Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Help with Vacuum Tube Circuit

Status
Not open for further replies.

EBSDallas

New Member
My dad and I had just started working on a vacuum tube audio amplifier when he passed away three years ago. He was an EE. I'm not. I'm just now getting back to it and am going from some of our sketchy notes. I'm having trouble with the attached circuit.

It looks to me like the plates of the 12AX7 would see 480 volts which is considerably higher that they're rated. Unless: do the R4/R6 and R5/R7 pairs act as voltage dividers whose values would allow one to select the plate voltage for the tube?

Also, can someone point me to a good regulator circuit for the bias supply?

I know this is real basic stuff for most of you; thanks for your help.

Everett
 

Attachments

  • Tube_Question_001.jpg
    Tube_Question_001.jpg
    126.7 KB · Views: 247
I don't know alot about tube amps, but I can tell you that R4/R6 and R5/R7 do not make a voltage divider. The plates will indeed see the full 480V.
 
I think I see a grounded grid phase splitter...or is that cathode coupled phase splitter?
Anyway, There is a 70 volt bias on the cathodes (Limit = 180V), so that subtracts from the plate voltage.
Then there is the rating: 330 V @ Wikipedia, 300 volts @ GE, 300V @ Sylvania, 300 @ Tung-Sol 250V @ Phillips
I suspect the right answer is 300 volts and I know that tube amps are designed with the tubes at max voltage during idle.
Then there is the voltage drop across R4 or R5, not R4 AND R5.

480 - 70 = 410 so you need to lose at least 110 volts in R4.
That's not unusual. I expect the design, as intended, will use R4 to drop 170 volts, but I also expect Vb will have some resistance to limit the plate current of each tube.

Basically a valid design, but a major PIB when it has no parts values written on it :-(

Now...what was your question?

Oh yeah...bias supply. Build a voltage divider from the 480 volts to get to 70. Add a decoupling capacitor and put a resistance between the 70 volts and the cathodes.

I'm afraid I'm running out of Login time so, quick post.
 
Something very wrong with that circuit. The grids are at +70V with respect to their cathodes (not allowed). The plates are >500V with respect to the cathodes (not allowed).

Look at the Maximum Ratings section of the attached.

btw- This came from my copy of the "RCA Recieving Tube Manual"
 

Attachments

  • 12AX7a.png
    12AX7a.png
    453.1 KB · Views: 207
  • 12AX7b.png
    12AX7b.png
    488.6 KB · Views: 184
I see that Leo Fender used 360 V and 410 V for 12AX7 supplies. After the plate resistor, the plates were running at 270 V plus/minus their voltage swing during amplification. Your design starts with 410 volts difference, so, it's been done, and it went into production. The rest of the design will dictate values for some of the resistors. Do you have the rest of the design?
 
Oops. I didn't think about cathode to grid voltage. Yeah, something is very wrong there. Add that to assuming the 70 volts was positive, and most of my rambling looks pretty weak. Try not to believe that most of what I said is about YOUR circuit.
 
R2 or R3 is redundant and there must be a capacitor or 2 missing so the grids can develop a voltage similar to the cathodes. I still think the 70 volts must be positive and mis-labeled.
 
That diagram is DEAD wrong.

First off...since R2 is already tying V1b's grid to ground, R3 is not needed. With R3 there it does appear to be a long tailed pair phase inverter (i.e. "grounded grid" phase inverter) but in reality it's two parallel gain stages with a shared cathode arrangement that more than likely get "mixed" at some point in the circuit downstream.

Second off, with the grids referenced to ground via the leak resistors and the cathodes referenced to a -70V source, that would place the grids at +70V relative to the cathode and we all know that alone could never work.

Furthermore, with a 480V supply, I suppose you could use a 220K plate resistor and bias the grid to -2V relative to the cathode to get it to sit idle at around 240V, but this would require at least a 2K cathode resistor to ground to make this happen. Also, when the valve gets driven into cutoff, the plate will be sitting at 480V relative to the cathode, which will more than exceed the relative to cathode plate voltage rating of a 12AX7.

Now here's a much more realistic version of the circuit -

**broken link removed**
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top