Tube-rolling power tubes for clean

[Dusty Chalk]

So I have a very special MDT 50 and a very purple Archon 100/50, and I'm going to experiment with tube-rolling, mostly on the clean channel/settings. Does anyone with experience with tube-rolling know the answer to any of these questions?

• What does underbiasing or overbiasing the amp accomplish? Will either of those provide more or less headroom? High frequency extension? I'm thinking I probably shouldn't go much beyond...25mV-35mV?
• What does loading tubes with much higher or much lower wattage ratings accomplish, assuming I don't go completely out of spec? I've seen KT88s go just higher of 40W rated, and I've seen 6L6G rated as low as 19W. Assuming that I can bias these within nominal working parameters, which do you think would have better results?
• Volume is irrelevant -- I can never dime the master in my apartment, anyway, so would the 19W tube be a better choice anyway, or is that a function of ... whatever the word for gain is, where gain is the amount the amplifier actually literally amplifies the signal...? I know the word is gain in audiophile amp design, but that it's used differently on guitar amps.

I'll be attempting to answer these questions myself through experimentation, but any guidance would be appreciated. I've already emailed customer service about the limits of my amps parts of the questions.

 

[bodia]

So I have a very special MDT 50 and a very purple Archon 100/50, and I'm going to experiment with tube-rolling, mostly on the clean channel/settings. Does anyone with experience with tube-rolling know the answer to any of these questions?

• What does underbiasing or overbiasing the amp accomplish? Will either of those provide more or less headroom? High frequency extension? I'm thinking I probably shouldn't go much beyond...25mV-35mV?
• What does loading tubes with much higher or much lower wattage ratings accomplish, assuming I don't go completely out of spec? I've seen KT88s go just higher of 40W rated, and I've seen 6L6G rated as low as 19W. Assuming that I can bias these within nominal working parameters, which do you think would have better results?
• Volume is irrelevant -- I can never dime the master in my apartment, anyway, so would the 19W tube be a better choice anyway, or is that a function of ... whatever the word for gain is, where gain is the amount the amplifier actually literally amplifies the signal...? I know the word is gain in audiophile amp design, but that it's used differently on guitar amps.

I'll be attempting to answer these questions myself through experimentation, but any guidance would be appreciated. I've already emailed customer service about the limits of my amps parts of the questions.

Clueless....but good luck. Looking forward to you discoveries.

 

[elvis]

There are two parts to the bias point of a tube: current and voltage. The current bias is usually designed to find a place on the transfer curve (input voltage vs. output current) with the desired gain and/or linearity over the intended swing. Plate voltage can be modified to alter the linearity curve (and change the ideal current setting) as well as to modulate power.

You are working with fixed plate voltage, so many people will try to increase power to "heat the tube up" by increasing current. This can increase distortion by pushing the bias point to a less linear region on the V/I curve. Or it could go to a more linear region, but reduce the lifetime by increasing temp.

Most people try moving bias "blind", meaning they have no idea what the design parameters are for the amp, what the tube's operating curves look like, or what they are doing at all. I have been told that you can alter the crossover distortion with bias, which is sometimes true, but that is hard to quantify even with the right tools, and often seems optimal at a bias point that is nowhere near where it is suppose to be.

Running lower current is equally difficult to predict blindly, but at least you can bet that the tube life will be longer.

In my experience, people really want to believe they have that extra "edge", so they really want to modify stuff. Sometimes there is real science behind it. Sometimes it is placebo effect. I recommend running the amp the way it was designed to be run. If you disagree with the amp's designer, maybe that amp isn't for you.

 

[Dusty Chalk]

Taken to heart. I by no means want to operate the amp outside of its nominal operating parameters for precisely the reason you state. And I started this thread to hear exactly this sort of information that you stated at the beginning.

Even just biasing the amp from ~24mV range to 30mV with Groove Tues mediums (6s) has made an improvement, can't wait to get some hards in there.

 

[László]

I can't speak to bias, since I always bias to the recommendation of the amp designer. My experience is that higher power rated tubes will have more clean headroom, and this is consistent with the usual thing one reads in the literature.

So a 6L6 will have higher headroom than a 6L6WGB, etc. Volume isn't irrelevant; the volume at which a tube breaks up is a function of its clean headroom. The lower the headroom, the lower the volume at which the tube reaches saturation and distortion.

Generally, biasing a tube hotter than the manufacturer's recommendation will also cause it to break up a little earlier, at the expense of tube life.

