Comments on my 5E3 build..... possible ultralinear?

[akimball442]

Ok, please don't flame me!!!!!


I built a fairly stock 5e3 from a kit, and purchased a nice tweed cabinet for it. Stock. Yeah, kind of. I increased the 250 Ohm cathode resistor on the output tubes to 10W, so I can run 6L6 tubes. Does that make it a Pro?

I'm also running a 5V4 Rectifier. B+ is a nice even 400V.

The preamp tubes are a 12AY7 and a 12AX7, stock.

The bypass caps have all been reduced to .022 from the original .1 spec, and the cathode bias on the preamp tube has been switched out from the stock 25uF/820R to the 'Marshallesque' .68uF/2700R.

Cabinet is stock 5E3 size and style, containing an EV-12L speaker which barely fits.

With the help of this forum, I have eliminated all but a very little bit of hum which is only present with the volume controls just about cranked. It's very tolerable.

The amp hates effects, but sounds fantastic. Plug a Tele straight into it and enjoy.

I actually did one gig in a small club with it, and I was quite pleased. Cranked up it has a nice crunch, back off the volume on the guitar, and it cleans up nicely. Once you get the controls set you never have to touch the amp. Simplicity at its best. A major deviation from my 5f6-A modified BF Bassman +212 cabinet for sure. Pleasant though, and so far reliable.

Enough bragging.

So the kit I purchased contained an output transformer with UL taps. It's rated at around 25 Watts, so I'm guessing it could handle the 6L6's in UL right? (I hope).

I'm more or less curious about the results, but I have a question regarding hooking it all up...... as in How???

Looking through various schematics, most of the 6L6 UL amps (at least by Fender) are fixed bias, and contain a 470R resistor from each screen tied to B+ after the first dropping resistor (or choke), with the CT on the output tranny tied straight to the first B+ node.

On the 5E3, the screens are tied directly together with each other, and to the node after the first dropping resistor.

In order to convert this to UL, can the screens be tied directly to the UL taps, or should the 470R resistors be added?

Thanks!

[Chuck H]

Ok, please don't flame me!!!!!

It depends on how you take it I suppose.

I increased the 250 Ohm cathode resistor on the output tubes to 10W, so I can run 6L6 tubes. Does that make it a Pro?

Nope.

I'm also running a 5V4 Rectifier. B+ is a nice even 400V.

The 5V4 may not hold up with 6L6's.

The bypass caps have all been reduced to .022 from the original .1 spec

I think you mean "coupling" caps. And reducing them is generally a good thing for this design if you OD the amp a lot.

and the cathode bias on the preamp tube has been switched out from the stock 25uF/820R to the 'Marshallesque' .68uF/2700R

Reducing the "bypass" cap values on the preamp is a common mod for the 5E3 and can help tighten the amp up under OD. With your higher preamp voltages (because your using a 5V4 rectifier) the bias shift may also be appropriate. I think that even with the 5V4 you would get closer to a correct bias with a 1k or 1.5k resistor on the 12ax7 preamp tube. 2700 seems pretty high for a 12ay7 in that circuit.

I It's rated at around 25 Watts, so I'm guessing it could handle the 6L6's in UL right? (I hope).

I'm more or less curious about the results, but I have a question regarding hooking it all up...... as in How???

You'll blow up your OT if you try to bias the 6L6's correctly. You'll need to run 6L6's cold in that amp or upgrade the OT. UL taps are typically connected to the power tube G2 (screens) omitting any seperate screen supply voltage or screen resistors.

Chuck

[akimball442]

It was late when I wrote this post.

I know it's not really a Pro, I was being silly- There's not a grand difference between the Deluxe and the Pro in terms of circuitry. The Pro got a choke, a bigger speaker, the 6L6's and a different OT.

I chose the 5V4 because I had it on hand, and it would be no great loss if it failed. The filament current was 2A, which was acceptable to my PT. I suppose a 5AR4 would be a better choice. But I think that might further increase the B+. It did survive being on for Four straight hours at just about full tilt, though. It's a brand I have not seen or heard of (maybe someone else has) Graylock?

I did mean coupling caps, not bypass caps between stages.

As for the cathode bypass on the first stage, I peeled the back off the amp, and I must have realized this at the time... I did not use a 2700 ohm resistor, I used 1200- roughly in the middle of the 1K to 1.5K you suggested. The cap is .68uF.

