Adding a tube

[jeff]

 Hi guys
Say you had a preamp and you wanted to add one more stage before it and one more stage after it. Is there any potential problem with using the two sections in one tube for the first and last stage?
 
In other words is it ok for the signal chain to go: V1a V2a V2b V1b
or should you always go in order: V1a V1b V2a V2b
 
I don't think that should be a problem but thought I'd ask those who are more experienced than me
 
Also any problem with using one triode for LFO and the other triode for signal gain stage of the same tube?
 
I guess I'm wondering how isolated the to trioeds are in a single tube and any potential problems with either the last and first stage in one tube(some kind of bad feedback) or LFO and signal chain in one tube(LFO superimposed on signal).
Thanks
Jeff

[eleventeen]

There should be no inherent issue with that, but it's certainly possible that the *external* components "crisscrossing" might cause some issues. If you're asking if there's a preference to go in sequence, sure there is.


I don' think there would be an issue with the LFO > gain tube-split either. *Could* be an issue if the oscillator was a spiky kind of oscillator, like a blocking oscillator or multivibrator where the tube is slammed all the way on/off. With a mild low-freq sine wave typical of a trem osc, I doubt there'd be any problems.

[kagliostro]

The only thing I can say is that there are tubes like 6N2P (6.3v russian 12AX7) that has a shield between the two triodes






I don't know if this will be of some benefit for your purpose


K

[Jack_Hester]

The only thing I can say is that there are tubes like 6N2P (6.3v russian 12AX7) that has a shield between the two triodes

Same for the 6N6P, 6N5P(6CG7), and 6N1P.  I have these, as well as the one you gave.  All have the internal shield connected to pin 9.

Jack

[sluckey]

Say you had a preamp and you wanted to add one more stage before it and one more stage after it. Is there any potential problem with using the two sections in one tube for the first and last stage?

Well, you will have a very tiny signal at one half of the tube and you will have a huge signal at the other half. And the tiny input signal will be in phase with the huge output signal. Lot's of possibilities for squealing feedback just due to the close proximity of these two signals. Maybe these potential issues could be overcome with careful layout. You would want to use different B+ node to feed each half of the added tube.

I would never purposefully design a layout such as this, but if I needed to shoehorn another tube such as your example, I'd certainly give it a try. I'd be hopeful that it works but not disappointed if it had too many issues.

You would likely be more successful to keep the progression v1a, v1b, v2a, v2b and using one B+ node for v1 and another B+ node for v2.

Also any problem with using one triode for LFO and the other triode for signal gain stage of the same tube?

Lot's of amps do this, especially if the LFO will be used to modulate the other triode.

[tubenit]

FWIW,  I have successfully used a 5751 with a 5879 pentode in the middle.  I've tried it two different ways:

5751 triode ......... 5879  ............ 5751 triode gain stage.   I've done this on two different amps.

5751 triode ........... 5879/5751 as a cathode follower.   I've done this on maybe 2-3 different amps.

I didn't have any oscillation problems.  I don't think it's really the "best" approach for reasons mentioned above but I did get away with it.

With respect, Tubenit

[MakerDP]

I've seen it done. Kevin O'Connor does it in his London Power Standard for example. He uses V1 as a "Fenderish" clean stage and switches-in V2 between V1a and V1b to create a 4-stage "lead channel."

[kagliostro]

Seems to be an articulated project

http://www.fydamps.com/photo.html

K

[MakerDP]

Seems to be an articulated project

http://www.fydamps.com/photo.html

K

What do you mean by "articulated project"?

[tubenit]

What do you mean by "articulated project"?

I understand this to mean that it is a "documented & proven" project.

With respect, Tubenit

[MakerDP]

Ah, OK. Just never heard that used in this context before. Thanks.

Yeah, it's also the design he sells in his preamp kits. The design can be found in his books TUT, TUT5 and the preamp in "Tonnes of Tone." The preamp version of it was actually my first attempt at an amp build about 20 years ago. I failed miserably and gave up because I knew nothing about proper lead dress, grounding, etc. and there were no Internet forums around to guide me back then.     

[jjasilli]

I think the problem resides here, quoted from Wikipedia under Motorboating:  "One common cause [of motorboating] is feedback through the plate power supply circuit. The power supply provides DC current to each tube's plate circuit, so the power supply wiring (power busses) can be an inadvertent feedback path between stages. The increasing impedance of the filter capacitors at low frequencies can mean that low frequency swings in the current drawn by output stages can cause voltage swings in the power supply voltage which feed back to earlier stages, making the system a subaudio oscillator. This is caused by inadequate power supply filtering or decoupling."


Early signal stages getting mixed in with distant B+ nodes, especially within the same tube envelope, may give rise to motorboating.  But as Sluckey says, it may work out fine.


If not, I would add a filter stage, re-figure dropping resistor values as may be needed, and re-connect all the gain stages to the B+ nodes in proper order.

[MakerDP]

OK I just looked-up the Standard preamp I alluded to earlier. It is similar to what was asked about but it is also a little different...

The topology goes like this:

Clean channel: V1b -> V2a

Lead channel: V1b -> V1a -> V2b -> V2a

So, while there are two more gain stages inserted between the two from the clean stage, it's not just a matter of sticking a new tube in between an  existing two-stage preamp's tube.

As was alluded to in the previous post by jjasilli, he does use a separate filter for each triode's B+, with the order of "more filtering to less filtering" following the topology above.