• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Tube Rolling: Does it Make a Difference?

D

Deleted member 43441

Guest
I recently reviewed the Mhdt Pagoda tube DAC. Owner was kind enough to send me three other tubes to test with it. Here are the set of tubes I received:

View attachment 168967

Let's start again with the tube supplied with the unit (GE 5670)

View attachment 168968

Now let's switch to Tesla 6CC42 (GA):

View attachment 168969

Other than slightly more mains hum at 60 Hz, I see no other difference. Gain is reduced negligible amount.

Next up is the Western Electric JW2C51:

View attachment 168970

Distortion is 0.7 dB lower. So nothing significant there.

Finally I tested the Western Electric 396A tube which produced a surprising result:

View attachment 168971

Channel 1 is the same but channel 2 has 6 dB lower distortion. Is half the tube better or is there some asymmetry in the design?

That's all I have for you. :)

Conclusions
It is clear that by far the source of distortion is the design and not choice of tube. In three cases the difference is negligible and inaudible. In the forth example, the WE 396A, there is reduction of distortion in one channel. In the larger picture, there is so much distortion here that reducing it with this tube is not going to make a difference one way or the other.

Notice the level of instrumentation it takes to see the impact of the tube. You need to measure to see if there is a difference and not just assume there is.

Personally even if I signed up for a tube product, I would not waste time "rolling tubes." You are likely to suffer more from "rolling placebo" by your brain than any variation in such tubes.

----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Thanks for this, and I’ve suspected it for quite awhile. I’ll tell you that there is a strong placebo effect, plus if you listen to enough people they’ll tell you everything from 1957 XXX tube to 1958 of the same tube sound different because they’re a year apart. Personally I think the differences that people believe they hear come from mismatched tubes, or different gains from different manufacturers possibly.

That being said, I just picked up a Freya + and I quickly swapped in a brand new NOS set of Sylvania‘s from back in the day. I have a good tube supplier/friend so I know what I’m getting.

How do they sound? probably just like the ones I pulled out…. but that placebo effect is strong indeed. I swear it sounds twice as good LOL.
 
D

Deleted member 43441

Guest
There is something about running a set of tubes from the 1950s, 60s, 70s. Even if it doesn’t sound any better, I just like the looks of them, and they sound excellent. Probably just like the new ones… but my new production ones are back in the box ;-)
 

DanaGer

Active Member
Joined
Feb 12, 2021
Messages
138
Likes
250
Location
USA
I'm not going to change anyone's mind here but 1950s and early 1960s tubes do sound better than the later tubes and these triods are actually a right and left channel in the same tube so it is very understandable that the sides would test differently. The good ones are also over 70 years old and most of them are used. That being said I have a MHDT Orchid dac and have upgraded the tube to a D Getter 6922 with an adapter and its unfortunately substantially better sounding and a lot more expensive. It's not a placebo effect (everyone who feels better says that) but when tubes fell out of being used as mainstream the quality of the internals changed and in my experience not for the better.

Tubes impart a specific sound signature which is why they are used in recordings. Tube microphone preamps and tube compressors are still being used today by major recording artists. Unfortunately no one is making 70 year old tubes anymore so all the SS only guys will eventually be able to stay pure to their beliefs but will they still feel the same way when it happens?
 

EarlessOldMan

Member
Joined
Apr 11, 2022
Messages
50
Likes
83
People would be better experimenting with putting different electric guitar pedals into their system if they like all their music to sound like it's being played by Edge from U2
Coming to this late . . . But, man, now I need to add a couple of Snarling Dogs Super-Bawl Whine-O Wahs into the signal path of my audio systems! Yeah, baby!

And, yeah, I have a Super-Bawl Whine-O Wah. Super-nasty. Love it!
 

57gold

Active Member
Joined
Feb 15, 2021
Messages
100
Likes
84
Anybody now convinced that NOS tubes don't make a difference can get rid of their stash of matched pairs of Mullard EL-34 xf2s and Blackburn factory 12AX7s, and Amperex brown based double getter EL-34s and Bugle Boy 12AX7s by sending them my way for $s.

