Tube Rolling: Does it Make a Difference?

[dfuller]

Interesting, weird how they make them and yet call them beam tetrodes. They also are rated for 30 watts dissipation instead of the normal 25 watts. Have you smashed one open or can you see the third screen grid through the plates? I have really no first hand experience with them but I remember when they came out and were touted as a beam tetrode and so suited better for hifi.

You can see them through the plates, plus you can see 3 sets of grid posts up top.

 

[captainbeefheart]

You can see them through the plates, plus you can see 3 sets of grid posts up top.

Thanks for the information!

No idea why analog metric puts the EL34-b in that category but it must be a mistake or the sales team (email address linked in the datasheet) didn't quite understand the engineering team? Who knows.

Funny I found videos I'll post below showing the tube being made in the factory and you're correct, I don't see any beam plates and you can see her install the suppressor grid before adding the first half of the plate (timestamp 1:15). It's in the very first video.

Birth Of The Shuguang EL34-B Vacuum Tube - thetubestore Blog

Behind the scenes video of the EL34-B tube production process. Step by step from internal assembly, to sealing the bottle, and finally flashing the getter.

blog.thetubestore.com

 

[Poultrygeist]

That makes sense. Thanks for the clarification.

Do you know if the later units use both 12AX7 and 12AU7? I believe that's how they are being made now.

When Dared first switched from the 6N3 their next version came with four 12AX7 tubes. After that I didn't keep up with any changes.

The sound was so bad when I first bought the Dared amp that I called the Illinois importer in an effort to make a return. He was quick to recommend I replace the no name 6N3 tubes with WE396's which made all the difference in the world. The Dared power tubes were 2A3C Shuguangs which I've enjoyed for years in other DHT amps.

 

[captainbeefheart]

When Dared first switched from the 6N3 their next version came with four 12AX7 tubes. After that I didn't keep up with any changes.

The sound was so bad when I first bought the Dared amp that I called the Illinois importer in an effort to make a return. He was quick to recommend I replace the no name 6N3 tubes with WE396's which made all the difference in the world. The Dared power tubes were 2A3C Shuguangs which I've enjoyed for years in other DHT amps.

I would love a peak at the inside to see the circuit they used. I have used the 6N3 many times and it's a pretty decent tube, my test setup was 3x the plate impedance to see how they differed, 22k plate load, datasheet says 40k plate to plate for class AB1 operation which works out to 20k Class A and 10k Class B. Typically like I said to test triodes and how they do with decent loading I like to pick 3x the plate impedance which matched up with the datasheet quite well. Bias was -3v and 5mA of plate current. My test results showed that the 6N3 types I had were as good as any other variant in the family. I had similar findings as Amir with getting the same numbers as I got with GE JAN 5670, I had some random 2C51 mostly Tung-sol and GE then of course the WE 396a.

I'm not doubting your experience, I always think there is a reasonable explanation for something so drastic like in your experience because a normal functioning 6N3 really shouldn't be much different in terms of electrical characteristics between 2C51 types. My gut either says Dared got a box of bunk 6N3 tubes that they didn't screen them before installing. Sadly I think the culls in a production run don't get thrown out, they still get sold off at a reduced price. Other common tubes from China are 6N2 and 6N1 but they have different pinouts and wouldn't work so I doubt they were mislabeled tubes which I have seen. The electrical parameters are pretty much identical except for heater current, the 6N3 does use more heater current, I have seen near 400mA from them where a 2C51 is 300mA. Possible if they have the preamp tubes setup to run DC for the heaters there could have been current starved with the 6N3. This is a head scratcher.

If you are decent with a multimeter one quick thing would be to compare operating voltages between the 6N3 and the We396a and see how different they are, it's possible the batch was very weak in regard to emissions and the low transconductance causes those tubes to sound bad.

