Full tone stack off of an EF86?

[dehughes]

Hey all,

So, why is it that I don't see that many full tone stacks (Treble, Middle, Bass, etc...) coming off of an EF86? I mean, Vox, Matchless, et al...use a tone knob almost exclusively. Why is that? About the only exception I can think of is the Dr. Z KT45/Rt.66....

I'm asking in part because my "Normal channel Brown Bassman" amp idea is (possibly) turning into an EF86 creation similar to maybe a Dr. Z KT45/Rt.66, but non-ultra-linear, and with 5881s, and with a SS recto...

[jaysg]

The usual tone stack you're thinking off, is too low an impedance for that tube. This is to say that if you connected it, you would wonder where all the treble went. I've not heard of anyone attaching a follower (which should help). The Traynor circuit DrZ employs, uses 1Ms iirc.

[Alex R]

The Leak Varislope preamp uses 2 x EF86 per channel. A hi-fi preamp, it uses switchable NFB and inline filters. I'm working on one at the moment, is why I thought of posting it. Guitar content, zero probably.

[dartanion]

Ideally you would want a CF after the EF86 as James mentions, as this lowers the output impedance to where common tmb stacks work well. A good option is to use a pentode/ triode tube so only one socket is used and you don't give up a triode from somewhere else.

[dehughes]

The Leak Varislope preamp uses 2 x EF86 per channel. A hi-fi preamp, it uses switchable NFB and inline filters. I'm working on one at the moment, is why I thought of posting it. Guitar content, zero probably.

Very cool. Do you have a clearer schematic? I'd like to understand what I'm seeing...

[dehughes]

So, Jay and Dart....if I understand you correctly:

1) An EF86 needs to see a higher impedance or else one will notice a lack of treble, for low impedance would bleed treble off to ground, yes?

2) Would a higher impedance after the EF86 take the form of higher resistance pots (1M, as mentioned) for Treble, Bass, Mid, etc.., and/or a cathode follower (as mentioned) in the form of, say, a 5F6-A Bassman tonestack, though coming right off the EF86 circuit of, say, an AC15?

I'm still mentally building an understanding of impedance in relationship to various tubes, and the effect that has on tone, etc..

[dartanion]

In a typical tmb tonestack, there is quite a lot of gain loss, add on the impedance mismatch and you will likely have too little gain.

[collinsamps]

Not sure of the tube compliment of the channel in question, but you can always use a full tonestack just in front of the EF86 and after a single side or both sides of a 12ax7. After the tone stack and before the EF86 is also a good place for a master or reverb insertion. Take a look at webers 6S100 and also his ef86 add on with the rotary cap selector switch. Thats a great way to feed a PI or whatever would be next.

[dehughes]

Interesting. Right now, I have only two spots for preamp tubes, so I'm thinking either some iteration of EF86 into PI, or 12AX7 with TMB in between the triodes a la Normal Channel Bassman. I coooould cut out another hole, but the chassis is so nicely setup as-is...

Part of me wants to live dangerously and try something out of the proverbial box, but then another side of me wants to go Normal Bassman...especially as I just played a decent Blackface Bassman head today, and really liked how "right" the Normal channel felt.

[Merlinb]

You can run a TMB off an EF86. However, for reasons mentioned above, I wouldn't use the 'traditional' TS values. Bascially you'll want to redesign the stack to have higher input impedance. Probably the easiest way to do that is to increase the middle pot to at least 47k, and the slope resistor to, say, 150k. Then use Duncan's TSC to model it and find suitable cap values.
Yeah it'll still load down the EF86 a bit, but you probably won't be dissapointed.
And if you are dissapointed, you coudl always tack in a MOSFET follower to buffer it, rather than undo all the hard work.

[Andy Le Blanc]

The high impedance means that you can use a much smaller coupling cap
and still have the same response as a larger one, say a .001 instead of a .1.

The simple vol. and ton. setup works nice with a high impedance. I've tried
that with a presence and resonance controls on the feedback loop around a
long tail cathode coupled inverter, call them vol., tone, treb., bass. Makes
for a responsive and useful amp. With tube electronics you should respect
impedance issue's around the device, not everything has to end up sounding
like a fender.

[dehughes]

You can run a TMB off an EF86. However, for reasons mentioned above, I wouldn't use the 'traditional' TS values. Bascially you'll want to redesign the stack to have higher input impedance. Probably the easiest way to do that is to increase the middle pot to at least 47k, and the slope resistor to, say, 150k. Then use Duncan's TSC to model it and find suitable cap values.
Yeah it'll still load down the EF86 a bit, but you probably won't be dissapointed.
And if you are dissapointed, you coudl always tack in a MOSFET follower to buffer it, rather than undo all the hard work.

Interesting...this is what the Rt. 66 does, yes?

So, I could still have a "mid" pot if I choose....it would just need a "floor" of about 50k..so a 50k pot on top of a 50k resistor, maybe?

[dehughes]

The high impedance means that you can use a much smaller coupling cap
and still have the same response as a larger one, say a .001 instead of a .1.

The simple vol. and ton. setup works nice with a high impedance. I've tried
that with a presence and resonance controls on the feedback loop around a
long tail cathode coupled inverter, call them vol., tone, treb., bass. Makes
for a responsive and useful amp. With tube electronics you should respect
impedance issue's around the device, not everything has to end up sounding
like a fender.

Okay. So if I need to design a TS with a high impedance, that's just a function of the slope resistor and the plate cap off the tube? I suppose I'm not getting what attribute of the design makes a TS high impedance...

As well, yes, everything shouldn't end up sounding like a Fender, but sometimes a Fender just sounds "right". However, you're saying you prefer a tone pot on the EF86 and then messing with feedback/presence, etc..? I like that idea... As for pot "names", I was contemplating eschewing all traditional names....call a pot whatever I want to.

What does a "resonance" control look like?

[briane]

You can run a TMB off an EF86. However, for reasons mentioned above, I wouldn't use the 'traditional' TS values.

glad someone said it. You can do it, but your going to be using different value's of components. You need to run the ef86 in pentode mode, as triode mode has a very high insertion loss, and generally lack of gain to drive the TMB. I tried to run in triode mode, and while the math seems to support this, the reality does not.

I built my version of the SSS based on this idea. it took a long time to work out the component choices to achieve the goal. I run a parelled 12ax7 after the ef86 in that build. Its a very clean pre-amp, with a bit of punch due to the pentode mode operation.

[dehughes]

You can run a TMB off an EF86. However, for reasons mentioned above, I wouldn't use the 'traditional' TS values.

glad someone said it. You can do it, but your going to be using different value's of components. You need to run the ef86 in pentode mode, as triode mode has a very high insertion loss, and generally lack of gain to drive the TMB. I tried to run in triode mode, and while the math seems to support this, the reality does not.

I built my version of the SSS based on this idea. it took a long time to work out the component choices to achieve the goal. I run a parelled 12ax7 after the ef86 in that build. Its a very clean pre-amp, with a bit of punch due to the pentode mode operation.

Interesting. Would you be so kind as to tell me how you built the preamp? I don't think I'll run a paralleled 12AX7 after the EF86 like you did, but the component values would probably point me in the right direction. And yes, I was thinking of running the EF86 in Pentode as well.