The output transformer does not seem to really even get warm after prolonged use of the amp in this configuration.

Thanks for the response on wiring it up UL, I'm more or less curious about it, I have not read any rave reviews about UL guitar amps. I figured since the wires were there bundled up for a rainy day it might be fun to experiment. (The weather has been lousy here.)

I'm also interested to see what happens to the tone of the amp with the change to UL. If the tone changes drastically, then it'll immediately be put back to the way it was. If it gets louder and cleaner with roughly the same tone, I might consider adding a switch to change the modes.

And if I do let the smoke out of the PT, it wasn't expensive, that just means it needed to be replaced anyway (with a good one)

[Chuck H]

UL or not isn't really the issue. The use of 6L6's in an amp designed for 6V6's can stress other parts IF the voltage is high enough to get more watts from the 6L6's than you were getting from the 6V6's AND you drive the 6L6's such that they will produce those higher watts. Plain and simple, if you set up the 6L6's to produce more watts than the OT can handle and drawing more current than the PT can handle then something has to give. Perhaps your trannies are up to the task.

As far as UL tone...? Not sure other than to say that this is a way to get an improved distortion figure from a tube. But don't confuse that with less clipping. Tubes make all manor of waveform distortions just operating. Hi fi designs use tubes UL or triode wired with more NFB and more careful OT design and matching to the circuit conditions to get rid of this distortion. A typical guitar amp is probably between 3% and 7% THD on the unclipped wave form. Some hi fi amps have a distortion figure of .01%. UL operation is one method used to reduce this distortion on the unclipped wave. It won't make your amp clip less. It will make your clean tone cleaner. And that's not always a good thing because guitar amps are used more as an instrument than for pure reproduction. In other words, guitar amps intentionally color the sound they are reproducing and not trying to give a mirror image. Reducing harmonics is generally considered a bad thing for guitar amps. Some harmonics lost will be 2nd order. These are the harmonics that seem to make tube amps special. I don't discriminate and like all harmonics be they odd or even.

OTOH some builders including Dr. Z have used UL wiring to good effect. But I do think that those amps are running at or near class A which doesn't cancel 2nd order harmonics like an AB1 circuit does. So maybe running UL on a class A amp would seem to have the effect of accentuating 2nd order harmonics because other harmonics are reduced so low that the 2nd harmonic stands out. Not sure and I haven't experimented with it. But I don't think UL is something that should be added to a guitar amp without other design considerations. But if you have UL taps it should be easy enough to try it and see if you like it anyway. It won't make the amp more powerful or improve headroom though.

Chuck

[pdf64]

My guess is that a UL arrangement would be more powerful, similar to subbing a choke for the 4k7 dropper resistor to the screens, because the screens would have a much more solid, higher voltage supply.
6L6 may run too hot because of this and overload the power supply. If you are going to try UL, check it out with 6V6 first.
UL often don't sound good when overdriven as too much ripple gets to the screens. That could be mitigated by using a pi type CLC B+ filter between rectifier and OT. That would need a choke which could cope with the full B+ current.

[Chuck H]

+++ On UL hook up being more watts due to the increase in screen voltage. I failed to consider that above. But just to be clear it's the higher screen voltage that would increase headroom and not the UL operation. All things being equal UL operation isn't somehow more powerful than a more typical choke arrangement. I don't think I'd try it using 6L6's with an OT designed for 6V6's. The BAC (big a$$ choke) would be both expensive and clumbsy large. I've never played a Dr. Z. They basically live overdriven. Do they have a lot ripple on the screens, and I would guess ghost notes?

This still comes back to what I was saying above though. UL isn't the kind of thing you can add to a vintage amp as a simple mod. It should be designed into the amp.

[Rhodesplyr]

My guess is that a UL arrangement would be more powerful, similar to subbing a choke for the 4k7 dropper resistor to the screens, because the screens would have a much more solid, higher voltage supply.

I'm not sure it works this way. I've read an analysis of UL vs. pentode output stages that showed why you got maximum power output running beam tetrodes in pentode mode. Check out the schematics of some Bogen amps rated 100W from two 6550s. They run them in pentode mode--keeping screen voltages at typical levels, but increasing plate voltage levels to > 600V. They use fairly large transformers.

