I know, there must be hundreds of such discussions everywhere, but they all seem to be regarding guitar amps where the interest is in distortion characteristics. I need to know the subjective and the objective differences between the two, especially in SET application.
Any subjective listening opinion on the qualitative differences?
What should I know about the technical differences to build a SET amp around them?
Thanks
Any subjective listening opinion on the qualitative differences?
What should I know about the technical differences to build a SET amp around them?
Thanks
I have Winged C EL34 on my SE UL amp with Silk opt comparing to 6p3S-E(6n3C-E in Russian). The EL34s have a more prononuced mid-high, so they are good for pop-rock. While the 6p3s-E(=6L6CG) sound balanced form low to high with good detail and any kind of music is OK.
Some times, more power is needed. A KT88 SEUL is better if you listend to classical music. I have another KT88 SEUL amp with popular James JS6123HS opt. KT88 Gold Lion Reissue is great, but it has nothing close to original GoldLion, so I rename it as Russian Lion. Svetlana KT88 sounds sweeter with bit weaker bass. Shuguang KT88-98 is not bad, sound like Chinese 300Bs.
Anyway, my favorite tube is 6p3s-E. It performs excellecnt in PP UL amp.
Some times, more power is needed. A KT88 SEUL is better if you listend to classical music. I have another KT88 SEUL amp with popular James JS6123HS opt. KT88 Gold Lion Reissue is great, but it has nothing close to original GoldLion, so I rename it as Russian Lion. Svetlana KT88 sounds sweeter with bit weaker bass. Shuguang KT88-98 is not bad, sound like Chinese 300Bs.
Anyway, my favorite tube is 6p3s-E. It performs excellecnt in PP UL amp.
EL34 was designed with as little curvature as possible, for linearity in
single ended and class A push pull. They cross with a slight wiggle in
class AB, the "Marshall" sound. HiFi amps operating in AB might need
a lot more global feedback to get rid of that wiggle.
6L6 was deliberately curved to cutoff on a square law, optimized for
better linearity in A and AB push pull crossing, "California clean". Most
beam power tubes share this same curve, for same reason. Ex: KT77
replacement for EL34. In SE, 6L6 type requires more feedback than
EL34 type to remove that curvature where it isn't actually helpful.
According to Schade, square law was a feature designed into his G1.
Its not at all clear that every beam power tube or upgraded clone of
his 6L6 will automagically inherit the original curve? I got feeling from
Schade's writings that he could just as easily have made it cutoff like
a normal Pentode if he'd wanted it that way. And if so, may be that
regular pentodes could be designed with square law curvatures too...
I have noticed EL84 pentodes crossing suspiciously smooth in AB.
Looking at XRays of G1, you see the grid coiled tighter at the ends.
Same what I see in 6L6. Though EL84 is not a beam power type.
Except one oddball RCA I have, actually is beam type. Go figure...
In short: For SE, best use EL34. Least curve is what you want.
single ended and class A push pull. They cross with a slight wiggle in
class AB, the "Marshall" sound. HiFi amps operating in AB might need
a lot more global feedback to get rid of that wiggle.
6L6 was deliberately curved to cutoff on a square law, optimized for
better linearity in A and AB push pull crossing, "California clean". Most
beam power tubes share this same curve, for same reason. Ex: KT77
replacement for EL34. In SE, 6L6 type requires more feedback than
EL34 type to remove that curvature where it isn't actually helpful.
According to Schade, square law was a feature designed into his G1.
Its not at all clear that every beam power tube or upgraded clone of
his 6L6 will automagically inherit the original curve? I got feeling from
Schade's writings that he could just as easily have made it cutoff like
a normal Pentode if he'd wanted it that way. And if so, may be that
regular pentodes could be designed with square law curvatures too...
I have noticed EL84 pentodes crossing suspiciously smooth in AB.
Looking at XRays of G1, you see the grid coiled tighter at the ends.
Same what I see in 6L6. Though EL84 is not a beam power type.
Except one oddball RCA I have, actually is beam type. Go figure...
In short: For SE, best use EL34. Least curve is what you want.
Last edited:
Do not agree with you after years of listening to SE amplifier diy or factory made.
In short : use 6P3S for best sound in SE and music of different type .
El 34 tend to be very unevenly for some music ....
It is only in my years maybe , but this is what I am hearing when compare .
And 6L6 original (NOS) I will never use but for guitar amplifier !
In short : use 6P3S for best sound in SE and music of different type .
El 34 tend to be very unevenly for some music ....
It is only in my years maybe , but this is what I am hearing when compare .
And 6L6 original (NOS) I will never use but for guitar amplifier !
Hi Sprinter, I am a hi-fi listener as the username suggests. I'll be happy to touch on the good questions you raise concerning 6L6 vs EL34 in a SET topology. First I will give some background info.
