coupling the trem ?

[Chuck H]

I'm starting a custom build for a friend. I'm looking to do a tube design (by request). I'm going to modulate the power tube bias. The standard Fender design for a pair of big bottles uses a cathode follower with a 470k cathode!?! I know this is going to be a problem for many new preamp tubes known to fail when faced with high cathode/heater differential. I've been reading about the problem and it seems to be an issue with many Russian offerings. I CAN use a Chinese tube but I'd rather not build a circuit that is brand specific. The reissue Fender amps that use bias modulating trems look to include the second triode in the oscillator and have a 100k cathode on the coupling triode. Better, but still over 150V on the cathode. Andy Fuchs reports problems with as little as 30V differential. That seems really poor to me. possibly exaggerated. My current plan is to use the reissue type circuit, DC elevate the heaters 40 or 50V and hope for the best . Unless anyone knows a better way to couple an oscillator to a pair of big bottles with an all tube trem.?.

TIA

[Tom Phillips]

I'm starting a custom build for a friend. I'm looking to do a tube design (by request)....TIA

Chuck,
My input doesn't directly answer your question but is there any chance you can explain to your friend that the audio signal does not pass thru the trem tube in the circuits that you propose? If so your build would be a good opportunity to use MOSFETs in place of the trem tube triods. Result = all the goodness of a tube amp and higher reliability for the tram circuit. I've been thinking of trying that when I need trem in my next build from scratch.
Regards,
Tom

[Chuck H]

I was actually holding my breath (knowing I wouldn't suffocate) counting the time it would take for someone to read "tube design" and "all tube trem" and still suggest FET's instead of addressing the subject I have already considered SS design alternatives for both the trem and reverb circuits on this amp. As it is, with a tube reverb, a tube trem and an AC50 type design I'm looking at 5 preamp tubes and a pair of el34's (requiring 5A filament supply for a 50W single channel amp!). If I went SS for the trem coupling and the reverb recovery I could do it with with three preamp tubes. And that would be fine by me. The all tube aspect is a matter of aesthetics for this guy. Not much to be done about it. I don't mind anyway since I get to sell him a bunch of stuff that's just laying around on my shelves now.

[Rhodesplyr]

Chuck, In a way, if that's what your friend wants, then, in certain ways, he deserves what he gets :-) Running tubes with heater-cathode voltages that high was always trouble.

This is just a non-thought-out concept, but why does the oscillator tube *have* to be direct-coupled to the cathode follower? I realize that it's to avoid roll-off at low frequencies, but we do have many more sources of capacitors in the several uF range nowadays than they did when the circuit was designed. Could a 12AU7 stage achieve the same drive with higher current and a smaller value cathode resistor? Gain isn't an issue in a cathode follower. How large an amplitude does the LFO need to have?

That said, this does seem to be a circuit where solid-state devices would work just fine.

[riz]

Wondering out loud if any of the following would work:

How about using an NOS-only but still readily available single triode, that would hold up in the stock circuit? No risk of interchanging a sub-par new tube, and a lifetime supply should be what, 2 or 3 tubes tops?

A super-elevated heater supply for just that one tube.

[Chuck H]

Chuck, In a way, if that's what your friend wants, then, in certain ways, he deserves what he gets :-) Running tubes with heater-cathode voltages that high was always trouble.

I think there are plenty of cranks out there with vague criteria. The "all tube" aspect does have some charm.

Could a 12AU7 stage achieve the same drive with higher current and a smaller value cathode resistor? Gain isn't an issue in a cathode follower.

I had considered this this there will be a 12au7 in the amp with one half as the reverb driver and I was going to use the other as the preamp cathode follower. Not sure if there's any advantage to using my other au triode as the preamp cathode follower or the trem coupler.?. I did read the au data and it indicated a filament/cathode differential of 100VDC with a 200V peak. So it looks like the tube type perfomance vs. working conditions becomes apples to apples.

How large an amplitude does the LFO need to have?

I'm guessing about the same as the Vibrolux circuit indicated above. I'll be using el34's (which require little less bias voltage than 6L6's for an equal plate voltage) but I'll have about 460Vp. So...? probably about the same for similar performance. I haven't heard the reissue trem. I don't know how useful an extream on/off effect would be. I've never been a trem user but that sort of thing seems goofy to me.

