I got yo big ol' Ultra Linear right here.

[gingertube]

The best sounding twin reverb I ever heard was UL. The key was trashing the feedback and cathode bias portion, returning to straight fixed bias. Besides the reverb sounds best delivered to the other input to the phase inverter. It sounds good even at lower volume but is certainly overkill for small gigs, but the owner does mainly big stage and outdoor gigs. Sorry Reel, I guess we don't all share the same tonal ideals, I just can't believe you wouldn't love the tone also. How do you feel about the new UL Tone King amps?

Use of the UL Tap drops output impedance by a factor of about 3 over Pentode connection. That means that (in general) you do need to ditch any global feedback loop - you just don't need it with UL connected output tubes. Size your screen resistors (from g2 to UL tap) to suit the level of compression you want, don't be afraid to really try some bigger values. 1K is minimum for EL34 but try 2K2, 4K7 etc. you just may like it.

Cheers,
Ian

[TUBEDUDE]

Matt, I really dig the SRV Lenny tone. When you run the twin with 2 tubes, are you changing taps for the 15, or is a mismatch part of the tone recipe?
I've always wanted to play with a set of EL-37's, but when they can be found the cost curbs my enthusiasm.

[matt h]

Tubedude, I'll never use el37s, but been looking into them because David Root mentioned them. I'd heard of them before, but really didn't know squat about them. I'm poor, so they're, uh, they'll just be something that stays an academic discussion for me! haha However, it's worth noting that they were considered "obsolete" with the development of the el34. I don't think it's worth discounting that.

As for the "mismatch"- you'll notice I specified an 8ohm 15" speaker. Yah? So a stock twin runs a quad of 6L6's at 2k into a 4ohm load (two 8 ohm speakers in parallel). So, if you pull a pair of 6L6's and continue to run the 4ohm load on the output, there's a mismatch. If, however, you do what I do, and run an 8ohm load on the secondary *while* pulling that pair? It actually un-mismatches, you've now got a 4k:8ohm (same ratio as the 2k:4ohm before).

One thing I try to get everyone with a twin who gripes about the volume to do, but no one ever does... re-wire your speakers! Run each speaker into its own jack on the back of the amp. That way, when you pull a pair of output tubes, you can just disconnect one of the two speakers. No switches, no BS, and even non-tech minded players can typically remember "one pair of tubes to one speaker." without thinking about "zomg! MATH! AHHH" factor.

Fender sorta "undersized" their OTs for cost... well, shazzam- you've just effectively oversized it. Headroom galore.

[gingertube]

matt h,

The poor mans EL37 is an 807, which is to say that the EL37 is an 807 with the anode top cap redirected to one of the base socket pins.

807 are cheap - recently bought myself a box of 12 Amperex 807 for $60 Australian (about $70 US these days). They are intended for a guitar amp, probably throw a quad of them at a 2nd hand Marshall JCM900 100W Output Tranny I have on the shelf.

The Standell used 807 at one time - not sure of any other Git. Amp. that did.

The "definitive" 807 Application Report is here:
http://www.retrovox.com.au/STC807.pdf

On the UL tap percentage - 20% is for max power, 43% is for minimum distortion.

Cheers,
Ian

[matt h]

Gingertube, hey- thanks for reminding me of the 807. I don't mean to be contrary-- but how are they EL37s? I ask because an EL37 is, at least per the data sheets, an honest pentode. 807s, to the best of my knowledge, are beam tetrodes like 6L6's.

I know they're great sounding tubes but was under the impression they were *really* more like 6L6's in a funky anodecapped package.

[gingertube]

Hmm! - My bad, you are quite correct, the EL37 is a true pentode, made "obsolete" by the introduction of the EL34.
I was lead astray by my equivalents book - It seems that "some animals are more equal than others."
Cheers,
Ian

[matt h]

ahhh "equivalents" hahah making the world simultaneously, paradoxically, and counter-intuitively more *and* less complicated!

I wonder if there's any truth to your observation, though. Not that the 807 *is* an EL37, but that they might be more likely to be in the same sonic area within the 6L6 variant/replacement/equivalent ballpark. I guess I can sort of imagine some scenario where by lead dress alteration and internal tube construction, that certain inter-electrode capacitances shift, transconductance gets nudged a lil here or there, and suddenly the weirdo plate-capped beam power tetrode does end up sounding a bit more like the pentode. Certainly not a ruling, but a point to consider- even if only hypothetically.

I guess that ends up being analogous to this UL discussion. UL ends up being a form of NFB. Ok. Amps often use NFB. Ok.