It's just you shouldn't have to pay out of pocket for tubes that actually make your amp function properly, it should come working within stated performance specifications. I got into SET amps for a while ( complete designs ) and with zero feedback types which this most likely is you can get slightly different performance when changing tubes out, typically small differences the spectrum analyzer shows but the ear it's not so black and white. The worst I have seen this happen is with 2 stage designs where the first stage is a voltage amp with a ton of gain, the linearity of this tube can make a difference in the performance more so than the output tube, for example you probably need 120v peak to peak to drive those 2A3's, bringing up line level (1v peak) to full output you need a gain of 60 or +35.5db. That's asking a lot from one stage so we typically use a very high impedance load like a gyrator or CCS to get maximum linearity and make variations of tube performance less of a variable in performance but with resistor loading there can be some pretty large performance differences. But, this isn't a two stage design, there is actually 4 triode sections each channel to drive the 2A3, this is so much easier than trying to do it with just one triode stage. With so many triodes at your disposal one should make it pretty easy to drive the 2A3 and not push the 6N3 to it's limits. With weak tubes you may not only have lower gain and higher distortion, you might also run into some slew rate limiting trying to drive the input capacitance of 2A3. We know slew rate is a large signal phenomena and the 2A3 needs a very large signal to bring it to full power, the larger the signal amplitude the more current you need to drive the input capacitance and if the tubes are weak then you could also be getting not just bad performance at all frequencies, it will get worse as frequency increases. Did the amp sound worse the louder you pushed it? This would show the theory that the tubes were very weak and they just let amps out with bad batch of Chinese tubes which is not rare at all. I know a well known guitar amp builder that bought an entire box (100 count) of 6N2's from China cheap for guitar amps every single one of them were bad, they worked and passed a signal but instead of gain of 60 it was like 20, and gm was supposed to be 1.6mA/V but was .8mA/V, half of what it should be, they sounded awful. This is something he did regularly and typically the 6N2 is a fantastic sounding 12AX7 for guitar duty but this one batch was just pure junk, they were duds that were boxed up and sold and this probably happens far too frequently than it should. The simplest answer is most likely the correct one and that's where I got my money, bad batch of tubes. I order the 6N2 directly from Analog Metric for $4.75 each and usually get 30 of them at a time for guitar amp builds and also for replacements in hifi gear. I screen the whole batch when they arrive and put them into two piles, higher distortion variants go into guitar amp builds and the lower distortion ones get put into hifi.

If you're in any way interested in figuring this out feel free to throw them in the mail and send them my way, I'll rigorously test them and see how they behave and what their performance is compared to my stash of 2C51 types I have here. I'll post results here and then mail them right back to you. First Class USPS padded envelope is about $5, I'll even pay the return shipping as I'm interested in this also. I don't know what you pay for the 2A3C but I get the PSVane 2A3C for $68.31 each. I typically see them being sold from vendors for $200 a matched pair but I get it, they need to make profit on them where I am getting them from the same source they are.

 

[srkbear]

When Dared first switched from the 6N3 their next version came with four 12AX7 tubes. After that I didn't keep up with any changes.

The sound was so bad when I first bought the Dared amp that I called the Illinois importer in an effort to make a return. He was quick to recommend I replace the no name 6N3 tubes with WE396's which made all the difference in the world. The Dared power tubes were 2A3C Shuguangs which I've enjoyed for years in other DHT amps.

How is the sound different? Could the stock tubes have been faulty? Did you notice a difference in both channels, or just one? Did you try isolating the channels when listening and swapping the tubes to see if one channel sounded worse than the other? And did you swap out a matched pair? Tubes can have subtle differences between them, or one set may outperform another if one pair has higher gain than the other. But unless the tubes were faulty, gassy or otherwise underperforming a transformative difference is highly atypical in clean operating ranges.

 

[avanti1960]

Matt, you have just had a real first hand experience with your ear/brain fooling you into thinking it had a huge change in performance. You would swear on a stack of Bibles that the change was real, very large and anyone could hear it. But, when tested, it would not be noticeable. That is how over powering and amazing the effect is. This is why subjectivists have fits about measurements. To them and your own experience it is real. Your brain is saying this is real and huge. That is why it is so tricky. Even Amir is NOT immune to the effect. It affects everyone. There is no way to out think it.

It is amazing isn't it? When you really get to wrap your head around the idea, it is an eye opener to discount so much of the audio worlds marketing mumbo jumbo crap. Until it happens to you it is hard to believe. Heck, I was tuning in an EQ for my sub awhile back and made a bunch of big changes to really dial it in perfectly. Then when I had "arrived" and was all set to call it a day, I checked the sub and found out it WAS TURNED OFF. I had turned it off a few hours before and forgot to turn it on again. That was yet another slap in my face at how strong the brain is in fooling us. It really is amazing when you think about it. Thanks for a great post!

This position is truly laughable and I sincerely mean no offense by it. It clearly shows a gross ignorance and lack of experience with tube amplification products.