That said, this does seem to be a circuit where solid-state devices would work just fine.

Noted. In fact I just did a repair on an old Kustom K250 that had what I thought was (and have read others think too) a very good sounding SS trem. Alas...

[J M Fahey]

The purist approved solution is to build it "the way Leo did" *with* "the tube Leo used".
Build it the vintage way, plug there a vintage (NOS) tube and call it a day.
Your friend will be happy.

[Steve Conner]

Yep, if you elevate the heater supply, it should be fine. I've been running my hi-fi amp at +40V DC for ages, partly to quiet the hum, partly to take the stress off the direct coupled cathodyne stage.

You might want to go further than 40.

[tubeswell]

What the others already said: simplest way is a good NOS tube and/or elevate the heater for the cathode follower. A source follower also works fine and an IRF820 is virtually a drop in replacement for that CF triode.

There are other ways, like using a DC-coupled level-shifted CF stage, or a cathode-biased CF stage with a lower value (47k-100k) load (so that the cathode ends up at 100V or less) - both of which will require mucking around with any voltage divider you may have as part of the intensity pot. I have tried both of these ways and they can end up doing something peculiar to the trem (which you may or may not like). But Leo's way is the neatest

[Gingertube]

Do you have a copy Of KOC TUT5.
Look at his proposed circuit for the Traynor Guitar Mate Clone.
He uses a dedicated 12AX7, 2 triodes across the output tube grid drive lines to 0V, one triode on each of the push pull sides and then feeds the tremlo oscillator output to both 12AX7 grids.
Jus one more option which fits the all tube design philosophy but requires one extra tube.
If you don't have it and want to see it, reply here and I'll scan that page for you and post it.
Cheers,
Ian

[Chuck H]

Thank you. I won't trouble you for the scan. I don't have the extra heater current for another tube on this one.

Using an NOS tube to operate a vintage circuit doesn't meet my criteria for current production tubes. Should I die or become otherwise unable to service the amp myself I wouldn't want the amp owner to have too many problems getting it serviced. I'm no repair tech but I'm shocked by some of the service experiences I read about. There are a lot of apparently qualified, but really stupid techs out there. Building a circuit for a friend that requires some specialized knowledge just seems like a bad idea.

I'll probably go with my first scenario. I'll build the reissue trem circuit (who wants to source a 3MRA pot anyway). And I'll elevate the filaments. I'll get the filament/cathode differential down to about 90V and hope for the best. Obviously the New Sensor spiral filaments won't do. But everything else should be OK. I should probably design in a fail safe so that if an incorrect tube is plugged in, it won't be able to damage other components if/when it fails.

[loudthud]

Weber sources the 3MRA pots and the tapped ones used on Blonde/Brown Fenders.

[tubeswell]

Chuck - in case you want to reconsider, a mosfet source follower is really easy to implement - saves using a triode, doesn't affect the sound of the amp or the performance of the trem in any way, and you don't have to worry about h-k voltage.It goes straight in place of the CF stage, and all you need is the FET (a high-voltage type like IRF820), a 100-200R resistor in series with the gate (to snub ocsillation) and a 12V zener between the gate and the source (to protect the gate-source insulation)See schematic (which I have implemented on a number of bias-vary tremolos)

[Chuck H]

Weber sources the 3MRA pots and the tapped ones used on Blonde/Brown Fenders.

Yeah. I looked those up AFTER posting just to see if they were around. When I started doing amps the 3MRA wasn't available. Looks like things have changed. I've wondered about the tapped pots too. I may pick up a couple just to try them out since I'm kinda picky about my tapers and settings. I use whatever makes an amp sound like I think it should with controls set at nominal positions. So different tapers, taps, resistor bridges, etc. are always useful for me.

[Chuck H]

Thanks tubeswell. That's something "I" will definitely be using for myself. I don't have a trem amp of my own but after going through my parts inventory on this project, and finding I have what I need for two such amps, I decided to build myself one. I rarely build an amp for myself so I'm stoked.