Screen Supply Voltage.

[Jag]

OK, got the "spitty" thing worked out (forgot to install a grid stopper ).

Here's another question regarding the same amp. Screen voltage is about 2 volts above Plate voltage (once again I am at work and forgot to bring my notes so can't quote exact voltages). I've got a 3H choke between the filter cap supplying the plate (40uf) and the one supplying the screen (20uf). Would it be best for me to install a 500-1K resistor between the present screen supply point and the subsequent filters adding another 20uf filter cap and take the supply off the new filter, or can I just put the new resistor in series with the choke going to the second filter?

Going from memory, measuring voltage from cathode to ground I get 38volts, plate to ground I get 474volts and screen to ground is 476volts. The amp is cathode biased.

[MWJB]

Do you have a 470ohm 1W screen grid resistor between the screen supply node and pin 4?

[Jag]

Sorry, I should've been more clear.

Thanks MWJB. Yes, I have a 1K screen resistor on there already.

Also, I posted that information rather quickly yesterday right before heading off to a meeting. It was a complete typo regarding 470+ volts on plate and screen. Here are the real measurements.

Plate Voltage = 438v (to chassis ground)
Screen Voltage = 440v (to chassis ground)
Cathode Voltage = 34v (to chassis ground)

(Do I subtract the cathode voltage from the screen to ground voltage on an SE amp to get the actual screen voltage? In otherwords, I know the plate voltage, in this case, is 404v. Would I say the screen voltage is 406v in this example?)

I could increase the screen resistor but, what I was actually wondering was would it be better to add another supply node/dropping resistor? I ask this because I was reading some posts where some people found the tone was adversly affected by increasing the screen resistor above about 1K, or is this a six in one half a dozen in the other scenario?


Jag

[Jag]

Here's a partial schematic.

http://www.jonesong.com/images/temp/part_schem.bmp

[MWJB]

Yes, you can consider the difference between cathode and screen/plate voltage as the working voltage.

In reality, many SF Champs & Princetons probably run their screens a little in excess of the plate voltage, though it's not "ideal". Try another resistor & filter node and let us know what you get.

Out of interest, are you still running a 6V6 and what cathode resistor/bias are you getting?

[Jag]

Yes, I am still running a 6V6. I've been using a JJ 6V6 because they are a tough little tube.

I'm getting 32.74ma current draw with 34.4 volts across the 1K cathode resistor bypassed with a 25uf cap.

By may calculation, the tube is dissipating 13.22 watts at idle. I'll probably tame this down a bit but the tube seems happy enough for the moment, no red plating.

[Bruce / Mission Amps]

Here's a partial schematic.

http://www.jonesong.com/images/temp/part_schem.bmp

Get rid of the 5AR4 and use softer rectifier to get the B+ down lower.

What is the DC voltage drop across the primary of the OT with the tube idling?
What is the voltage drop across the choke and the 22K preamp node resistor?
These will tell you how much voltage you need to drop across the B+ node that feeds the screen.
1K will not be enough if the screen current is 2ma to 4ma... that will only drop 2v to 4v!!!
You could use the 1K after the choke and in series with the B+ rail to the screen node and then another resistor to the actual screen.
I'd be careful about putting too much series resistance in the B+ rail with out further filtering but up to degree, it shouldn't matter too much what the final DC resistance of the screen resistor is.
The B+ rail is where you need to drop this screen node voltage.
You are looking for a resistance that drops a little more or the same voltage as the current saturated OT winding does, then decouple it with a high voltage cap and feed the screen from there.
It is not critical that the screen voltage be lower then the plate voltage but I think these circuits sound better when they are about 10v-20v less then the plate.
Remember that a 6V6 screen doesn't draw much more then a couple miliamps of current so your screen node resistor will need to be bigger then you think.
1K will do next to nothing.

[Jag]

Thanks Bruce, that answered alot of my unasked questions and filled in some gaps in my knowledge.

I was going to try a 5Y3 tonight as well as a few other things. You've geiven me some more good things to check out as well.

Something I forgot to mention earlier is, I've got MM iron in there. The PT is beefier so I can run 6L6, EL34, KT88, etc. The OT is quite a bit bigger as well. I use this amp to try out various things and as a platform to check out how different tubes change the tone, etc.

I did a bunch of things last night. Here's a general summary.

The first thing I did was up the screen resistor to 4.7k. As you implied, that didn't do much.

Next I reinstalled the 1K screeen resistor and put a 6L6 in the amp. This made me more comfortable with respect to plate voltage and dissipation, but the screen voltage was still 1 -2 volts above the plate.

Next, I siliconed in a 20uf 500v cap and terminated the choke there. I installed a 1K resitor from there to the node supplying the screen. I also replaced the 1K screen resistor with a 470. This brought the screen voltage down equal to the plate voltage. I reinstalled the 1K screen resistor and the screen is now 1 volt less than the plate.

