Still messin' with my 52fh

[slohand42]

The last time I posted about this amp was almost a year ago in late January of 2009. You can refer to the thread entitled "How to make my 5F2H a little Browner". Through forum feedback and trial and error, I got it runnin' pretty good (actually nearly awsome), but. . . . .and that's a big but . . . .I can seldom leave anything alone, so here I am again.

Last night I finally decided to try one more thing with the preamp circuit. I decided to set the first stage up like the ever so popular Premier Twin 8 with a "grid leak" bias on the first stage of the 12AX7 preamp tube.



The grid leak bias on the preamp tube was known to break up very easy and is preferred by many harp players. The interesting thing is, going to this bias drops the B+ voltage at the plate 90V without messing with the voltage dropping resistors. This voltage is in the range that Gerald Weber says is great for harp tone.

Has anyone else tried out this concept? Do you have any feedback? It was so late when I got it done last night I didn't try it out.

I will be trying it out tonight and will give some feedback on tone etc.

Ted

[ThermionicScott]

I'm not too familiar with the Premier Twin 8, but what purpose does the 100k resistor serve, besides forming a divider with those input resistors and throwing away a bunch of signal? Similarly, I don't think the 68k input resistors will be able to suppress RF when separated from the tube by a cap, but I could be wrong. I'm sure the amp will sound cooler for harp though.

The reason that the plate voltage drops so much, is that grid-leak biasing only provides a fraction of a volt of bias (usually 0.5-1.0V), so the tube is biased hotter and consumes more current.

[slohand42]

I'm not too familiar with the Premier Twin 8, but what purpose does the 100k resistor serve, besides forming a divider with those input resistors and throwing away a bunch of signal? Similarly, I don't think the 68k input resistors will be able to suppress RF when separated from the tube by a cap, but I could be wrong. I'm sure the amp will sound cooler for harp though.

I'm truly not familiar enough with grid leak biasing in preamps to know exactly what everything does. Though this circuit is directly out of the Premier 8, it is indicative of most preamps of this type that I have researched. One of them had the 68K's on the instrument inputs and a 33K on the "mic" input. I may try that to see what it does.

One note. The amp is dead quiet (no hum or buzz) with or without signal, except for the inevitable feedback in my small basement room when I get too roudy.

Thanks for the input.

Ted

[slohand42]

So after gathering all of my harps, mics, and cords, I finally tried out the amp last night. I was very impressed. To start off with, I wasn't sure I had wired it correctly since it was totally quiet (no hum at all) with or without a signal. Cranking it and trying out the mics, it really rocked. A very good and early break up. Dirty (but not muddy) and somewhat honky. A very good sound.

There is still a feedback issue, as with any amp used with the old mics, but the sound is way cool and the tone great before any feedback is encountered. My test room is a small room in the basement, so my feedback is an even worse issue. A more open space would help a lot.

To get the amp a little "browner" with the original preamp circuit, I had added some extra tone filtering which now seems to be unnecessary. In fact I have to turn the tone to the brighter side to get what I like, so I will be tweeking the tone circuit.

I will also, now, go back through the process of testing it with different 12AX7's and different speakers to tune it in even better.

To this point, my sure black label and hot white label sound the best through this amp, but the crystal JT30, though not as crisp, is still pretty good.

So far, I like it!

Will follow up with more info as I have it.

Ted

[loudthud]

Grid leak bias and low plate voltage seem to go hand in hand. There is a school of thought that says it's the low plate voltage that encourages grid current to flow. I've tried using different grid leak resistors and the plate voltage didn't seem to change that much.

I've used grid leak bias on one input of my last two 5E3 based amps. I used a 10 Meg resistor on each side of the input cap. You might find that Crystal Mics work very differently when the input impedance is very high. Experiment with the 100K on the input. The grid stopper still does it's job of supressing RF, I used 100K but would be tempted to use something higher if I wanted a browner sound.

[slohand42]

Experiment with the 100K on the input. The grid stopper still does it's job of supressing RF, I used 100K but would be tempted to use something higher if I wanted a browner sound.

