Merlin's morphing Pentode/Ultralinear/Triode EF86

[surfsup]

tubeswell, I have an amp design Im messing around with that is similar to your second schematic. I am curious how much loss you are getting across the bandaxall. In other words I figure if you get a gain of ~60 on a 1Vpp from the first parallel tube, what's going into the EF86. Also, what gain are you getting with the 100k/680k resistors on the EF86?

I am actually going to use a 6SJ7 instead as my second tube, as well as 6V6s for the power stage, but Im curious if you could share a bit more (if you ever measured and documented that stuff) I would appreciate it.

Ahhh...while I'm at it how's that tremolo?

[tubeswell]

tubeswell, I have an amp design Im messing around with that is similar to your second schematic. I am curious how much loss you are getting across the bandaxall. In other words I figure if you get a gain of ~60 on a 1Vpp from the first parallel tube, what's going into the EF86. Also, what gain are you getting with the 100k/680k resistors on the EF86?

I am actually going to use a 6SJ7 instead as my second tube, as well as 6V6s for the power stage, but Im curious if you could share a bit more (if you ever measured and documented that stuff) I would appreciate it.

Ahhh...while I'm at it how's that tremolo?

The gain of the EF86 stage in that amp is about 100 in pentode mode (and 30 - maybe more - in triode mode I'd say). The PP gigging amp is more than sufficiently loud enough for gigging in a band situation in a medium sized club wit a loud drummer and another geetarist. I have tested it so several times.

The 'Bone Ray' tone stack in the PP gigging amp is not very lossy at all, and is even better when driven by the parallel V1 stage. It has a clear mid scoop-hump shift. I have since changed to a James stack which provides more of a classic Fender scoop. However the Bone Ray stack was quite useful and I would use it again in an EL34 inspired amp

The tremolo is nice - like a 5G9

[brewdude]

tubeswell, I have an amp design Im messing around with that is similar to your second schematic. I am curious how much loss you are getting across the bandaxall. In other words I figure if you get a gain of ~60 on a 1Vpp from the first parallel tube, what's going into the EF86. Also, what gain are you getting with the 100k/680k resistors on the EF86?

I am actually going to use a 6SJ7 instead as my second tube, as well as 6V6s for the power stage, but Im curious if you could share a bit more (if you ever measured and documented that stuff) I would appreciate it.

Ahhh...while I'm at it how's that tremolo?

I have been working on an amp with a parallel 12ax7 input stage into a 5693 with a variable voltage screen grid, then a parallel 12au7 cathodyne PI feeding a pair of 6v6's.

Since I drew the attached schematic, I have tried (and removed) a bass control and upped the pentode's plate load to 100k. Currently it has no tone controls and needs an adjustment to the pentodes screen voltage control circuit. It is still a work in progress.

[Reeltarded]

Selmer Zodiac meets JTM-50 is what I am doing right now.

[gingertube]

Lars over at DIYAudio (posts as "Revintage") is a European Guitar Amp Designer Manufacturer.

http://www.revintage.se/gui.html

He has a patented scheme he calls TULP, no doubt for "Triode Ultralinear Pentode". he is quite protective of the design but has let out these facts:
1) he uses it as the input stage
2) he runs 1/2 a 6DJ8 and 1/2 a 12AX7

That leads me to take WAG that it is a cascode with the 1/2 6DJ8 on the bottom and the 1/2 12AX7 on the top with pot control to the top triode grid.
If you conceptualize the cascode as a pentode then the top triode grid is the "screen" connection.

This WAG is based upon the fact that the gain of a cascode is basically the gm of the lower tube x the RL on the top triode. That would be why he chose to use a 6DJ8, to get large gm for the bottom tube and hence maximize gain.

So there is an experiment for someone if you are feeling a bit bored.

Other guidance in looking at the tubes he uses - I know he uses the ECL86 (6GW8) as the Reveb Driver / Recovery. Also as most will know the 12DW7 is basically 1/2 a 12AX7 and 1/2 a 12AU7 in one envelope.

Cheers,
Ian

[dorrisant]

I'm starting to feel like I fit in... Great stuff here poeple, thanks for posting!