 

[avanti1960]

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/

You may have since changed your viewpoint but the conclusion is based on a very limited sample size and a product that may be sonically insulated from the effects of different tubes.
To really hear the differences one would need a product that has tubes placed in the circuit that allow their effects to be more audible.
Tube based preamplifiers, especially those using tube rectifiers are especially noted for how tubes influence the sound. Also the input stage of a tube or hybrid amplifier is another influential position for a tube circuit.
I bought my first tube based product ~ 10 years ago and was so impressed with the sound quality that it started a minor revolution in my home audio hobby. Since them I have owned several tube preamplifiers, tube integrated amplifiers and tube power amplifiers. I eventually landed on a 6SN7 based tube preamplifier driving a Class A solid state amplifier.
Having gone through the research of the NOS and new production tube world including drawers full of 12AU7, 12AT7, 12AX7, 6SN7 GTA, GTB, 5AR4, GZ34 variations the idea that different tubes do not influence the sound is based on lack of experience and use of tubes and tube products, nothing more, nothing less.
I would love to send you my preamp and a set of tubes for your review but at $500 Fed Ex round something I am not willing to do. Peace.

 

[SIY]

This position is truly laughable and I sincerely mean no offense by it. It clearly shows a gross ignorance and lack of experience with tube amplification products.

Do you have data and evidence or just insults and anecdotes? I don't see anything incorrect here, but perhaps 50 years of experience with tubes, designing, building, and using, is insufficient.

 

[Spkrdctr]

This position is truly laughable and I sincerely mean no offense by it. It clearly shows a gross ignorance and lack of experience with tube amplification products.

I think you may have accidently highlighted my post and you wanted the one above mine.

 

[avanti1960]

Do you have data and evidence or just insults and anecdotes? I don't see anything incorrect here, but perhaps 50 years of experience with tubes, designing, building, and using, is insufficient.

Sincerely no offense meant to anyone. However I did not make any assumptions about what others may hear and I feel insulted when others make assumptions about what I hear and experience.
You have 50 years of detailed experience with tube electronics and sincerely cannot hear the differences among different tubes?
If that were the case for me I truly would believe everyone who does hear differences are fulfilling expectation bias / placebo effect.

 

[SIY]

If that were the case for me I truly would believe everyone who does hear differences are fulfilling expectation bias / placebo effect.

Often they are, yes. In the absence of a pathological out of spec tube, probably always. So one could take the claims of baseball card style tube collectors with extreme skepticism until someone actually demonstrates it ears-only.

The big differences between tubes are noise, noise corner, microphonics, and matching. And the spread of these within a tube brand/vintage is nearly as great as the spread in a multibrand population (see, e.g., Merlin Blencoe's paper on triode noise). This, of course, is only important for phono and mike input stages.

 

[captainbeefheart]

^^^^^^^^^^^^^^^^^^^^^^^^^^ This

Even some extreme designs, say a two stage SET amp with zero global feedback where the first stage has it's work cut out for it in terms of gain and driving the output triode differences in linearity of each tube in the signal chain can give slightly different distortion content. This is the probably the most extreme situation where tube rolling 'may' cause a slight performance difference but most likely not audible enough to warrant some of the claims I have heard.

Most decently designed amps tube rolling is pointless, unless you have a strong desire to play with electronics yet don't know enough to actually do anything besides pop tubes in and out of their sockets.

How do I know perception bias is so strong, I hear people claiming ridiculous changes in sound when swapping out a like for like rectifier tube in a pure Class A circuit. For example swapping a run of the mill 5AR4 and then to a Mullard GZ34, the latter is said to have better bass and midrange, smoother treble, soundstage you can reach out and touch the piano etc..... The rest of us roll our eyes and know like for like rectifier tube in a Class A circuit isn't going to change anything enough to make any audible difference. I will attest that the Mullard GZ34 is a remarkably tough tube, I use one only for in the shop for troubleshooting and building because I know it's tough and can handle on/off cycles and just built well and have a very long service life. That's really the only advantage to them and why I once used them a lot but when the price went to over $100 per I quickly sold them all but one to fund other projects. Anytime rarity and price out weighs the performance I sell them off, it's crazy what I have purchased from just selling tubes to audiophiles. I don't market them with any hype, they ask me for them and I sell at a lower rate than market but still crazy money for me.

 

[dfuller]

Within the same type though say 6L6GC vs 6L6GC the clipping should be relatively the same unless like you say the construction of the tube is very different but then it wouldn't be the same tube type.