This is why Mesa, for example, factory-biases a bit on the cold side; to avoid complaints about tube life from customers.

 

[Dusty Chalk]

So would it follow that biasing a little on the lower side would also increase apparent headroom?

Unfortunately, it also changes the tone. The MDT was biased a little high (~34mV). After biasing both to within tolerance (the MDT down, the Archon up), they sound a little more similar. The MDT slightly less sweet, the Archon slightly sweeter.

Well, that may have to be alright, I'll practice with them down, and experiment with them slightly up for recording. Just for tube life and frugality reasons.

 

[Dusty Chalk]

...My experience is that higher power rated tubes will have more clean headroom, and this is consistent with the usual thing one reads in the literature.

So a 6L6 will have higher headroom than a 6L6WGB, etc...

How did you determine this? Do you use Duncan Amps' TDSL site? Or is there something better, now?

 

[elvis]

Changing the bias slightly should not have a huge effect on headroom. I'm always surprised at the massive changes in tone and headroom ascribed to a relatively trivial change in bias.

 

[László]

How did you determine this? Do you use Duncan Amps' TDSL site? Or is there something better, now?

Never heard of Duncan Amps. I'm dating myself here, but I read books on guitar gear back when people did that, before the internet.

I also rolled power tubes with my old Two-Rocks; in this endeavor I was advised by Bill, who designed the amps.

But since I got my first tube amp back in 1965...yeah, I've been around awhile.

 

[László]

Changing the bias slightly should not have a huge effect on headroom. I'm always surprised at the massive changes in tone and headroom ascribed to a relatively trivial change in bias.

It's not a huge effect. It's a small effect. But it exists.

Differences don't have to be huge to matter.

 

[Dusty Chalk]

Never heard of Duncan Amps. I'm dating myself here, but I read books on guitar gear back when people did that, before the internet.

I also rolled power tubes with my old Two-Rocks; in this endeavor I was advised by Bill, who designed the amps.

But since I got my first tube amp back in 1965...yeah, I've been around awhile.

Yeah, yeah, we get it, and you used to read stone tablets and carry a sundial watch around on your wrist and have a pet dinosaur.

The Tube Data Sheet Locator. It's not perfect, but it helps. Not sure it's any better than google, though, if one doesn't know how to read a datasheet. What I really want is that "close or identical" and "different rating or performance" with the crucial numbers -- wattage rating range? recommended plate voltage range? -- somewhere. And more complete (if you search on 6L6GC, you don't get EL34, KT66, or 6CA7).

No, seriously, so you just remembered? The reason I ask is because you made it sound like categories of tubes were rated for wattages, but Groove Tubes has a way of categorizing the same tubes of the same type from the same factory along a 1-10 scale, from "soft" to "hard", where "soft" has an early breakup character (for those who live in the dirty channel) and "hard" has to be cranked before it breaks up (for those like me who live in the clean channel), which sounds to me analogous to what you were talking about with wattage ratings, so ... yeah. That's why I was asking. I'm just trying to figure out all my options that allow me to remain within parameters without ruling out something that might be that magical combination for me. I don't necessarily want to have to read a whole data sheet (and I suspect most guitarists who weren't interested in the guts of their amps would, neither), so I was trying to figure out if there was one or two sets of numbers that I should look at.

elvis -- yeah, I should probably be more methodical and record samples after a constant warmup time or something, but that sounds too much like work. But I can A/B the MDT and Archon and "more" or "less" dissimilar is a pretty audible difference. But they shouldn't be taken as science, I'm definitely not being scientific about this. I still appreciate your input -- your skepticism is well placed, when someone says they can hear a difference -- I've called people out on making the same kind of hyperbole between different versions of Stax headphones in audiophile circles. So for that, I should clarify: I do still find them different, significantly so.

I'll try to be more critical in my listening though. And see if I can find a way to record samples. But I honestly think the Archon sounds better now, slightly less dry.

 

[László]

No, seriously, so you just remembered? The reason I ask is because you made it sound like categories of tubes were rated for wattages, but Groove Tubes has a way of categorizing the same tubes of the same type from the same factory along a 1-10 scale, from "soft" to "hard", where "soft" has an early breakup character (for those who live in the dirty channel) and "hard" has to be cranked before it breaks up (for those like me who live in the clean channel), which sounds to me analogous to what you were talking about with wattage ratings, so ... yeah. That's why I was asking.

All tubes have a wattage specification! It's part of their original published design spec. I guess you'd call that a wattage rating? Look on the internet, I'll bet the original RCA 6L6GC spec is available.