I decided to get the GC version of the 6L6 to try in my SEUL topology because at a rated 30 watts dissipation it is near identical to an EL34 in this respect. I also thought I might add that I keep plate supply voltage at 400 volts and use fixed bias.
Subjectively speaking, the 6l6GC became the clear winner due to phenomenal detail when compared to EL34 but this was not something that was readily apparent at the beginning. For about a month I been auditioning first one then the other, unable to tell much difference when I stumbled onto to something. That something was that the 6L6GCs gave a beautiful performance with bias set for 40 ma. By comparison the EL34s would sound the same to me set anywhere between 35ma and 60ma.
The technical differences between 6L6GCs and EL34s as far as building around them goes is just about mute. By the time you provide suitable bias adjust range for one you have done it for the other as well.
But there is a technical difference between 6L6 and EL34 that just might tip the balance in favor of the 6l6 in terms of listening quality. Owing to its' beam tetrode design the 6l6 has a far flatter plate current vs plate voltage curve than does an EL34. For more information on this try searching on The National Valve Museum.
In conclusion, you will not go wrong with either tube but for now, for these ears, the 6l6GC is the clear winner. Enjoy!
I decided to get the GC version of the 6L6 to try in my SEUL topology because at a rated 30 watts dissipation it is near identical to an EL34 in this respect. I also thought I might add that I keep plate supply voltage at 400 volts and use fixed bias.
Subjectively speaking, the 6l6GC became the clear winner due to phenomenal detail when compared to EL34 but this was not something that was readily apparent at the beginning. For about a month I been auditioning first one then the other, unable to tell much difference when I stumbled onto to something. That something was that the 6L6GCs gave a beautiful performance with bias set for 40 ma. By comparison the EL34s would sound the same to me set anywhere between 35ma and 60ma.
The technical differences between 6L6GCs and EL34s as far as building around them goes is just about mute. By the time you provide suitable bias adjust range for one you have done it for the other as well.
But there is a technical difference between 6L6 and EL34 that just might tip the balance in favor of the 6l6 in terms of listening quality. Owing to its' beam tetrode design the 6l6 has a far flatter plate current vs plate voltage curve than does an EL34. For more information on this try searching on The National Valve Museum.
In conclusion, you will not go wrong with either tube but for now, for these ears, the 6l6GC is the clear winner. Enjoy!
It is false impression to think that the output tube only makes the final performance or sound.
The voltage amplifier/driver stage desing can affect as much.
An example: a triode connected EL34 produces some 5 % of THD.
With suitably designed driver stage the overall THD can vary from some 1 % to 8 %.
The voltage amplifier/driver stage desing can affect as much.
An example: a triode connected EL34 produces some 5 % of THD.
With suitably designed driver stage the overall THD can vary from some 1 % to 8 %.
How can you ever have the same distortion signature with totally different impedance devices, biased totally differently, driving a totally non linear reproduction system, frequency dependent load impedances, non linear speaker core magnetic fields & current dependent speaker coil back EMF?
The whole "validation procedure" is an oxymoron, and makes me laugh.
Vacuum tubes are designed to distort.
They can't do otherwise, which is why most guitar players & SE amp owners love the melodious fuzz, and are deaf.
Last edited:
Look at for example 2A3 or 4P1L triode mode curves. Do you think it is hard to get very low distortion from these devices?
6l6s have the better characteristic's for that imo, I would be damn Happy to use them in anything though.
EL34s to me always still sound like they just want to get more pissed off/unbalanced etc imo
Regardless of whatever loads, just a usual characteristic to EL34s to me. Or maybe just how I hear it I don't know.
EL34s to me always still sound like they just want to get more pissed off/unbalanced etc imo
Regardless of whatever loads, just a usual characteristic to EL34s to me. Or maybe just how I hear it I don't know.
Last edited:
Different tubes really are different
The EL34 and 6L6GC really are different.
If you look at the characteristics for these tubes when strapped in triode, you'll see that they need different bias to achieve the same operating points. That means they have different transconductance and internal plate resistance, and they also have different mu (gain).
For a single 6L6G(A, B or C) in triode, with plate voltage of 350V, plate current of 40mA, load Z = 5k, grid voltage will be around -34V. Mu will be about 7.2, gm = 3.6mA/V, rp will be about 2k ohms.
For EL34 with Vp 350V, load Z = 5k, grid voltage will be about -29V, Ip will also be about 40mA. Mu will be about 9.7, gm = 6.2mA/V, rp will be about 1.6k ohms.
As you can see, the grid bias has to be different to get the same operating points into the same load with the same plate voltage. Gain and transconductance are quite different, and so is plate resistance.
