Grounded cathode pre circuits

[majorbacon]

My absolute all time favorite small amp is a 1941 Oahu/Dickerson melody master. Anyone who hears it freaks. I opened it up to compare inputs with a 1951 rauland with the same 6sf5 v1 and a 6n7. Then I noticed the v1 cathode was directly grounded. The input had a .05 capacitor followed by a 5.6M grid leak resistor. I have several questions. Are there any guitar amp builders using this style input? Can this input design and directly grounded v1 cathode by applied to fender or other common guitar amps? Why would designers have left this design behind in favor of cathode resistors with bypass capacitors?

[Zozobra]

The Bad Cat Hot Cat 30 has a similar arrangement. There was quite a lot of discussion about the pros and cons of it on here fairly recently.

[jazbo8]

Grid leak bias relies on the very small leakage grid current (thus the very large value Rg used) to work, so the bias has more drift with time, and also with tube changes/swaps.

[Chuck H]

Grid leak bias relies on the very small leakage grid current (thus the very large value Rg used) to work, so the bias has more drift with time, and also with tube changes/swaps.

It's also said to be noisier and doesn't handle preamp driven pedals as well.

But it's absolutely the right input for the Melody Master. If it sounds good, it IS good. However that amp is used that makes it sound good is the right way to use it. The input circuit is part of a larger circuit with ideals in mind for it's operation. Would a grid leak bias input be a good choice for a Peavey 5150??? Nope. But it's just right for the Oahu.

[printer2]

Can this input design and directly grounded v1 cathode by applied to fender or other common guitar amps?

Will this work?

[nashvillebill]

I'm no expert, but I'd leave the 1.5K on the cathode permanently grounded. The switch would shunt around the 1.5K to ground, that way you never lose the cathode grounding if the switch gets dodgy. Then on the grid side, put a 4.7M in series with a 1M, going to ground; the switch would shunt around the 4.7M so you always have a grid leak.

Oh, and drop the 68K grid stopper to 34K...

p.s. You might want to look at the early Ampeg V4 to see how they did their sensitivity switching with the large grid leak resistor.

[EFK]

Valco used the grid leak input on channel 1 for the Supro 1624T, 1690T, 1694T and 1615T up into 1961-1962. Furthermore, they stuck with it on the little Supers, Spectators and Comets into the mid-1960s. It's one of the reasons the earlier 1624T circuits are so prized in comparison to the later cathode biased versions and subsequent 6424 (dual tones). It's wild and wooly and sounds great. Later versions sound emasculated in comparison. The Supro version of it on the 1624-and-up bigger amps used a small .01 grid input cap, which allows the amp to handle pedals such as... oh, lets say a Tonebender... without going immediately into grid blocking on the input.

[jazbo8]

If the grid leak bias sounds better then why did it not caught on with other manufactoruers? Inquiring minds want to know...

[J M Fahey]

It may sound better in some *very* specific situation, it also has lots of drawbacks.
Once popular, some 60 years ago, now obsolete.
Hey!!!, battery biasing was also popular once, nowhere to be found today.

[EFK]

Well it sounds better to me, I should insert that caveat! As far as drawbacks, I'm aware of the theoretical issues relative to aging tubes and the use of different tubes, however in practice I have never found any of this to be a problem and I've been building quite a lot of them. Many old Valcos still retain their original tubes and still sound great! About the only issue I've encountered is that *occasionally* when loading up an amp with tubes, I come across a new manufacture 12AX7 that gets hissy or static-y sounding in the V1 slot and I need to move it to a different slot. Others may have different experiences, if anyone has been using a grid leak front end and has run into consistent problems of some kind I would be very interested to hear about it.

[Jarodowen]

I agree with you! if you see yours let me know I"m trying to make a wall of 6

[Gingertube]

Noise is the issue with grid leak bias, mostly from coupling from AC heaters. You don't want too much gain after it. OK on the input triode in a 2 stage preamp. Becomes problematical for 3 stage preamp and just about impossible on a 4 stage preamp.
You can read quite a bit about it (grid leak biasing) and its problems in RDH4.
Worst problems with high mu tubes like 12AX7 (mu = 100) and 6SL7 (mu = 70).
BTW 6SF5 has mu = 100 too.
If you want to try grid leak biasing may I suggest you use DC heater on that stage.
Cheers,
Ian

[Chuck H]

Noise is the issue with grid leak bias, mostly from coupling from AC heaters. You don't want too much gain after it. OK on the input triode in a 2 stage preamp. Becomes problematical for 3 stage preamp and just about impossible on a 4 stage preamp.
You can read quite a bit about it (grid leak biasing) and its problems in RDH4.
Worst problems with high mu tubes like 12AX7 (mu = 100) and 6SL7 (mu = 70).
BTW 6SF5 has mu = 100 too.
If you want to try grid leak biasing may I suggest you use DC heater on that stage.
Cheers,
Ian

Interestingly... Some say that grid leak bias should be quieter because of the cathodes relative proximity to 0V! But this has never been my experience. Though I can see where an idealized circuit may imply that. I Submit this for discussion with no intent on defending it

[Gingertube]

I would hazard a guess that teh problems stem from the high impedance node at the grid. The higher the impedance of any circuit point the more prone to stray noise from small stray capacitance and straight radiated noise. with a 5M6 I would expect it to be 5.6 times more susceptible to noise than the standard bias arrangement with 1M grid leak.
Also:
Grid current is statistical in nature, it has a DC component and a AC noise component. That AC noise component will generate more noise across 5M6 than across 1M.
The 5M6 will be shunted by the guitar output impedance so you will find that it may be noisier at some setting of the guitars volume control, theoretically best a half volume and worst at low or near max volume.

Grid leak bias was originally used when battery heaters were used so noise (at least the coupling from heaters component) was then less of an issue, thats why I recommend DC heaters.

Cheers,
Ian

EDIT:
If you wat additional understanding of this I posted an essay a while back:
http://music-electronics-forum.com/t32978/
One of the things explained in that essay is that grid leak bias does not work well with old gassy tubes.

[tedmich]

If you want additional understanding of this I posted an essay a while back:
http://music-electronics-forum.com/t32978/
One of the things explained in that essay is that grid leak bias does not work well with old gassy tubes.

cool essay Ian and the paper Jazbo8 linked to is also quite interesting, the audio files from the paper (2011) are still online:
Department of Signal Processing and Communications :: Tube Modeling

and also Gitarreneffekte files!
Allgemeine Nachrichtentechnik :: Gitarreneffekte

funner than my undergrad instrumentation class ever was...