Issue with Magnatone 450

[Wes]

http://elektrotanya.com/PREVIEWS/634..._450.pdf_1.png

I've got one of these I'm working on and am stumped on how the reverb is operating. The whole amp is working
just fine and the reverb sounds good even. The issue that I'm concerned with is the fact that the 4k 10W plate
resistor in the reverb driver gets damn near 300 degrees while in operation. I disconnected it and put my ammeter
in between the plate and it and measured 500mA of current.

Plate voltage: ~187 V
Grid: ~0.1 V
Cathode: 15.62 V

- The resistor reads right at 4K
- I have a new coupling cap between the plate and the tanks input
- I've got a new cathode resistor and bypass installed
- The 1.5M resistor to the grid is ok
- There is no red plating
- it sounds good!

Any ideas why this is happening is greatly appreciated!

Totally stumped....

[g1]

It can't be 500mA current or you would have 2000 volts across that resistor. The resistor will run very hot, that is why they need to use a 10W there.
What is the DC voltage you measure across the 4K resistor?

[Wes]

Damn zeros.....

Looks like it was 50mA....

Does that seem right?

Voltage before resistor: 379
Voltage after resistor: 186
Voltage across resistor: 193

Would that be nearly 9 watts dissipation?

http://frank.pocnet.net/sheets/049/6/6DR7.pdf

It is unit #2- Anode pin 1, grid on 2 and 3 (internal connection), and cathode on 9.

The datasheet says max 7w dissipation... What a weirdo tube!

[g1]

It does seem the tube is running too hot. They would not spec a 10W resistor if the circuit were running that hot.
Likely suspects are the cathode cap or the tube itself. You can disconnect the cathode cap (20uf/25V) to see if it is a problem. If the voltage across the 10W resistor decreases when the cap is disconnected, replace the cap.
If there is no change with the cathode cap disconnected, the tube may be on the way out, I doubt you have another to check with.

[Jecarroll7]

Is the grid voltage too high (positive)? I might be reading the data sheet wrong but it seem like it should be a bit more on the negative side?

[Enzo]

Good question. Ground the grid with a clip wire and see if it drops the tube current.

[Wes]

Thanks for the help, guys! Good ideas all around.

So that math does look right?

I have already changed out that bypass cap (20/25) and I actually have three of these tubes.

So plenty of good ideas to try! I'll report my findings!

[Wes]

Grounding the grid directly resulted in no change.

Disconnecting the bypass cap resulted in no change.

All three 6DR7s pulled exactly the same- 55mA or so.

Would going to a higher value plate resistor help?

[Wes]

When I monitor the voltage drop across the 4k plate resistor it is continually changing- it's rising. This means that the current is increasing, correct?

[Jecarroll7]

If you're looking for a way to reduce current in the tube, and you've checked the cathode and plate circuits, and eliminated the tube itself, the grid is a good next suspect.

I believe you reported the grid voltage was 0.1volts, the data sheet for this tube suggests -3 to -17 depending upon application.

Maybe look at the grid bias resistors to the voltage source?

[Wes]

Thank you for your reply Jecarroll7. I think I may have voiced that grid voltage reading wrong earlier.
That .01V was with respect to ground and the cathode at nearly 16V.

All the resistors around the grid check ok. I even have subbed in others in their place.


Voltage from cathode to plate (not ground)= 159.4V

Voltage from cathode to grid (not ground)= -15.5V

Voltage from ground to cathode= 16.3V

Am I missing something? Could it be possible that I have THREE bad 6DR7s?

[g1]

Just to be sure, you should probably disconnect the input to the tank, or lift an end of the .1/450V coupling cap off the plate of the reverb driver. See if that has any effect on the tube current.

[Wes]

So I just changed the cathode resistor and got it within spec. No problem.

Thanks for all your help, guys.

[Wes]

Just to be sure, you should probably disconnect the input to the tank, or lift an end of the .1/450V coupling cap off the plate of the reverb driver. See if that has any effect on the tube current.

I did try these things too. No difference. Apparently my ammeter is out of calibration/busted and I relied too much on it's
false readings. For now on I'll look at voltage drops.

Thanks, man!

[g1]

So I just changed the cathode resistor and got it within spec.

So what is now the voltage drop across the 4K/10W ?
The very strange thing here is: resistors going bad increase in value. An increased value of the cathode resistor should decrease the current through the tube, not increase it.
Or are you saying you modified the circuit by changing the value of the cathode resistor?