High-Gain Amplifier Circuit

Nothing fancy... Just two fixed-gain feedback amplifiers with a volume control.

Schematic(48kB GIF)

I needed 50dB of gain makeup for my passive mixer design, and this is what I came up with.

The THD figure at max output is a simple estimate based on THD readings I took of the two amplifier stages separately. It's just a summed figure, and the THD of the signal generator was not subtracted. Since some of the distortion is noncoherent, and thus does not simply add, actual THD should be a bit lower. I haven't had a chance to test the entire circuit as "one piece" yet.

Nice circuit Dave,

What's the theory behind the 47pF's across the two anode loads?

And did you experiment with the feedback networks, or was that straight calculation?

Thanks for sharing!

Mark

[quote author="Mark Burnley"]What's the theory behind the 47pF's across the two anode loads?[/quote]

They're more of an "academically correct" touch than a downright necessity, especially if you're not going to try to monkey with the feedback values to get lower gain. The breadboard worked just fine without them.

The very short answer is that it's a common approach to tailor the open-loop gain of the amplifier to start falling off at the extremes of the frequency range you're planning to amplify, for the sake of stability. Read the article I posted in this thread for a fairly straightforward explanation devoid of heavy math.

And did you experiment with the feedback networks, or was that straight calculation?

Click to expand...

They were calculated, and I was pleased to find that reality agreed with my calculations. :grin:

The open-loop gain of each of the two feedback amplifies is 20dB greater than the closed-loop gain. The open-loop frequency response of each amplifier is 2dB down at 20Hz and 25kHz, and flat with feedback applied.

Oh, I almost forgot the main reason I posted the circuit here in the first place...

Put a 1:10 or 1:15 input transformer in front of it, and you'll have a slammin', high-gain preamp for your favorite low-output mics! :grin:

Yeah Dave! One day I'll be gutsy and attempt to build some of your designs. :wink:

me too...

yet to start with any valve projects, mebbe this be a good intro :thumb:

Have you thought of making some boards of this???

This is going to come as a shock, perhaps; but although I've been working with electronics for years, I have never made a PC board! :shock:

I really have to give that a try someday. I downloaded Eagle a while back and farted around with it for a little while, but the learning curve is steep and I just don't have the attention span.

A circuit like this is hand-wired easily enough, though. You could wire ten of them in the time it would take to lay out and etch a PCB.

Dave,
if you feel there is too much waisted time with the learning curve then it might be better to get one of us here to knock up a layout on the computer for you.

You could scribble on paper and/or make a version on breadboard and I'd be happy to give it some time. What you do with it after that is all your choice.
Do you have a site ?? One of us can help ther too.

I know there are people here that would love to see a NYD DIY Project.
:thumb:

What value were you thinking of for the unlabel input grid resistor, 1M?

Cheers,
--
Don

It can be any value that suits your application. 1M is a fine "general purpose" value.

Kev, thanks for the offer. I don't consider this a finalized circuit yet, which is why I put "PRELIMINARY" in big letters across the bottom. I may very well be making tweaks as the project develops. I have so little free time to work on this "fun stuff" that it could be many months or a year from now, for all I know. Soundguy Dave can vouch for the fact that I've been talking about this tube mixer concept of mine since well before the 2003 AES convention!

Dave,
I'd like to reaffirm Kev's offer. As a newbee in all this, it is often hard to feel like I can ever bring anything to the table, but I DO know how to do website designs, etc, and would be happy to help with that sort of thing to help you get a page up and running-if that was something you wanted to ever do.

Joel

Thanks for the pointers there Dave...

...I must have missed that link- ta for the scans!!

:thumb:

Mark

Such nice schematic, drawn with a straightedge and shit... :green:

Thanks Dave! :thumb:

Peace,
Al.

PS: Hey, do you know of any other tubes that will work good in a WCF besides the 12BH7? I've seen 6DJ8's on schematics, but their maximum Vp is a bit on the low side. With +B=~300V, seems like a 6DJ8 will die rather quickly... Sorry if this is too off topic :?

Remember that the maximum plate voltage refers to the voltage at the plate with regard to the cathode. In the circuit I posted, the plate-cathode voltage of each triode in the WCF is only about 130V. The main thing you gotta worry about is the maximum heater-cathode voltage, since the cathode of the upper triode is sitting at an elevated potential.

Tubes with high transconductance, low plate resistance, and fairly high maximum plate dissipation make for good cathode followers or WCFs for driving low-Z loads. I know the hi-fi guys have been successful in using 6DJ8 for WCFs.

[quote author="NewYorkDave"]Remember that the maximum plate voltage refers to the voltage at the plate with regard to the cathode. In the circuit I posted, the plate-cathode voltage of each triode in the WCF is only about 130V.[/quote]

The maximum Vp for the 6DJ8 is 130V (at least on the datasheet I have :?). I've seen it used in WCF's with +B voltages of about 150 to 200 Volts (about 75 to 100V per triode)... but I have no idea what tube to use with higher +B voltages (except for the 12BH7).

Peace,
Al.

Well, then, why not use the 12BH7? They're not hard to find.

what kind of caps/resistors should be used for a valve ckt like this? and what would be an appropriate psu? Do the valves in the schem as posted show the power connections? When i place say a 12BH7 in an eagle .sch I get 2 symbols which is fine as it's what shown in the ckt, but there's also another symbol showing pins 4, 5 and 9 (for the power i assume) when u hit the invoke cmd...

[quote author="NewYorkDave"]Well, then, why not use the 12BH7? They're not hard to find.[/quote]

:green:

No particular reason. I've only seen 12BH7's and 6DJ8's as WCF's, and was wodering why.

Asking for the sake of asking... :wink:

Peace,
Al.

Pins 4, 5 and 9 are the heaters. It's assumed you know how to hook these up, so no need to show them.

Peace,
Al.