Tweedle Dee - another latecomer (MODS: P12Q +Buzz +trim adj)

[bluesfendermanblues]

Even though Mr. Dumble use Celestion (Rocket or Greenback) in his Tweedle Dee, I wanted to try a reissue (italy) Jensen P12Q in my amp instead of the Cel 12L80 i had put in there. Celestions (with membran dope) sounds fine, but maybe a little to polite for a tweed. BTW thank you Charlie Wilson for posting the pics of a Tweedle Dee.

The Jensen has a little more compression (due to the alnico), which suits the amp, unfortunately also a little lower sensitivity (95dB vs. the Cel's 99) - but the increased top end of the Jensen, made me realise that the amp has a strange octave-like buzz riding on top of the signal when pushed. Not good!

My first instinct was oscillation due to lead dress, but I did a little forum searching and stumbled on:

1) Reference by user "Rock_Mumbles" (https://tubeamparchive.com/viewtopic.ph ... 70k#274956)

and

2) Merlin Blemcowe's http://www.valvewizard.co.uk/cathodyne.html:
"The main problem with the cathodyne occurs when it is itself overdriven. When grid current flows into the cathodyne and into Rk it causes the voltage across Rk to be artificially 'jacked up'. By this point the valve can't pull the anode down by an equal amount, so this voltage boost appears at the anode too, creating a sort of full-wave rectified or frequency doubled signal at the anode. This is shown in the lower photograph. It is usually this effect which causes the ugly 'blatting' overdrive tones sometimes heard in amps using this kind of phase inverter. Fortunately, the cure is simple. Just add a large grid stopper to the cathodyne, to limit the grid current. A value of 100k to 1Meg is usually necessary."

suggesting to use a 470k-1M grid resistor on the input of V2b. In order to not overdrive the cathodyne PI.

I tried a 1M in and that did that did the trick. Success!

Since this cleaned up the buzzy octave distortion, I now found that adjusting the trimmer for increased overtones made sense (using the classic procedure of gently brushing the strings at max treble setting - for this particular my Strat - middle PU w no tone control engaged). I found that max bloom was at 3 o'clock on the trimmer, producing an almost blooming tone quality - before the 1M mod, I couldn't hear any difference between the various trimmer settings.

Mods yet to try:
a) - Cathode resistor is 270 ohm (22.9 volts) and not 250 ohm (which might make the JJ 6v6 to run a little hotter.
b) - Increased Grids on the 6v6. Pt. I have 1.5K, but Merlin suggest 47k-100k. - why not use a little Miller effect to remove to dampen the frequencies above 8kHz altogether? (Like this Marshall Studio 15 with 6v6's http://www.drtube.com/schematics/marshall/4001m88.gif)

[Smokebreak]

Very cool, thanks for the report. The last cathodyne I did I used a 1M pot wired as variable resistor to tweak the grid stopper value, as you definitely start to lose a little high end after 500K or so. I think I had 56K stoppers on the 6V6s, as well.

[Structo]

Claus, are you saying the 1M resistor goes on the grid of the PI?

I see the original Tweedle has a straight shot to the grid there after the .022uf coupling cap.

I din't notice much variance adjusting the trimmer myself.

[M Fowler]

Tom,

That was the wiring error Steve-67plexi was talking about in one of the early Tweedle Dee threads makes the amp more open not so 5E3 sounding.

See the attached layout red circle and red line connecting .1 cap to 1m

Mark

[rp]

are you saying the 1M resistor goes on the grid of the PI?

https://robrobinette.com/5e3_Modificati ... ifications

[bluesfendermanblues]

are you saying the 1M resistor goes on the grid of the PI?

https://robrobinette.com/5e3_Modificati ... ifications

Exactly.

I'm talking about adding a 1M resistor connecting the board and pin 7 on V2.

[bluesfendermanblues]

Follow-Up.....After playing the amp for a few days, I could still hear a buzzy distortion character when upping the volume above 3 (goes to 12). So I took out the above mentioned added 1M resistor.

Mods in order to reduce paracitic oscillation:
a) After reading a lot of posts on this forum and on other forums, I decided to a Gerald Weber (the big mister No No on this forum) mod (he has labeled it Flaw #1) - moving the middle (0.022 uf) cap to the V2 socket and ditching 4 inches of grid wire. Eawsy mod that to my ears improved the sound.

b) Another very interesting layout can be found on el34world.com: http://el34world.com/Hoffman/files/Hoffman_5E3.pdf

Hoffman has done a lot of experimentation and his 3rd layout moves the two 0.1uf from the PI down on the board, as far away from the volume pots as possible, (which is also along the lines of Gerald Weber Flaw #2). So I decided to take inspiration from both Hoffman and Gerry Weber and move both the 0.1uf caps.