AFAIK, the point of ultralinear output stages is find a compromise between the lower sensitivity of triodes to variations in load (speaker) impedance with the higher power output of pentodes. This works well for Hi-Fi where you want flat frequency response, but for guitar? There are so few ultralinear guitar amps out there that it's hard for me to say.

Has anyone tried to use a Dynaco MkIII or MkIV as a guitar amp?

[Rhodesplyr]

I hit reply before I remembered that a few Sunn amps from the late 1960s were Dynaco MkIIIs with additional input gain stages and tone controls.

Look up the Sunn Model 60, 100S, and 200S.

I've never seen one. Anyone played through one?

[Rhodesplyr]

And the Sunn Solarus is a Dynaco MkIV with additional circuits for guitars/basses.

[Chuck H]

That's what I've read. UL is a way to get better "performance" from a tube. In this case performance would be the finite distortion figure. You won't sacrifice power to a certain % of the OT wind for the screen taps. I think the magic number is between 20% and 40% depending on other operating conditions. Past that your output power will drop closer to true triode operation. It may well be that UL offers the best distortion figure per watt but it doesn't make more watts than any other method and certainly less than some (like the bogen you sited above) due to limitations of most power tube screens. The lower damping and phase shifts also improve finite performance as it applies to high end audio for reproduction of a program. I never considered it an appropriate avenue to explore for guitar amps but it has been done by a few builders. I've never played a Super Twin or the big Sunn's or a DR. Z. All respectable guitar amps and in different viens too.

[akimball442]

Well, it seems to be holding up, but wrong is wrong...

I'd like to have it long-term reliable, so what would be the recommended output transformer to keep the 6L6's?

I like them better than the 6V6's.

[Rhodesplyr]

I'd like to have it long-term reliable, so what would be the recommended output transformer to keep the 6L6's?

According to the datasheets, you're not likely to get more than 25W from a pair of cathode-biased 6L6s. What I'd check, if possible, would be the maximum DC bias current of the transformer. Make sure you're not exceeding that. I can't be sure, of course, but I'd suspect it would be adequate. Why? You aren't going to get 25W from a pair of 6V6s.

Of course, transformer ratings can be tricky to interpret. Triad rated the output transformer used in pretty much all Leslie amps at 24 Watts, 30-6,000 Hz; yet, Leslie built 40W amps around them. I think the key is that you're not going to sit there running 40W through them continuously. You will, however, have a constant DC bias current through them.

I'd be more concerned about the power transformer. I worked on a friend's Fender Princeton Reverb in which the previous owner had installed Sovtek 5881s in place of 6V6s. The extra demand on the power transformer burned it up.

[pdf64]

' I've read an analysis of UL vs. pentode output stages that showed why you got maximum power output running beam tetrodes in pentode mode.'

I'm not sure that's directly applicable to the scenario here, as there's no question of getting the max theoretical power output from the power tubes. Rather, as I see it, it's how the tubes will react to their screens increasing 100V or so. Given that the load remains the same, I would have thought that the power output pretty much has to increase (going from stock to UL).

'I'd be more concerned about the power transformer. I worked on a friend's Fender Princeton Reverb in which the previous owner had installed Sovtek 5881s in place of 6V6s. The extra demand on the power transformer burned it up.'

That's what I was thinking too - 6L6s in UL will be able to conduct much harder than the stock arrangement that the PT may have been specified for. And then there's the extra heater load also.

[Rhodesplyr]

I was recently reading up on pentodes in preparation for building an amp that uses a 6JS7 as the input tube. If you connect the 6SJ7 as a pentode, you can get a gain of >100. If, on the other hand, you connect the screen directly to the plate, wiring it as a triode, its mu drops to 19, equivalent to a 6C4. That's using the same power supply voltage.

I found that pentode behavior is really not that intuitive, and I still don't claim to understand it all that well. Check out this site:

The Valve Wizard -Small Signal Pentode

One thing the curves show is that pentode input sensitivity actually *increases* with low screen voltages. So, returning to the output tube stage, the closer you get to triode mode, the less gain I think you'd have. Thus, you'd need a bigger input signal to drive the output stage to full power. I've read a few descriptions of Single-Ended Triode power amp circuits, and they usually seem to need fairly high amplitude drive signals.

Please correct me if I'm wrong!