"There's more than one way to skin a cat... just remember, whatever the method: you can only skin it once before ending up with an unrecognizable pile of distastefulness."

[pdf64]

Fender sorta "undersized" their OTs for cost

The only Fender guitar amps with seemingly undersized OTs that I can think of are the Tremolux, Bandmaster etc type.
Most of the other amps seem to have suitably sized OTs for the application, ie appear to put out full clean power at low E / 80Hz.
Not sure if the same can be said of the various Bassman types at 40Hz though?
As a comparison, models from the Hammond 1600 series of an equivalent power rating don't seem to be any beefier, whilst being rated down to 30Hz.
Pete

[pdf64]

Use of the UL Tap drops output impedance by a factor of about 3 over Pentode connection. That means that (in general) you do need to ditch any global feedback loop - you just don't need it with UL connected output tubes

Does the (local?) NFB from UL operation reduce crossover etc distortion, in a similar manner to which an equivalent degree of global NFB would with a more regular arrangement?
Does a UL arrangement make any impact on the characteristics (eg output impedance) when the power amp is overdriven?
Pete

[gingertube]

Pete,
The short answer to that is I don't know.

Ultralinear is a "local" feedback between the output tubes and the output transformer only.

It is still not fully understood as the reduction in harmonic distortion and intermodulation distortion is a little larger than a purely mathematic treatment using feedback theory indicates it should be.

Feedback effects (on distortion, output impedance etc) rely on a gain reduction. That is the "open loop" gain is larger than the gain when "negative" feedback is applied. When you saturate either the output tubes or the output transformer you no longer have any gain (increasing the input further gives no additional output) and ALL of the effects of the negative feedback will disappear. The distortion will go back up (although in saturation you'll never notice) and the output impedance will go back up which you might notice as a lossening when overdriven.

More broadly, ANY sort of feedback system ceases to function when overdrive/saturation in the forward path reduces the gain to zero.

With ordinary global feedback to the input of the power amp, the gain reduction from the fedback stops happening and accordingly drive to the output tubes is increased, this tends to make the overdrive quite hard sounding and on an oscilloscope you will see the waveform "Square Up" as saturation becomes more abrupt and harder.

With just UL and no global feedback I would expect overdrive to not be as hard or agressive - BUT this is a guess, an "educated" guess but not based upon experience. If this is the case I would expect less of the objectionable high order odd harmonics to be produced.

What this does suggest to me is that if using UL in a git amp I would err on the side of a larger output tranny, one that would be unlikely to be driven to saturation before the output tubes saturate.

I put switchable UL/Pentode Mode into the 6V6 PP Guitar Amp I built for myself (6SL7 Concertina driver /splitter and 6SL7 version of Kevin O'Connors London Power Standard Preamp). With that Amp I find I never use it it in the UL mode. Having said that, the power amp also has some global feedback and I think the UL would be beter with that removed. The Output Tranny I used is a low cost Chinese HIFI Output Tranny designed for Ultralinear Mode EL84 (rated at 10W for HiFI but being pushed to 14W for guitar).

Maybe someone else has more practical experience and can offer other advice.

Cheers,
Ian

[renshen1957]

Me thinks ol' Suz is a spambot.

Hi Structo,

Maybe a Fembot?

Best regards,

Steve

[renshen1957]

Hmm! - My bad, you are quite correct, the EL37 is a true pentode, made "obsolete" by the introduction of the EL34.
I was lead astray by my equivalents book - It seems that "some animals are more equal than others."
Cheers,
Ian

Hey Ian,


As I recall (sleep deprivation affects the memory and spelling)

The 6L6 is a descendant of the "Harries Valve" developed by British engineer J. Owen Harries and marketed by the Hivac Co. Ltd. in 1935. It was further developed by MOV. MOV's engineers did not feel the tetrode could be successfully mass-produced, they licensed the design to RCA.

The 807 is a derivative 6L6 with an Anode/Plate cap. In push-pull pairs in class AB1 or AB2 giving up to 120 watts of usable power. The 807 is not a good choice of tube for typical Ultra-Linear amplifier circuits, as the maximum screen dissipation rating is easily exceeded when the screen is operated at or near plate voltage.

The 6L6 and later the 807was originally promoted by RCA for PA applications, but had some life as a radio tube.

The EL37 is a Power Pentode equivalent for a beam tetrode 6L6, as you wrote, developed after the 6L6

Best regards,

Steve

[renshen1957]

Pete,
The short answer to that is I don't know.

Ultralinear is a "local" feedback between the output tubes and the output transformer only.