And while were at it, $s for a matched pair or two of Visseaux 6V6s, last produced in 1952 for the French military. They kill in a 1959 Deluxe and a recent Carr Mercury V. As a backstop, recently picked up a quad of military issue Mazdas, also made in France around the same time for the Department de la Guerre...something about the Frenchie's great food, wine and....6V6s?

All set for wah wahs, have an 1970 Vox and a more recent Fulltone...but have to look up the Super-Bawl! Need to do some Jimi, Stevie and maybe the theme from Shaft at next gig.
 

DualTriode

Addicted to Fun and Learning
Joined
Oct 24, 2019
Messages
893
Likes
593
Hello All,
In a previous decade I had a AP2522 Audio Analyzer on my bench.
I made a study of the signature Harmonic series of different types of 12AU7 vacuum tubes.
The 12AU7 is famous for a ample amount of H2 distortion. My tests also confirmed this.
Some sample 12AU7's had almost zero H3 distortion. Put these low 3rd Harmonic tubes in a adjusted SRPP circuit the H2 harmonics would nearly cancel plus with the inherent low 3rd harmonic we are talking very low (for vacuum tube) levels of distortion.
This is more like tube, or if you like, FET matching than tube rolling.
Thanks DT
 

welsh

Senior Member
Joined
Jul 11, 2020
Messages
363
Likes
364
Hello All,
In a previous decade I had a AP2522 Audio Analyzer on my bench.
I made a study of the signature Harmonic series of different types of 12AU7 vacuum tubes.
The 12AU7 is famous for a ample amount of H2 distortion. My tests also confirmed this.
Some sample 12AU7's had almost zero H3 distortion. Put these low 3rd Harmonic tubes in a adjusted SRPP circuit the H2 harmonics would nearly cancel plus with the inherent low 3rd harmonic we are talking very low (for vacuum tube) levels of distortion.
This is more like tube, or if you like, FET matching than tube rolling.
Thanks DT
I can quite easily tell the difference between different types of valves in my guitar amplifiers, because the amps can so easily be pushed into serious distortion. American sound: 6V6, 6L6. British sound: EL84, EL34. However, I would never listen to a recording of said valve amps using valve audio gear; why introduce more distortion?
 

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,871
Location
Santa Fe, NM
I can quite easily tell the difference between different types of valves in my guitar amplifiers, because the amps can so easily be pushed into serious distortion. American sound: 6V6, 6L6. British sound: EL84, EL34. However, I would never listen to a recording of said valve amps using valve audio gear; why introduce more distortion?
The point of using a valve amp in playing a recording is not necessarily to introduce additional distortion. True some people use them specifically to do that, but don't paint all valve amps with that brush. The devil is in the details, and the problem is that many people (and you in this post) are glossing over these details.
 

Russm535il

Member
Joined
May 1, 2021
Messages
32
Likes
19
Location
Pittsburgh PA
I am so happy that this was posted !!! I have owned a tube integrated Rogue Audio Cronus Magnum 2 amp for approximately 5 years and really enjoy it paired with Golden Ear Triton 2 plus speakers. I have seen the term “ tube rolling” and thought it literally meant taking each tube out of the amp and rolling it between my hands !! LOL !! I could not figure out how that would change its sound ! Thanks to this post I now understand that it is simply switching out tubes for different models . I just resubscribed to Stereo Review !!
BTW I believe my tubes are Russian made so I might have to do some rolling eventually!
Best to all !
Russ Dejulio
Pittsburgh PA
 

DualTriode

Addicted to Fun and Learning
Joined
Oct 24, 2019
Messages
893
Likes
593
You have completely lost the nuance.

Some people like a little spice in their pumpkin pie latte.

I prefer my coffee black. Then you have your French Roast or Breakfast Blend.

A little H2 distortion may just be what you are missing.

The secret about a pinch of H2 is phase. Think of sonic stereoisomers.

Hook up your analyzer and watch the amount and phase of the distortion and listen to the variation of the sound stage.

Thanks
 

captainbeefheart

Senior Member
Joined
Apr 18, 2022
Messages
360
Likes
446
I didn't read all 22 pages so what I say may have been covered.