This is where I left off.

I don't really like the current wiring mess with the additional cap glued in. I also think the amp sounds a little bit less open to me now. That could be my imagination , although I did change to a 6L6 so there should be some difference.

I think tonight I might disconnect the new filter and try the original circuit with a 1K in series with the choke. If I read your post correctly, you said that was okay to a point? I'll also try a new 6L6 in there. I'll also measure the voltages you mentioned and post them back here as well as trying a 5Y3. If teh voltage drops enough I'll probably to the 6V6 also.

I really liked the way the amp sounded before I made the changes I'm just concerned that the screen voltage was a little higher than the plate and the voltages/current draw were a little too high for my liking with a 6V6.

[Jag]

Update.

Did some more work on it last night. I've updated the partial schematic with the voltage measurements and component changes.

Here's what I did;

- replaced the 5AR4 with a 5Y3
- reinstalled the 6V6
- removed the added supply node and put a 1K in series with the choke
- reduced the screen resistor to 470

Here are the current stats:

Plate (working V) = 403V
Screen (working V) = 402V
Cathode V = 34V
Current Draw = 34ma (I'm not subtracting Screen current but, close enough)
Plate Dissipation = 13.7W

[Bruce / Mission Amps]

Try this too:

[Jag]

Excellent Bruce! I already did move the CT for the filaments to the top of the cathode resistor. I think the rest of that looks great, I'll give it a try in the morning.

Thanks
Jag

[Jag]

OMG Bruce!!! Does that ever sound GOOOOD!!

I did exactly the changes you suggested. It brought the plate voltage down about 135V (it's around 295v now) and I can run a NOS RCA 6V6 in there and it's happy as a clam.

I have a question though. Why did moving the choke before the first filter, in line with the rectifier, drop the voltage so much more than having it between the first and second filters? Absolutely no complaints the amp sounds fantastic this way, just wanting to get a little theory.

Also, how advisable would it be to install a switch that would parallel another resistor to the one on the output tube cathode so I could run a little hotter with a 6L6 in there? Kind of like the low power switch on the THD Univalve.

[Jag]

Got the answer to the choke question! Chokes

[Bruce / Mission Amps]

OMG Bruce!!! Does that ever sound GOOOOD!!

I did exactly the changes you suggested. It brought the plate voltage down about 135V (it's around 295v now) and I can run a NOS RCA 6V6 in there and it's happy as a clam.

I have a question though. Why did moving the choke before the first filter, in line with the rectifier, drop the voltage so much more than having it between the first and second filters? Absolutely no complaints the amp sounds fantastic this way, just wanting to get a little theory.

Also, how advisable would it be to install a switch that would parallel another resistor to the one on the output tube cathode so I could run a little hotter with a 6L6 in there? Kind of like the low power switch on the THD Univalve.

You are probably running the 6V6 in true class A now.
You might be able to increase the current through the tube for a little more output.
What is the actual B+... plate (lug 3) to cathode (lug 8) voltage now?
And what what is the cathode voltage as measured to ground?
Did you put the GZ34 back in?
If you put a little 1uF-500v cap at the rectifier output, in front of the choke, it will charge up and increase the plate voltage a little... you can experiment with different value capacitors on the rectifier output (before the choke) to tweak your B+ rail a bit.

[Jag]

I've now reproduced the whole project on it's own chassis and it's really sounding great. I figured it was so not a Champ that it should be in it's own box. I also want to restore the Vibro Champ to stock.

Anyway, as I said, it's working great, no issues transferring the guts to the new chassis (everyting but the tube sockets was transferred, I had already built a new eyelet board when I started this).

Yes, I put the GZ34 back in

With a 6L6 in there
- Working plate voltage (pin 3 to pin 8) is 295v
- Cathode voltage is 24.1v (470 5w cathode resistor)

With a 6V6
- Working plate voltage is 341v
- Cathode voltage is 20.1v (470 5w resistor).

*NOTE: The measurements for the 6L6 are from my notes. The 6V6 measurements are from memory as I forgot to write them down. I do remember that, with the 6V6, the cathode voltage was 20.1v and the plate dissipation was about 14.6w at idle. I went backward from this to get the plate voltage.

20.1v/470ohm = 0.0428a
14.6w/0.0428a = 341v



I'd like to install a Hi/Low Power switch in there to optimize the current between larger tubes (6L6, EL34, KT66, etc.) and the 6V6. I'm thinking installing a switch that would parallel another 470 ohm 5W resistor with the existing 470 ohm 5W cathode resistor would get me roughly in the neighborhood. I'll have to tweak the valuse but does that sound like a good idea? I could then go low power for the 6V6 and switch it back to high power for the 6L6, etc. I know I could just optimize it for the 6V6 and swap in the 6L6 without changes but, I'd like to get the increased wattage with the 6L6 in there.