• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

FINALLY I can start building my GM70

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
As I understand what digitrax wrote, you can use the choke and the tube. With Ia=8mA and serial resistance of the choke RL=3700, voltage drop is V=3700 ohm*0,008 A = 29,6V. Ia=8mA is set when Vg=-3.5V and ca. 148V Ua for 12GN7. So 29,6V+148V=177V Ub voltage. Then driving 12GN7 2,5V rms you will have 240V peak-peak driver output voltage. Enough headroom to drive GM70 :)

Winnie the Pooh
 
Last edited:
Do I need to get a bigger choke if I run 350v to the top of the choke?
Not necessarily. Usually there is a resistor in series with the choke to bring the voltage to an appropriate level and to even out the response (all chokes have a resonant frequency which should be compensated for – hence the parafeed calculator at voltsecond's website).
OK, looking at the graph, I take it that I either need a different input tube or a different choke. Am I seeing that right?
Again, not necessarily. I was assuming an operating point (on the ΓM-70) of about -120V and that you wanted to drive it to full output (in A1); this may not be the case in your instance.

Here's what we need to know:
What kind of sensitivity do you require?
What kind of output are you looking for? (full A1 potential? into A2? just x watts?)
What is your proposed operating point on the ΓM-70?
 
OK, looking at the graph, I take it that I either need a different input tube or a different choke. Am I seeing that right?
Here ya go:
12GN7A Driver.JPG
 
digitrax
What is the purpose of connecting g2 via 100 ohm resistor to the choke and output? Some NFB?

Edit. Silly me. Now I can see
As you've surmised, that's your triode strap.

The resistor is there to stop oscillations, ensure that G2 stays negative WRT the plate, and (sometimes) to ensure the screen voltage isn't exceeded. Since you're operating at a relatively low plate voltage, you could probably get away without it by using a ferrite bead to kill any oscillations (or nothing at all – the GN7 is pretty stable). You could substitute a diode for protection, if you wish, but short of a bad socket connection, your screen can't rise above the plate anyway in this cicuit. Lowering the resistance or shunting the connection might increase your sensitivity and output swing a little, while raising the resistance might help improve linearity (at the expense of sensitivity). I was working on the assumption that you needed to swing 120V in either direction from the ΓM-70's bias point. Depending on where that is, that may not be the case. Have you determined a grid bias for the '70 yet?
 
I'm baaaaaAck. It's been a wild ride these last two years and the only thing I could do in audio is turn on the radio. Now I have a couple months of down time and I am determined to finish this amp and other projects.

I will be ordering all the rest of the parts I need Monday.
 
Administrator
Joined 2004
Paid Member
I know this is a stupid question, but in the shot below, the Center Tap ends up at the lower portion of the two circles. SO, is that lower tap line ground? Please don't laugh.

Yes, it can be. I would place the chokes in the negative side of the supply and ground only after the second choke, this eases electrical stress on the choke insulation and is what I have done in my GM70 amplifier supplies.
 
For some reason, I can't edit that last post so here is the screen shot I wanted to add.

So that's about 6mV of ripple? Is that low enough for a DHT power stage?
I'm currently putting together a power supply for an 845 power stage and I had kind of decided on using 2 x 10H chokes in a CLCLCRC configuration.. In that configuration it's about 100μV of ripple, but I'm wondering if that's maybe overkill?
 
Administrator
Joined 2004
Paid Member
Definite overkill, not that that really hurts unless it drives the costs up substantially.

It depends on the amount of buzz you can tolerate at the output terminals which is a function of the speaker efficiency.

The ripple at the output terminals is the product of two ratios, that of the tube rp vs primary impedance of the transformer and the turns ratio of the transformer. A few mVpp of supply ripple should translate into very small numbers of uV at the output - likely below the noise floor.

Magnetic coupling into the output transformer from nearby power transformers, and ripple from the driver stage if not well filtered is more likely to dominate.
 
Yes, it can be. I would place the chokes in the negative side of the supply and ground only after the second choke, this eases electrical stress on the choke insulation and is what I have done in my GM70 amplifier supplies.

So the chokes would go on the bottom line in the picture I posted and then go to ground after the last cap?

Also you wrote "Magnetic coupling into the output transformer from nearby power transformers, and ripple from the driver stage if not well filtered is more likely to dominate." So should I put a fariday cage type wall between the Pxt and Oxt?
 
Last edited:
So that's about 6mV of ripple? Is that low enough for a DHT power stage?
I'm currently putting together a power supply for an 845 power stage and I had kind of decided on using 2 x 10H chokes in a CLCLCRC configuration.. In that configuration it's about 100μV of ripple, but I'm wondering if that's maybe overkill?

I am thinking that 6mv would not be detectable, I can add more CLs if needed. I'll keep my ears tuned to that to see if it is noticalbe.
 
Definite overkill, not that that really hurts unless it drives the costs up substantially.

It depends on the amount of buzz you can tolerate at the output terminals which is a function of the speaker efficiency.

The ripple at the output terminals is the product of two ratios, that of the tube rp vs primary impedance of the transformer and the turns ratio of the transformer. A few mVpp of supply ripple should translate into very small numbers of uV at the output - likely below the noise floor.

Magnetic coupling into the output transformer from nearby power transformers, and ripple from the driver stage if not well filtered is more likely to dominate.

Well, I've already got the chokes which honestly weren't all that expensive, kind of seems silly to not use them now.. Maybe I can do some sort of dual rail with them or something.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.