Does tube rolling make a difference?

Well yes and no, it depends heavily on the design and type of circuit.


One major difference between designs is feedback. Some tube amp builders and sadly consumers are afraid of negative feedback, yet little do they know that their beloved directly heated triodes are only linear because of the internal feedback mechanism, which is why there were advancements in tube technology to the tetrode/pentode to remove this mechanism.

The construction and subsequent tolerance of tubes can actually be better than transistors, I get batches of transistors where hfe is all over the place far worse than tube tolerances hence why they need to be screened for their hfe before installing or else you will have imbalances. But that's not to say tubes themselves within a production batch are all the same, they are not. So it's not even a matter of one brand vs another brand, or production year vs production year, there are differences within the same production run. This means you could install 10 different tubes from the same company, same production batch and have different results. To many this is not surprising since they deal with gross variations in transistors all the time.

Moving on the the circuit. A good design shouldn't rely heavily on such small differences within the same type of tube, say 5670/2C51/WE396A/6N3/6N3P. But some of these tube amps, typically the single ended triode designs do not have any feedback. These types of amps have the first problem of needing a lot of gain to get the large enough signal to drive the power triode, 140v peak to peak is common. That's a large signal and linearity is of the utmost importance, small differences can lead to pretty drastic distortion profiles. Then there is the output triode, it doesn't have much gain, maybe 3x, but it has a plate swing of 420v peak to peak or more, linearity is important again with no feedback. With these amps changing tubes can alter the distortion profile enough to be audible, these are the amps that if you are into tube rolling you'll get the biggest differences.

Then there are regular tube amps using feedback. We learned pretty quickly that feedback has a great advantage in performance so most of these single ended triode designs are from 100 years ago, that doesn't mean you can't add feedback to them, it's just most look at it like a religion and feedback is the devil. Take and amp like the McIntosh tube amplifiers, the output stage is under almost 100% feedback hence the "unity coupled" name. You can swap as many power tubes as you want and you won't hear or measure a difference. The circuit has nested feedback within the global feedback so the whole circuit is very clean and tube rolling really makes no difference at all. So long as the tubes are functioning in their normal operating conditions any tube you use will yield the same performance at the output.

To summarize tube rolling depends heavily on the type of circuit they are used in. Good designs only require a functional tube to make performance spec while other non-feedback types the performance can heavily rely on screening through tubes until you find the best specs.
 

EarlessOldMan

Member
Joined
Apr 11, 2022
Messages
50
Likes
83
I have three tube amps.

I use 'em because they look cool. I'm not going to pretend that they're cleaner or more accurate than solid state or digital amps.

But I ain't claimin' nothin' 'bout nothin' nobody else should do or not do. Y'all go out and do what pleases you. I ain't arguin' none with y'all.
 

DualTriode

Addicted to Fun and Learning
Joined
Oct 24, 2019
Messages
893
Likes
593
Have you done this ears-only?

Hello,

Not ears only, more like seat of the pants.

Nelson Pass planted this seed in my brain.

I have seen this topic discussed more using JFET’s to source the H2 and to a lesser extent H3 harmonics.

I have built the H2 V2 PCB with J113’s (from the store on the other site).

Using a Keysight B2962B bench power supply and an APx555 I have seen all the distortion and phase effects discussed by the One and Only. No ears involved here.

Next, minus the analyzer I put the H2 V2 in the stereo system chain across the room. My seat of the pants impression is that there are changes to the perceived dimensions of the sound stage. At less than 1% added distortion I cannot say that I noticed any perceived distortion.

At this point it is fun playing with it. This is in the direction of playing with error correction to obtain the best SINAD possible then adding a little flavor to sample on your own.

Listening to stereo is all about the illusion in the first place. You know the center channel and all the rest, plus all the sound bouncing off the walls and other surfaces.

I do not see any fault in playing with it. Dial in the amount of distortion, H2/H3 ratio and left or right phase. Just need a couple of dials to turn.

My wife did not walk in the door to say she could hear a difference.

Thanks DT

https://paia.com/proddetail.php?prod=9305KEN fun toy
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Hello,

Not ears only, more like seat of the pants.