But of course, no two tubes are exactly alike in performance, and there is substantial deviation from spec.

6L6GC's design specification includes wattage, and has since RCA originally spec'd them. The output power is rated at a given distortion level. The tubes can output more power on peaks.

Every 85-100 Watt amp I've owned (that's a bunch) contained four 6L6GC tubes, and I've had an awful lot of this type of amp for the past 50 years, so it's not hard to remember. Most of my 50 Watt amps had two of them (I've had mostly 6L6 50 Watt amps).

So this isn't much of a memory thing, after a while, you learn that a 6L6GC is spec'd at 30 Watts of output power, and in most amps, it's going to be operated around 22-25 Watts.

I've used the small bottle 6L6 WGB on several amps, as it's a direct replacement for the 6L6GC in a lot of amps, and you get an earlier breakup. So it's just a matter of fussing around with a lot of amps and remembering things.

I definitely don't memorize this stuff.

 

[Dusty Chalk]

You mean, like that ------------------------------------------------------------------------------------------^ ?

 

[Deleted member 5962]

I'm not surprised by your results so far. One amp was over spec and it cooled some when adjusted down. One was under and got warmer or sweeter when adjusted up. That is usually the effect of turning the bias up or down, at least to a point. Big adjustments start to change other things, as Elvis mentioned.

Thing is, EVERY amp is different in just what changes, just how much it changes, how much of an adjustment makes how much difference, etc., because it's all dependent on the whole circuit design and all the other factors. In fact, I'd say that it won't even be identical on the next set of tubes, no matter how closely rated the two sets are.

I used to be all about all this stuff. I jokingly once called it a hangover from my audiophile days. I quickly learned I spent more time tweaking and doing tone experiments than playing.

 

[elvis]

I believe that the GT ratings are largely about gain. More gainy tubes will clip earlier because they have a bigger output for a similar-size input vs a lower gain tube.

Increasing bias current also increases gain, and gets the operating point a little close to the saturation point of the tube and the output xformer. But these amps are class A/B, so their peak current is MUCH bigger than the idle current. If they were class A, then I would buy the idea that biasing hotter has a more significant effect, as the tubes always run max current in class A.

 

[Dusty Chalk]

Well I tried the dirty channel briefly last night, and it sounded horrible, so I still need to learn a few things about my amp (including dialing in tone), so I will remain skeptical about my own results until I can repeat them methodically.

PRS customer service got back to me. The long and short of it is, they recommend the factory tube set and bias; it is okay to experiment, but if you run it too hot for too long, you risk damage; they don't recommend more deviation than 30 +/- 5 mV; 6CA7 is also okay; KT77 up to 40 (but again, keep an eye on the temperature (I have an infrared thermometer, so I'll be taking some baseline measurements and keeping it handy)); KT66, 6L6GB, and 5881 are explicitly disrecommended.

 

[elvis]

I believe that 6L6GB and 5881 are the same tube (Archon 25 uses the 5881). Likely reasons for recommending against the tube types you mentioned would include much higher heater current and low plate voltage tolerance, among others.

 

[Dusty Chalk]

He gave good reasons, I just left them out. I probably should try to understand them, as I am curious, and have enough of an engineering aptitude to figure it out, but I'm more curious how to play guitar.

 

[Dusty Chalk]

Confound it, I'm confused. I found a page that said the Groove Tubes GT-6L6R (which I've bought several, trying to make a ~ matched quad) is the same as the 5881/6L6WGC aka the 6Π3C-E, and I'm beginning to think they're not compatible. Going to stick with my one pair of GT-6L6CHPs in 50 Watt mode (or perhaps the MDT, since it only takes a pair) until I receive a matched quad of proper 6L6GCs, and ordering them tonight.

Perhaps someone wants to try them in the Archon 25, see if they work better as 5881s?

EDIT: Apparently, I should stay away from Groove Tubes altogether. According to their website, their 6L6 tubes correspond to 5881. Click on "cross reference" if it doesn't automatically put you to the table that says this very clearly.

 

[bodia]

I'm sure you've inundated yourself with countless searches and websites. If you're still looking for info, check out this page:
http://www.audiotubes.com/

I was enlightened reading through his descriptions/explanations. As a rookie when it comes to any tubes that didn't come in the amp, I thought this was a great place to learn about what I was looking for. In addition Brent was very helpful on the phone with suggestions, and disqualifications on tubes I was interested in for an HXDA.