It is still not fully understood as the reduction in harmonic distortion and intermodulation distortion is a little larger than a purely mathematic treatment using feedback theory indicates it should be.

Feedback effects (on distortion, output impedance etc) rely on a gain reduction. That is the "open loop" gain is larger than the gain when "negative" feedback is applied. When you saturate either the output tubes or the output transformer you no longer have any gain (increasing the input further gives no additional output) and ALL of the effects of the negative feedback will disappear. The distortion will go back up (although in saturation you'll never notice) and the output impedance will go back up which you might notice as a lossening when overdriven.

More broadly, ANY sort of feedback system ceases to function when overdrive/saturation in the forward path reduces the gain to zero.

With ordinary global feedback to the input of the power amp, the gain reduction from the fedback stops happening and accordingly drive to the output tubes is increased, this tends to make the overdrive quite hard sounding and on an oscilloscope you will see the waveform "Square Up" as saturation becomes more abrupt and harder.

With just UL and no global feedback I would expect overdrive to not be as hard or agressive - BUT this is a guess, an "educated" guess but not based upon experience. If this is the case I would expect less of the objectionable high order odd harmonics to be produced.

What this does suggest to me is that if using UL in a git amp I would err on the side of a larger output tranny, one that would be unlikely to be driven to saturation before the output tubes saturate.

I put switchable UL/Pentode Mode into the 6V6 PP Guitar Amp I built for myself (6SL7 Concertina driver /splitter and 6SL7 version of Kevin O'Connors London Power Standard Preamp). With that Amp I find I never use it it in the UL mode. Having said that, the power amp also has some global feedback and I think the UL would be beter with that removed. The Output Tranny I used is a low cost Chinese HIFI Output Tranny designed for Ultralinear Mode EL84 (rated at 10W for HiFI but being pushed to 14W for guitar).

Maybe someone else has more practical experience and can offer other advice.

Cheers,
Ian

Hi Ian,

I don't have more practical experience. Some observations.

Gar Gillies recommended in his book 2004 to convert UL Guitar amps to a more conventional configuration. Dr Z in an interview discussed that the KT tubes respond better UL connection; the KT66 worked well enough and the KT88 especially worked well.

The original Williamson amp wasn't a UL amp; this was a later addition. Some of the Hi-Fi forums posts more recently state that UL connection has too high a distortion rating to be ideal and that other tube connections have lower distortion. Go figure?

The Sunn Amps (Dynakit circuits) used UL transformers, and were Ultra Clean. Jimi Hendrix received an artist endorsement deal from Sunn amps which he jettisoned the solidified a five-year contract after 14 months). Jimi didn't leave Sunn for being too clean. These amps were too noisy at his usual double digit settings for his liking.

My two pence's worth.

Steve

[Leo_Gnardo]

The Sunn Amps (Dynakit circuits) used UL transformers, and were Ultra Clean. Jimi Hendrix received an artist endorsement deal from Sunn amps which he jettisoned the solidified a five-year contract after 14 months). Jimi didn't leave Sunn for being too clean. These amps were too noisy at his usual double digit settings for his liking.

Story I got was Jimi croaked it before taking delivery of his Sunn rig. It wound up in the (capable!) hands of Felix Pappalardi who honked his bass thru it with Mountain. Maybe it's all boo-shee, I dunno. Anyway yes Sunn outputs were Dynaco. What's amazing is they used the MkIII OT for their 120W amps, just doubled up the output tubes, and halved the output impedance ratings. Worked on a Sunn 120W PA recently and it delivered as advertised. The MkIII OT was housed in a big box but had the same part @ as the kit & factory hi fi amp.

Yes without some mods the Sunns can be noisy hum buzz beasts but it doesn't take much to fix. They foolishly ran their B+ bundled with other wiring into the preamp department. Run a new B+ wire along another route and hum/buzz are cut to acceptable levels. After all these years, the usual suspects must also be addressed starting with filter caps.

Suzannes recommended shoes caused one of my big toes to suffer - nail blackened then fell off. True story. Got rid of 'em then recovered 100%. Take that Soooooozie Creamcheese! Hey I know another Suzie that beats the heck out of all the rest but I'll keep that to myself for now. Hint - she's from Perth. Yeh that one, on the Indian Ocean.

[LeftyStrat]

Actually you can see Sunn amps in a photo down a bit on this page:

http://psychedelic-rocknroll.blogspot.c ... -1966.html

He took delivery in 68. Around the time this clip was recorded:

http://www.youtube.com/watch?v=saLnLi69jDA

Sounds like a Sunn to me. And to give credit where credit is due, it was someone on here that pointed out the connection. Can't find the thread now.