12AX7 is possibly the worst choice for this application (not that it hasn't been done!). The reason is that a 12AX7 is a very low current tube with high internal resistance, about 0.5-2mA and ~62k respectively, combined with much more gain (100) than you need. You can use tricks to lower the circuit gain, but the high Rp and low current will remain. In addition, if you make a cathode follower section out of the 2nd half of the tube, it has very low transconductance (gm), so the stage will also have a gain of much less than one. Please ditch the 12AX7 idea.
I don't have spice drawing software on this PC, let me get to another and I'll draw you up a schematic. It is really simple, you'll love it.
As far as 12AX7s go, the old Telefunkens are as good as it gets.
Stuart
Here is the schematic.
Output impedance will be about 1k. Drive anything over 10k just fine (probably even less!)
Power supply needs to deliver 180-220V @ ~40mA. Use tube rectification please. Hexfreds are just okay, a 6X4 sounds nice. Look at the Edcor catalog for some affordable and really good transformers.
The 12B4 has a filament center tap, so the same 6.3V filament supply can supply all the filaments (6X4 & 2 x 12B4).
Stuart
Output impedance will be about 1k. Drive anything over 10k just fine (probably even less!)
Power supply needs to deliver 180-220V @ ~40mA. Use tube rectification please. Hexfreds are just okay, a 6X4 sounds nice. Look at the Edcor catalog for some affordable and really good transformers.
The 12B4 has a filament center tap, so the same 6.3V filament supply can supply all the filaments (6X4 & 2 x 12B4).
Stuart
Last edited:
There are many many different ways to realise a preamplifier depending on what characteristic the designer wants to emphasize.
Now the required feature is "tube distortion". All the rest is insignificant, since TDA2040 does not even need any preamplifier to work well with line level sources.
And what is your suggestion to this ?
Your design looks to me as a completely opposite to the main specs "tube distortion.
This is maybe irrelevant, but I still comment some of above.
The gain of 12AX7 type tubes can be adjusted by NFB. This is not a trick.
It is a part of good design and engineering.
Also I can not agree with you about the cathode follower done with 12AX7. Do you seriously claim that it can not supply to 100 kohms load ?
Now the required feature is "tube distortion". All the rest is insignificant, since TDA2040 does not even need any preamplifier to work well with line level sources.
And what is your suggestion to this ?
Your design looks to me as a completely opposite to the main specs "tube distortion.
This is maybe irrelevant, but I still comment some of above.
The gain of 12AX7 type tubes can be adjusted by NFB. This is not a trick.
It is a part of good design and engineering.
Also I can not agree with you about the cathode follower done with 12AX7. Do you seriously claim that it can not supply to 100 kohms load ?
You are right. Many respondents seem to ignore the OP's goal. But, the 12B4 might be a fine choice anyway. Just make a simple grounded cathode amp. Use a steep load line. In fact, you can dial in how much distortion you want by playing with the load line. Start with something like 3k anode resistor, 230 Ohm cathode resistor (unbypassed), 120V B+. That will give you a gain of about 3.5 (11dB). It won't have great PSRR, so you will need a fairly quiet supply.
Sheldon
Well, I must plead guilty as charged. I am one who has ignored the OP's request for tube distortion. I'm not smart enough to know what that is, or what it sounds like. Do we mean the sound of a soft clipping cathode biased stage? An insufficiently decoupled power supply? Lots of wax and ceramic caps? Out of spec carbon resistors? A heavily loaded high impedance stage, rolling off the extremes and compressing the signal? A 12AX7 driven into heavy grid current like a guitar amp? What?
I willingly admit that I ignored the request because I didn't know what it was. But I do know that a 12AX7 is inappropriate for driving anything other than very high impedance loads. 250k? Okay. Will it, as a CF, drive 100k? Marginally. Try another tube, almost any tube, and unless you prefer AM over FM, cassettes over LP, you'd not choose the 12AX7 as a follower. Otherwise, not to my satisfaction, and I've built a few (before I knew better). Much better to have a high current low impedance source. If you're just building a follower with no gain, there lots of choices that will sound just fine.
My use of the word "trick" was probably a poor choice. Of course global NFB is a time honored technique. Sounds like compressed crap, but hey, maybe that's "tube sound, eh?" Sorry, not for me. Global feedback is almost required in many instances, such as Class AB PP power amps to linearize the output stages. But, tube line stages work perfectly well without it. I prefer to avoid it when possible, but I usually strive to avoid adding distortion, so my thinking is backwards.
As to pinkmouse's suggestion of Stuart Yaniger's Heretical Line Stage. Here, here! A great design, especially if used with the servo. I like it very much using a 6SN7/6FQ7 as the follower tube. Not as low impedance as the 6DJ8, but still ~200R and nice sounding. SY's design is very nice indeed. Essentially transparent. "Tube sound?"
Sorry for stepping on toes, just expressing my opinion. It's no better than anyone elses.
One more idea, if you're looking to reduce gain and add a sonic signature, why not add an interstage transformer to the output of the tube gain stage? You could use a 12AX7, with it's thin tizzy treble (okay, not always) and drive a transformer that will reduce the output impedance as well as the gain. Don't know if anyone makes plate load transformers for such a low current application, but it might be worth contacting someone like K&K about.
Best of luck, OP. (and yes, I know, I can be an opinionated jerk at times.)
Stuart
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.