Nelson Pass planted this seed in my brain.

I have seen this topic discussed more using JFET’s to source the H2 and to a lesser extent H3 harmonics.

I have built the H2 V2 PCB with J113’s (from the store on the other site).

Using a Keysight B2962B bench power supply and an APx555 I have seen all the distortion and phase effects discussed by the One and Only. No ears involved here.

Next, minus the analyzer I put the H2 V2 in the stereo system chain across the room. My seat of the pants impression is that there are changes to the perceived dimensions of the sound stage. At less than 1% added distortion I cannot say that I noticed any perceived distortion.

At this point it is fun playing with it. This is in the direction of playing with error correction to obtain the best SINAD possible then adding a little flavor to sample on your own.

Listening to stereo is all about the illusion in the first place. You know the center channel and all the rest, plus all the sound bouncing off the walls and other surfaces.

I do not see any fault in playing with it. Dial in the amount of distortion, H2/H3 ratio and left or right phase. Just need a couple of dials to turn.

My wife did not walk in the door to say she could hear a difference.

Thanks DT

https://paia.com/proddetail.php?prod=9305KEN fun toy
I can absolutely believe the measurements. But Nelson's "distortion as colorant" stuff remains completely evidence-free. And with some of @pkane 's software, those distortion profiles can easily be dialed in and actually tested for audibility without laying out a pile of money for actual amplifiers.
 

captainbeefheart

Senior Member
Joined
Apr 18, 2022
Messages
360
Likes
446
I was reading a blind test on the effects of audible distortion and for the most part, it's personal but there was some interesting correlations from the results. Some listed in order of their preference and although I'd say the majority liked it on the cleaner side there were several people that preferred 3% THD.

The thing I found interesting was that for the most part the second lowest distortion test was averaged out to be the most preferred sound and often was answered as the cleanest which it wasn't. I believe it was somewhere around -60db or .01% THD. So there is something going on here where the majority of people thought the .01%THD was the cleanest in the bunch and sounded the best compared to -100db.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
I was reading a blind test on the effects of audible distortion and for the most part, it's personal but there was some interesting correlations from the results. Some listed in order of their preference and although I'd say the majority liked it on the cleaner side there were several people that preferred 3% THD.

The thing I found interesting was that for the most part the second lowest distortion test was averaged out to be the most preferred sound and often was answered as the cleanest which it wasn't. I believe it was somewhere around -60db or .01% THD. So there is something going on here where the majority of people thought the .01%THD was the cleanest in the bunch and sounded the best compared to -100db.
Where would one find this?
 

DualTriode

Addicted to Fun and Learning
Joined
Oct 24, 2019
Messages
893
Likes
593
I can absolutely believe the measurements. But Nelson's "distortion as colorant" stuff remains completely evidence-free. And with some of @pkane 's software, those distortion profiles can easily be dialed in and actually tested for audibility without laying out a pile of money for actual amplifiers.

Measuring with a resistor as a test load changes everything. I can measure things that I never could dream of hearing from real speakers over room noise let alone analyzer and or amplifier fan noise.

I can use a calibrated AP 376M03 microphone or a pair of GRAS46AG calibrated microphones, room noise and fan frequencies look like distortion to the APx500 software.

Even turn off the power amplifier and run an APx500 test sweep with microphones, the APx500 acoustic software quantifies room noise and fan frequencies as H2 and H3 distortions.

It amuses me that people can claim that they can hear low level distortion without a thought about the level of masking room noise or distortion created by the driver.

Turn up the distortion; you might start to hear the distortion. Perhaps the distortion phase makes a difference?

Thanks DT

An example:

This is the peerless mid-sub driver that I use on my test bench.

Some of what is labeled H2 and H3 distortion is APx555 and APx1701 fan noise.

Level and Distortion -_ Smooth 2.jpg


This one the amplifier is turned off and scale is zoomed in.
Level and Distortion -_ Smooth -_ Smooth room and fans.jpg
 
Last edited:

pseudoid

Master Contributor
Forum Donor
Joined
Mar 23, 2021
Messages
5,116
Likes
3,416
Location
33.58 -117.88
Top Bottom