In order the get enough real estate on the board, I decided to move the 3 resistors; 4k7, 1M, 3.3M and the 0.002 cap to the V2 socket.
And hereafter I moved the 0.1uf caps. The grid wires to the output tubes cut be trimmed down to 1/2-1/3 of the original length and they are no longer near the volume pots.

c) Another small mod done was moving the eyelets for the (0.022uf) anode caps from the V1 plates. The purpose of this operation is getting the grid wires from the caps away from the input jacks.

d) The last mod was using shelded wire on the two input wires to V1, pin 2 and 7. and from the V1 anode caps to the volume pot inputs.

I only changed the board layout and have stayed with the Dumble Tweedle Dee values.

Other small mods -
e) Since tube are expensive, I decided to put a couple (470 ohms) on the 6V6 screens. No difference in sound, but it saved the tubes a little. After all the are seeing 400 vdc.

f) For variation I upped the V1a anode cap from 0.022uf to 0.047uf. I'd be hard pressed to actually be able to hear any difference between the two channels.

g) In order to have tone control on both channels, I changed the tone pot to a stereo pot.

h) I installed Dana's VVR 5 years ago when I build the amp. It works great on small cathodebiased amps. I usually run the put a noon for around half-the anode voltage. I have scaled the whole amp and have added an extra PSU cap, but not the 0.1 on the input jacks. Never had any issues with scratchy pots.

The amp is totally free from buzzy paracitic oscillation artifacts. I can rund the volume pots at noon or fully open for that matter, and it sounds like 'clean' overdrive. The amp is very responsive with lots of nice compression and cleans easy when you reduce the guitar volume.

Back in the 90's I had a VCR tape with Jim Wieder'"How to get that classic Fender sound". Jim is playing a 53' Fender Tweed Deluxe. I believe my own amp is actually able to reproduce those sounds or should I say, I was reminded of those sounds when I play the amp now. Nice.

I still have the Jensen P12Q - its a little bass light in comparison to the cel 12L80 I had in there, but its a rightious blues sound. [/u]

All in all - I difficult for me to understand why cloner and boutique builder are not using the Hoffman layout in stead of the falulty old Fender 5e3 layout, that was not intended for overdriving the amp. The Hoffman layout is free from oscillation.

Lots of builders of the Tweedle Dee must have experienced the same oscillation isssues as I had before the above alterations. Beats me why Mr. Dumble stayed with the original layout - on his ODS layout you usually - if done corectly - don't get oscilaltions.

[Charlie Wilson]

Interesting Bluesfenderman. I would have thought the .002/3m3 LNFB would help prevent oscillation issues.
CW

[bluesfendermanblues]

Interesting Bluesfenderman. I would have thought the .002/3m3 LNFB would help prevent oscillation issues.
CW

Maybe, but if so, it wasn't preventing oscillation in my amp.

In the ODS's the LNFB is more of a compressor/bass cut. which IME is best with humbuckers.

Anywho the above measures
- shielded cables for in/out of V1
- moving eyelets from input jacks
- moving the 0.1uf closer to the 6V6 altogether

is the recipe or simply use the Hoffman board with Dumble mods. BTW I am not affiliated with Hoffman Amplifiers.

[ToneMerc]

Lots of builders of the Tweedle Dee must have experienced the same oscillation isssues as I had before the above alterations. Beats me why Mr. Dumble stayed with the original layout - on his ODS layout you usually - if done corectly - don't get oscilaltions.

I wonder if using the 5C3 input(horizontal) jack scheme makes a difference? I cloned a 5CE chassis and don't recall experiencing those issues.

TM

[bluesfendermanblues]

I wonder if using the 5C3 input(horizontal) jack scheme makes a difference? I cloned a 5CE chassis and don't recall experiencing those
TM

Horisontal Jacks are probably better, but using shielded cable runs and moving the 0.1 uf's made a Big difference in regards to solving the oscilaltion issues.

[drew]

Anywho the above measures
- shielded cables for in/out of V1
- moving eyelets from input jacks
- moving the 0.1uf closer to the 6V6 altogether

is the recipe or simply use the Hoffman board with Dumble mods. BTW I am not affiliated with Hoffman Amplifiers.

When the Tweedle Dee was first revealed here, I tried to draw a Hoffman-style layout for it. I worked inwards from both ends, but couldn't figure out how to link the two halves in the middle and incorporate the trimmer in a way that didn't look like a giant kludge.

[passfan]

I have to say I'm liking the growing list of mods and such for this thing. I have bent a larger chassis with the plans of including reverb and a few extras for this amp. To that end I've built a 2-12 combo for my chassis and dropped a couple Jensen C12PS drivers into it......Wow , what a difference , this thing turned into a raunchy sounding blues amp. I let my friend play it for a couple of days and he can't find a guitar that sounds bad through it. I can't wait to finish the build with the reverb and some other things.