Are you calling what is now a 68K resistor the "grid stopper"? And are you saying that should be upped to 100K or more?

Thanks,

Ted

[loudthud]

Yes. The grid stopper many times doesn't seem like it would do anything because it's just in series between the input jack and the tube's grid. It forms a low pass filter with the tube's input capacitance that will supress AM radio stations. In your schematic, the mixing resistors between the input jacks perform that function. You can experiment with the value, higher = browner. If you go too high, you might start to hear a little resistor noise.

[slohand42]

Thanks, Loudthud. I'll play around with it.

Actually, I only have one input as that's all I need.

Ted

[ThermionicScott]

Grid leak bias and low plate voltage seem to go hand in hand. There is a school of thought that says it's the low plate voltage that encourages grid current to flow. I've tried using different grid leak resistors and the plate voltage didn't seem to change that much.

I've read that it is the close-to-zero bias that encourages grid current, not the low plate voltage. If there is no ground reference whatsoever, the tube will tend toward a bias of around -0.5V, so the large resistor is just there to provide some stability. The fact that different tubes of the same type will stabilize (or not ) at slightly different voltages is one reason why grid-leak fell out of favor...

- Scott

[MWJB]

Compared to a cathode biased input with a 1500ohm cathode resistor, grid leak will run at around 10% less plate voltage given the same plate resistor value & B+ supply.

Grid leak inputs have a habit of choking when hit with a big signal, may not be such a problem with your set up thanks to the voltage divider before the input cap at the tube's grid. Most set ups would just have the cap between input jack & tube grid, with a 1-15meg grid leak resistor.

I like grid leak bias with a 5-10Meg resistor for crystal & ceramic mics, but usually prefer a cathode biased input for dynamic/CM/CR mics. In fact, what I usually do is a "quasi-grid leak" with a regular cathode biased input (say 1.5K-2K cathode resistor) - use a 0.022uf cap and a 5-10meg grid leak resistor, now bypass the 0.022uf cap with a very large value resistor, I use 33meg. Gives some of the fullness you get with grid leak and a crystal, but works well with any other kind of mic too.

[Bruce / Mission Amps]

JPG attached

Hmmm... do you mean something like this:

[MWJB]

Hi Bruce,

Yes, like the 2nd example.

[Bruce / Mission Amps]

Hi Bruce,

Yes, like the 2nd example.

Cool... mind if I try it out on one of my 5F2H amps... which runs a 4m7 input grid load with localized NFB.
I'm in the middle of a new'ish design a push pull 6L6, single 12" spkr, modded tweed 5E5A harp amp (using the same 5F2H setup on the input triode) and should like to try it there too.

[MWJB]

Please feel free to try/share it.

[slohand42]

Report on More Messin'

From some of the advice here and some more research, I did some testing, some more messin' and some more testin'.



Referencing the original posted schematic from a Premier Twin 8, I replaced the 68K resistor with a 33K. I had mistakenly left a 4.7 Meg resistor where the 100 K was supposed to be, so I corrected that. I also increased the 2.2 Meg Resistance on the grid leak after the cap to 7 meg.

Results were great. The tone is overdriven, somewhat honky with a grind that is not overbearing.

After realizing I was close to where I wanted to be, I tested many preamp tubes. On the original 52fh, an old 60's Silvertone 12AX7 sounded best. They didn't stand up to the new set up. Some old 60's GE 12AX7s sounded better, but an old ECC83 with faded markings turned out to be the best sounding 12AX7. I also tried a JJ ECC81 (12AT1) and an Electro-Harmonix 12AU7. The JJ ECC81 has very comparable sound to the best 12AX7. As is common with less powerful preamp tubes, I was able to crank it up higher before feedback. The actual pre-feedback volume increase is not significant, but with more of the dial to work with, I am finding I can "dial" in a good range of tone and sound. I will work with this some more.

Next, Speaker trials with the new set up.

Later,

Ted