They are different tubes. If you take a circuit that was designed to use EL34 and simply swap in a 6L6GC, the circuit will very likely not work as well, and that's likely to sound different!
--
The EL34 and 6L6GC really are different.
If you look at the characteristics for these tubes when strapped in triode, you'll see that they need different bias to achieve the same operating points. That means they have different transconductance and internal plate resistance, and they also have different mu (gain).
For a single 6L6G(A, B or C) in triode, with plate voltage of 350V, plate current of 40mA, load Z = 5k, grid voltage will be around -34V. Mu will be about 7.2, gm = 3.6mA/V, rp will be about 2k ohms.
For EL34 with Vp 350V, load Z = 5k, grid voltage will be about -29V, Ip will also be about 40mA. Mu will be about 9.7, gm = 6.2mA/V, rp will be about 1.6k ohms.
As you can see, the grid bias has to be different to get the same operating points into the same load with the same plate voltage. Gain and transconductance are quite different, and so is plate resistance.
They are different tubes. If you take a circuit that was designed to use EL34 and simply swap in a 6L6GC, the circuit will very likely not work as well, and that's likely to sound different!
--
Last edited:
This is all very good relevant info.
You will get cretins on youtube strumming away a guitar claiming they can demonstrate the "sonic differences" between the 2 valves and ask you to listen to their version of a cheapy microphone run through a mpeg coded in an effort to impress you with their complete technical ignorance.
It's a sad thing in an age of communication and easy access to good informational resources, the "everyman is an expert" has taken over to this extent.
I doubt most of the time whether 99% of "audiophiles" even have a basic grasp of physics and ohms law, but they are the ones keeping everyone up to speed on "skin & dielectric effects" in exotic speaker cables that simply can't exist at audio frequencies (unless your amp is oscillating heavily at 150khz!).
Impedance is just one of those determining aspects about most parts of a recording and reproduction system that matters, and matching those things is essential.
When you look at the chain, from the FET preamps in most high quality microphones, through the op-amps in the majority of studio mixers in through the ICs that do the A-D and back....you have all kinds of impedances to match from very high to very low.
...Ahum, never mind the distortion and crazy impedance changes introduced at the speakers from "motors" which are anything but LINEAR....
People then claim the only way to listen is through something that distorts even more... a chain of glass envelopes with red hot bits of metal emitting and bouncing off inconstant amounts of electrons...depending on lots of other semi-independent time constant influenced devices through TRANSFORMERS!
And you want to compare the sound of what?
A valve made in 1950 with one made in 1935, reproduced imperfectly by some Russians or Ukrainians 60 years later, or at a factory in Yunnan China?
You will get cretins on youtube strumming away a guitar claiming they can demonstrate the "sonic differences" between the 2 valves and ask you to listen to their version of a cheapy microphone run through a mpeg coded in an effort to impress you with their complete technical ignorance.
It's a sad thing in an age of communication and easy access to good informational resources, the "everyman is an expert" has taken over to this extent.
I doubt most of the time whether 99% of "audiophiles" even have a basic grasp of physics and ohms law, but they are the ones keeping everyone up to speed on "skin & dielectric effects" in exotic speaker cables that simply can't exist at audio frequencies (unless your amp is oscillating heavily at 150khz!).
Impedance is just one of those determining aspects about most parts of a recording and reproduction system that matters, and matching those things is essential.
When you look at the chain, from the FET preamps in most high quality microphones, through the op-amps in the majority of studio mixers in through the ICs that do the A-D and back....you have all kinds of impedances to match from very high to very low.
...Ahum, never mind the distortion and crazy impedance changes introduced at the speakers from "motors" which are anything but LINEAR....
People then claim the only way to listen is through something that distorts even more... a chain of glass envelopes with red hot bits of metal emitting and bouncing off inconstant amounts of electrons...depending on lots of other semi-independent time constant influenced devices through TRANSFORMERS!
And you want to compare the sound of what?
A valve made in 1950 with one made in 1935, reproduced imperfectly by some Russians or Ukrainians 60 years later, or at a factory in Yunnan China?
Last edited:
I have had good experience with both tube types...provided the circuit is properly optimized for the given tube type...
My feeling is that some tube types have gotten a bad review based on putting it in a non-ideal circuit...
Such as evaluating EL34's by substituting them into a 6L6 amplifier...may not produce ideal results...
Nailing down an optimum plate load for the given valve at particular operating conditions...Optimizing inductance, winding capacitance as well as leakage inductance, all must be taken in consideration...
Drive signals and feedback need different optimizations as well..
Feedback and gm for me affect the dynamics or "jump" of the amplified signal, due to input sensitivity changes..