I am currently trying to stuff a similar circuit into a "breakfaster"...

Paralleled 12ax7 --> ef184 pentode --> SE el84. Kinda close to a Matchless Clubman pre, but it is evolving to something different. I had been looking at the morph control... I think it would help to get the nice cleans but also allow the ef184 to do its magic as well.

A little off topic but... What would be the best way to use a 6v6 and an el84 in the same SE circuit but never at the same time? Besides pulling one tube or another... I thought about using a multi-pole switch to switch the cathode to ground and make or break the heater circuit as well. Not sure... maybe make/break the plates instead, I know there is HT on the switch that way, I'm just thinking out loud here... Don't know if it would ever work but if someone has pulled it off, I would love to experiment in that direction.

Maybe I don't fit in now, lol!

Tony

PS: Maybe it should be called the Mighty Morphing Pentode Ranger...

[martin manning]

DPDT switch grounds one cathode or the other (6V6 or EL84), and opens the filament circuit on the unused tube. Separate cathode resistors to bias each as needed, 8k load works for both, ~300v on the plate.

[dorrisant]

DPDT switch grounds one cathode or the other (6V6 or EL84), and opens the filament circuit on the unused tube. Separate cathode resistors to bias each as needed, 8k load works for both, ~300v on the plate.

Sweet!!! Thanks Martin!

Tony

[briane]

Was your amp something similar to the Foxy 18?

that one was a straight marshall 1974 - I'built the foxy 18 - different beast. The 1974x in 12ax7 form is all clean, little crunch.

the ef86 front end can give you a lot of 'honk' - I have one amp where I ran an ef86 into a paralled 12ax7 to smooth it out and add thicker harmonics - worked like a dream for a pre-amp for a clean king amp.

[chief mushroom cloud]

Selmer Zodiac meets JTM-50 is what I am doing right now.

LIKE

[roseblood11]

I've done this on a couple of amps now where the EF86 was used for the input and then the 2nd stage.

[img724]http://farm9.staticflickr.com/8029/8031 ... bd9c_b.jpg[/img]

[img725]http://farm6.staticflickr.com/5159/6949 ... 6d30_b.jpg[/img]

As jelle says, it mostly works like a gain control. However at the pentode end of the pot rotation, the bandwidth is noticeably wider as well. And on the lower strings you tend to get interesting tonal shifts.

FWIW, I have also tried the other circuits in that chapter (just before the morph control) and they also produce noticeable tonal effect variation.

I also did a variation where the EF86 had a CF stage and where I took the morph control input from the CF cathode. This helps with the loading of the EF86.

Very old thread...

Do you have a bigger version of the second schematic?

[stephenl]

Lars over at DIYAudio (posts as "Revintage") is a European Guitar Amp Designer Manufacturer.

http://www.revintage.se/gui.html

He has a patented scheme he calls TULP, no doubt for "Triode Ultralinear Pentode". he is quite protective of the design but has let out these facts:
1) he uses it as the input stage
2) he runs 1/2 a 6DJ8 and 1/2 a 12AX7

That leads me to take WAG that it is a cascode with the 1/2 6DJ8 on the bottom and the 1/2 12AX7 on the top with pot control to the top triode grid.
If you conceptualize the cascode as a pentode then the top triode grid is the "screen" connection.

This WAG is based upon the fact that the gain of a cascode is basically the gm of the lower tube x the RL on the top triode. That would be why he chose to use a 6DJ8, to get large gm for the bottom tube and hence maximize gain.

So there is an experiment for someone if you are feeling a bit bored.

Other guidance in looking at the tubes he uses - I know he uses the ECL86 (6GW8) as the Reveb Driver / Recovery. Also as most will know the 12DW7 is basically 1/2 a 12AX7 and 1/2 a 12AU7 in one envelope.

Cheers,
Ian

I've played around with the cascode in a couple of amps.
I built a RT66-like amp and replaced the pentode input stage with a 6SN7 in self bias cascode...sounded nice but nothing special.
I tried a 12AV7 in self biased cascode as the input stage on a wreck express circuit - That thing was WICKED