• 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.

Vacuum Tube Amp Circuit

Status
Not open for further replies.

Leftyretro

New Member
I built the following circuit:

Vacuum tube audio amplifier : DISCRETE SEMICONDUCTOR CIRCUITS

I tested the power supply voltages plugged it in and i could hear audio for about 1 second then it started to distort then it just started buzzing i think its the tube but i don't know what could have broken it. Any ideas?
Well I'm pretty old and I thought I've seen just about everything, but using a automotive ignition coil as a audio output transformer, now that is a first for me. Wonder if it really work? Not so far I'm sure you would add. :p

Good luck and sorry I can't help much, it's basic troubleshooting, measuring voltages and current with meters and looking at waveforms with a scope, hard to do via forum postings. Love the idea. If you got one of those chrome covers for the ignition coil that baby would deserve being mounted outside right next to the tube!

Lefty
 
Hi

Never yet seen 12V powering filament windings on any 12 series tube, the norm is 6.3V. Yes I know what the specs say but look through all your guitar tube amp designs, all are 6.3V.

Do your tubes still glow? If not it may be your filament windings that have blown.

You might want to register on Hoffman Amplifiers, Tube amplifier parts, tube amps, tubes and log into their forum to get their opinion. Bunch of tube guru's around on that one.

Cheers
Andrew
 

Mikebits

Well-Known Member
The number 12 in the pn# means the tube runs with 12vac filament.
 

Leftyretro

New Member
Hi

Never yet seen 12V powering filament windings on any 12 series tube, the norm is 6.3V. Yes I know what the specs say but look through all your guitar tube amp designs, all are 6.3V.

Do your tubes still glow? If not it may be your filament windings that have blown.

You might want to register on Hoffman Amplifiers, Tube amplifier parts, tube amps, tubes and log into their forum to get their opinion. Bunch of tube guru's around on that one.

Cheers
Andrew

No that's not a problem. The 12X tube has a center tapped filament so you can power the filaments with either 6 volts or 12 volts depending on how you wire them up.

Lefty
 

Nigel Goodwin

Super Moderator
Most Helpful Member
Likewise, I'm horrfiied at attempting to use an ignition coil for a speaker transformer :D

Mismatch, low quality, or what?.

But even 'worse' (if it could be) there's no bias for the valves, so they presumably ran flat out and fried themselves.
 

Willbe

New Member
I'd be concerned about the "freq. response" of an ignition coil.

Something failed in your circuit. Let's hope it failed by opening and not by shorting.
Take voltage readings and post them.
You might also check what kind of plate current is passing through this tube.
 
Last edited:

Roff

Well-Known Member
I checked the 12AX7 datasheet. Maximum plate dissipation is 1.2W for each section. From the V-I curves, it looks like each section is running at about 640mW.
I'm not endorsing the circuit, just commenting on the power issue. I haven't played with tubes in years.
 

crutschow

Well-Known Member
Most Helpful Member
Make sure the grid leak bias resistor is actually 100k ohms. The grid leakage current across this resistor is what generates the required negative grid bias for the tube.
 

i_build_stuff

New Member
Actually, an ignition coil *should* have an almost-OK frequency response, because the engine needs to run over a pretty wide range of speeds.

I will say that I don't like the bias scheme they're using. The electrons from the cathode will charge the grid when the input signal goes positive, and then will be stuck there when it goes negative, which could push the tube into cutoff on your low half-cycles and cause clipping (and eventually cut the tube off completely if your grid resistor is big enough). If it works for one second every time you turn it on, then this is probably the case.

Maybe you could try measuring the plate current, to see what it does? If it's the cutoff thing, then it will start out at like 3.75mA, and then fade away to zero as you apply audio. If this is what's going on, you could fix it by reducing that 100k resistor to maybe 10k or 1k, or adding a negative bias to the grid (and increasing the plate voltage for the same output power, but watch that you don't go over 1.2W per plate).

Do note that a 12AX7 isn't really big enough to drive a speaker though. I've made headphone amplifiers out of little tubes like that which sounded pretty good, using a 120V:9V power transformer at the output. The transformer doesn't need to be anywhere near as "good" for headphones, as their impedance is a lot higher than a regular speaker, and it can live without the air gap because the couple milliamps you're pushing isn't enough to saturate the core.
 

Hero999

Banned
Wow an ignition coil as an audio transformer!

Whatever next?

I would have thought a mains transformer would be better even though the treble response will be poor.
 

Nigel Goodwin

Super Moderator
Most Helpful Member
Make sure the grid leak bias resistor is actually 100k ohms. The grid leakage current across this resistor is what generates the required negative grid bias for the tube.
There is no grid leakage current - and the grid resistor has no effect on anode current in any method of biasing.

There are essentially two methods:

1) You provide a negative voltage to the grid, and ground the cathode.

2) You connect the grid to ground via a resistor (as in this circuti), and place a resistor in the cathode. The voltage dropped across the cathode resistor is the negative bias voltage for the grid. This is called auto-biasing.

This circuit runs the valve absolutely flat out at all times, not good - and I've NEVER seen it done - or using a low signal triode as an output valve.

It's really, absolutely, terribly, horrible - don't consider it for a moment :D
 

crutschow

Well-Known Member
Most Helpful Member
There is no grid leakage current - and the grid resistor has no effect on anode current in any method of biasing.

There are essentially two methods:

1) You provide a negative voltage to the grid, and ground the cathode.

2) You connect the grid to ground via a resistor (as in this circuti), and place a resistor in the cathode. The voltage dropped across the cathode resistor is the negative bias voltage for the grid. This is called auto-biasing.

This circuit runs the valve absolutely flat out at all times, not good - and I've NEVER seen it done - or using a low signal triode as an output valve.

It's really, absolutely, terribly, horrible - don't consider it for a moment :D
Now don't bust a vessel.;)

There is indeed a grid leak bias circuit that works for certain vacuum tubes (see http://www.angelfire.com/electronic/funwithtubes/Amp-RC.html "The Zero Bias Grounded Cathode Amplifier"). There is a very small grid current due to a small number of electrons that the grid intercepts, but it does require a large resistor value of typically 5 to 10 Megohms, to properly bias the tube. That may be the reason his circuit isn't working, it probably should have a 10 Meg grid resistor, not 100k.
 

Nigel Goodwin

Super Moderator
Most Helpful Member
10meg is pretty normal anyway in valve circuits, hoping blindly that using no cathode resistor will magically bias the valve anywhere near where it needs to be is overly optimistic :D

There's not a single thing about the so called 'amplifier' that is correct - it's really the worst thing you could try and build.
 

fernando_g

New Member
Please perform the following experiment:
Without an audio signal, measure the DC voltage across the 100K resistor. If there is not at least minus 1.8 volts, then your tubes ARE NOT working class A, which is required for this circuit.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top