I believe that when either valve is placed in optimum designed conditions, they can produce very similar results with respect to bandwidth...however each valve has their unique harmonic signature due to inherent non-linearities of the curves, due to dissimilar mechanical geometries... It would take a big dose of feedback to calm those subtleties ...At that point you have a tube amp colder than solid state amp...defeats the purpose... With reasonable amount of feedback both tube type would perform fairly similar in HiFi application...those with good ears will always here the signature of the valves...
Guitar amps is another story, since you are then pushing the valves in square waves and exploiting the harmonic signature of the particular valve..
My feeling is that some tube types have gotten a bad review based on putting it in a non-ideal circuit...
Such as evaluating EL34's by substituting them into a 6L6 amplifier...may not produce ideal results...
Nailing down an optimum plate load for the given valve at particular operating conditions...Optimizing inductance, winding capacitance as well as leakage inductance, all must be taken in consideration...
Drive signals and feedback need different optimizations as well..
Feedback and gm for me affect the dynamics or "jump" of the amplified signal, due to input sensitivity changes..
I believe that when either valve is placed in optimum designed conditions, they can produce very similar results with respect to bandwidth...however each valve has their unique harmonic signature due to inherent non-linearities of the curves, due to dissimilar mechanical geometries... It would take a big dose of feedback to calm those subtleties ...At that point you have a tube amp colder than solid state amp...defeats the purpose... With reasonable amount of feedback both tube type would perform fairly similar in HiFi application...those with good ears will always here the signature of the valves...
Guitar amps is another story, since you are then pushing the valves in square waves and exploiting the harmonic signature of the particular valve..
Last edited:
I know this is going off-topic, but I can't resist...
Everybody keeps saying vacuum tube guitar amps are only used for making fuzztone distortions by super-loud/metalhead guitar players looking to 'shred.' That just ain't true, dammit!
Many legendary jazz guitarists played and still play through push-pull 6L6 amps with clean tone.
- The inspirational Jim Hall played a really old Gibson GA-50, and got a supremely beautiful sound. I also heard him play through small solid state amps (Polytone). I personally thought he got his best sound through his old GA-50.
- Wes Montgomery played through Fender Twin Reverb amps much of the time. Wes played a Super Reverb on many occasions too. Supposedly that was one of his favorite amps. However, Wes went to an early solid state amp (Standel) late in his career.
- I heard Kenny Burrell play through a Twin Reverb (push pull, parallel 6L6's) on several occasions. He had a solid state amp with him on other occasions, but I think that was mostly out of the need to have a lighter, more portable amp.
- There are many contemporary guitarists who play through old Fender tube amps. Check out Peter Bernstein, especially. (He gets a spectacularly beautiful electric guitar tone.) I think he's still playing through a 1970s Fender Vibrolux Reverb.
There is a nice 'singing' warmth to clean tones that these tube amps can give, which is usually not forthcoming from amps based on solid state circuitry. A musician won't ask why and how, but will simply choose the tool that works best for expressing him- or herself in sound. Now, why is it that these ghastly old tube amps are so coveted for their *clean* sounds? I suppose it's distortion, but it's not from pushing the tubes into outputting square waves. It's more subtle than that. Give the musicians some credit, would ya?
--
Agreed...100% ..
I love the "clean" TUBE sound ..... I love tubes because they are not perfect...The warmth or "thickness" you hear is essentially harmonic enhancement...
Some audiophiles enjoy this harmonic enhancement while others are appalled at the slightest colorization.... Musicians thrive on it, whether it is from a tube amp or a old violin made from harmonically rich wood ..
I like a little warmth, "meat", to the audio...but just a modest amount to make the music smooth and sexy, without getting overly bloomy, sounding like a cheap tube amp...
Keep in mind the distinction between source harmonics and reproduction harmonics..
One day about 15 years ago...My friend came over my house with his girlfriend....She yelled at me for 10 minutes straight saying that every time they visited me they never heard music from my fancy homemade Hi-Fi system... Instead all she hears is test frequencies and sees network analyzer and spectrum analyzer displays... She said to just stop with all the calibrations and tweaking and just enjoy the damn music before they bury you with the scope probes still in your hands... Well it took a while to sink in but she was right... I now don't care as much as I use to, just as long as it sounds good...
Last edited:
"Clean" is all relative.
The vast majority of the distortion of a valve head is probably coming from the really basic design of speaker cabinet, & non linear behaviour of the cone and magnet, but enabling you to carry it around with a 15" or 18" driver inside without having a hernia.
Distortion is our way of life, it all depends which form you like most, and which is most socially acceptable.
Automotive technology is just the same.
We put 5L of gas in a modern high performance car, and throw 4.5L away into the environment in the form of heat, friction, noise, braking heat, & accelerating excess weight, then congratulate each other how advanced a society we have in the 21st century.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.