

Thanks, those photos are very clear. Cranking my head sideways to gaze into the Chinese (!), and from what I can detect, one cannot say that the triodes themselves are all that different. The Chinese tubes appear to have different mica spacers (more than two), with some construction to make things more sturdy. This might be for extra low microphony or such.

What Miles mentioned amazed me; I have not encountered that, also coming from the 50s. But we used mostly RCA, Sylvania and Mullard - a few Telefunken. Long and short twin triodes came later. But it is true that as long as the specs are the same ......

Regarding sound, I don't want to open a hornet's nest, but one must remember that the commentaries on "rolling" all refer to guitar amplifiers, where tubes are often worked into distorting regions for a particular sound, in contrast to hi-fi where tubes are definitely not supposed to sound - and does not except for some voodoo sometimes expressed. As the man said on the e-Bay commentary, it is quite subjective. I also mostly read there about differences in quality, microphony etc, governing choice, not so much sound

But I worry when I read about "special high-gain" 12AX7s or such. That makes a mockery of specs. A number is assigned to a tube so that a designer can know its specs. If you are making a few dozen of the same thing (never mind a production run of 100s), you cannot be expected to "listen" to every singe tube's effect. Often such were used in instrumentation, where sound did not apply. In the hey-day we rejected tubes deviating by more than +/- 10% from specs (and I must have used hundreds). I also see a 12AY7 compared to a 12AT7, and 6550s to KT88s. Those are different enough to work different (and then probably sound different) in the same circuits. Further, I once read a quite extensive "sound" test (guitars again), also giving the measured anode currents of some 12 different examples of the same number tube "rolled" (same circuit). That varied by more than 50% - no wonder they sounded different!

Coming back to the original question, and depending on how "more bass" is defined, it might be simpler to just include a bass augmenting filter somewhere than rely on this business of hunting for the right tube.

Thanks for the info Johan. What you say makes sense to me.I was a technician and I cannot understand how two tubes with the same specks would sound different but everywhere people are talking about rolling tubes.I don't think people are going to change their mind if they are convined that tube rolling works.

 

[Brett]

Thanks for the info Johan. What you say makes sense to me.I was a technician and I cannot understand how two tubes with the same specks would sound different but everywhere people are talking about rolling tubes.I don't think people are going to change their mind if they are convined that tube rolling works.

Most tube rollers are non-technical, even if they are DIYers so they don't know. Perceptual bias would nake a massive difference in that case I would think.

I would be very surprised if there were significant measured differences between NOS 6SN7's of various brands, with say current 12AX7 types, as the latter are mostly used for MI (largest part of the market) and the tubes may be deilberately mis-made to produce a desired result. Musos, almost to a failt, are completely technically ignorant.

 

[jnb]

Specs only give a generalisation of the story, unlike plate curves. I won't claim that one brand to the next is more significant than the circuitry, but brands can differ noticeably. To a degree, and not always noticeable depending where used, one may sound relatively lean and another warm, and yet another can sound thick. Not unlike the differences between amps in general.

The most significant case of this I found was when comparing, IIRC, a Philips, RCA and Mullard 12AU7 in a hifi role.

 

[Brett]

Specs only give a generalisation of the story, unlike plate curves. I won't claim that one brand to the next is more significant than the circuitry, but brands can differ noticeably. To a degree, and not always noticeable depending where used, one may sound relatively lean and another warm, and yet another can sound thick. Not unlike the differences between amps in general.

Measuring some in the same cct will give a good idea of how closely that follow design specs.
I still don't buy that they'll sound much different if they measure close to each other and airy descriptors mean nothing to me. My experience doesn't back up such differences when tested blind. If a 12AX7 varies significantly from the specs of a 12AX7, it is either faulty or should be given another designator if it's a deliberate change on the part of the manufacturer.

 

[jnb]

and airy descriptors mean nothing to me.

Wasn't trying to impress . I've found enough instances where I couldn't notice any dfference, that I doubt what a single blind test can find. What I heard was consistently repeatable. 12AU7s are plug in replacements. They measured the same Va in circuit. If I made a mistake, I can't see how.

 

[Wavebourn]

If the rig was well designed use tubes in specs with what was expected by the original manufacturer of the equipment, otherwise results are unpredictable.
Shape of getter and color of base does not matter; what matters is how close are electrical parameters of the tube (especially dependences of currents on voltages) to what the rig was designed for. Even when RCA manufactured 6L6's all like twin brothers anyway Fender and others used trimpots to adjust idle currents. Those days all manufacturers tried to follow specs so tubes would be interchangeable no matter which brand was used. However, Germans always tended to follow specs and standards more thoroughly than Russians, that's why say NOS Telefunkens' are more expensive than say NOS Saratovs'. They are not better, they do not "sound fat, sweet, ..." (insert desirable), they are STRICT. If it is labeled say ECC82 it IS THE ECC82, but neither ECC81.999, nor ECC82.0001. If the Original Designer tuned his rig, selecting certain resistors to get the end result he was working on, using tubes with certain tolerances of parameters, it means that the tube must have that parameters in ranges. Everything else means extra distortions, and never better sound. Period.

PS: if to be honest, some Russian tubes are as strict as Telefunkens; if you see rhombus or a triangle with a digit inside, or a star with a digit inside stamped, that means military QA. Such tubes from the same box are usually close in 2% tolerance.

 

[jnb]

Microphonics are not covered under the term - specs. For that matter, not even necessarily in the plate curves. This is a dynamic behaviour.

Whether excited by thermal means, magnetic means or by physical feedback, valve parts do not behave like Behringer measurement mics - they are typically resonant and frequency selective.

A quality unit would need attention paid to the reinforcements (such as the plate ridges), grid supports, suspension, size shape and weight, glass etc. etc.

 

[Wavebourn]

Jnb;
actually microphonics are covered by specs for tubes intended for low signal usage.

 

[ChopperCharles]

Come over. Listen. I'm located in Durham, North Carolina. Add @gmail.com to my userid. Come over and compare in back to back tests.

When I swapped out the chinese tubes for the Electro Harmonix tubes, I noticed that the volume level stayed the same for bass and midrange, but the treble was noticeably lessened. And my opinion is that it sounds better this way. Come and give it a listen. I'll A/B the 12AX7s and the 6SN7s for you.


Charles.