I realized I missed this bit - my understanding comes from guitar amps, where the 12AX7 is used pretty much ubiquitously. There are probably two dozen different physical designs of 12AX7, and they all sound a little different when driven into clipping. Again - not major, like... at all (very difficult to pick out, even sighted, it's not "night and day" nonsense at all), but definitely there.

The biggest differences though are between tubes that meet spec and those that don't. "This tube sounds anemic!" = something in the construction of that particular tube is off so it doesn't meet spec. Within reasonable (20%) tolerances, even the most golden ear type would be hard pressed to hear a difference.

When they're run in their linear region, they're all exactly the damn same.

 

[tomelex]

Tubes are highly variable within and between brands. Yes, I have heard differences as mentioned above #472, in a simple SET unit with a single driver, and you are driving it hard. However if the tubes are pretty well matched and measured properly, differences can be so small that you may be just fooling yourself to hear a difference. And so what if you hear a difference, other than you prefer one difference to the other it really does not make any points for or against a particular tube.

 

[Spkrdctr]

Do you have data and evidence or just insults and anecdotes? I don't see anything incorrect here, but perhaps 50 years of experience with tubes, designing, building, and using, is insufficient.

SIY, so you are saying your a tube beginner?

The big question is do you still do tubes? I guess a recommendation from a pro like you would carry a lot of weight for tube newbies. I would try tubes but they don't really lend themselves to surround sound very well.

 

[SIY]

SIY, so you are saying your a tube beginner?

The big question is do you still do tubes? I guess a recommendation from a pro like you would carry a lot of weight for tube newbies. I would try tubes but they don't really lend themselves to surround sound very well.

I stopped for a couple years because we had moved to Arizona, and we couldn’t deal with the heat load. Now that I’m in a cooler place, I’m starting two new power amp projects.

 

[Xulonn]

I would try tubes but they don't really lend themselves to surround sound very well.

5 pairs of KT90s in push-pull for 70WPC from Meixing Mingda, a well-regarded Chinese tube amp manufacturer...

 

[MattHooper]

As I mentioned I perceive differences in tube rolling with my CJ premier 12 amps. They seem very distinct, as I described earlier in the thread.

Given, as I also mentioned earlier, I have two sets of the same amplifiers at the moment (two sets of CJ Premier 12 monoblocks) I was considering doing a blind test for
tube rolling. I do like doing blind tests, unless it just gets too much hassle. I did get around to my recent blind test of my CJ tube preamp vs my Benchmark LA4 (I posted those results in a dedicated thread). I have a similar level of confidence I could pass blind tests for telling between certain tubes in these amps. (And, of course, I could be wrong, that's the point of blind tests).

Unfortunately looking in to how to achieve it, it looked like a hassle and for an electrical ignoramus like me I'm super worried about doing anything that could threaten the well-being of the amplifiers/speakers. But more pertinent at this point: I won't get a chance because current financial imperatives mean I have to sell one of the pairs of amps immediately.

So while I would have enjoyed doing a blind test for tube rolling for both personal reasons and to add some fun "data" to the site, I'm afraid it won't happen.

As to my impressions of tube rolling, they are strong and distinct enough (and so similar to the types of differences I established I heard when comparing my preamps) that I'm going to "roll with them." And keep having fun with it. Without, of course, any expectation someone on this site should take my impressions as solid data of any kind.

 

[piczus]

I am into tube rolling.... I reached extraordinary results with brand combination of tubes (Preamp - Miniwatt Ecc81, Brimar 12ATWA, Tesla Ecc802s, Poweramp - Tesla Ecc802s / RCA 5814A, Reflektor 6N23P / Tesla E88CC). In nutshell.... only NOS, ideally military production.

Below ..... my DIY mono poweramp project - SET (~17W) with Ecc82 + Ecc88 + GU50 ( my version in aluminium chassis including "magic eye", first version in wood of my friend). Cheers!

 

[beefkabob]

I am into tube rolling.... I reached extraordinary results with brand combination of tubes (Preamp - Miniwatt Ecc81, Brimar 12ATWA, Tesla Ecc802s, Poweramp - Tesla Ecc802s / RCA 5814A, Reflektor 6N23P / Tesla E88CC). In nutshell.... only NOS, ideally military production.

Below ..... my DIY mono poweramp project - SET (~17W) with Ecc82 + Ecc88 + GU50 ( my version in aluminium chassis including "magic eye", first version in wood of my friend). Cheers!

Welcome to the forum.
Where are your measurements of your extraordinary results?
No measurements? It didn't happen.