Mini amp build, 2 problems

[AmberB]

So, I've been building a mini amp that uses a 6SN7 with the plates in parallel single ended.  I finished the build today and powered it up.
 Let me start by saying that I used a power transformer that it way over-powered for a build like this. 
I have a 12 AZ7 in the preamp (it was handy for voltage testing) and a 6SN7 in the power tube socket.  There isn't enough heater current to bring the heater voltage below 7 volts.  I'm using one of those fancy new LED replacement bulbs in place of the standard incandescent bulb for the "power on" bulb, so I might have to put a standard bulb in the socket to see if that helps load the heater supply, but I doubt it will make a difference.  So the first question is, how do I add loading to the filament winding to bring the voltage below 7 volts?

The second thing... The B+ on this amp setup is 410 volts on the plates of the 6SN7, and the cathode voltage to ground is 13 volts.  The plate to cathode voltage is 395 volts.  The numbers don't look like they add up, but that's what I measure with my Fluke 77 meter.  I have a 470 ohm cathode resister that is bypassed with a 50uf-50 volt cap.  I'm testing the amp with an 8 ohm load resistor in the speaker jack.

Anyway, If I'm calculating this correctly, I'm getting .027 amps across the cathode resistor, which calculates to about 10 watts plate watts.  That seems a bit excessive for a tube that has a listed combined plate wattage limit of 7.5 watts with both plates operating.
So, either I'm calculating this wrong, or I need to increase the value my cathode resistor.
I took the value of the resistor from a schematic I found on line.  I'm guessing they weren't putting 400 volts on the plates in that instance, but that schematic didn't give a B+ voltage.
If my math is correct, I'll experiment with different resistor values to bring the plate current down to an acceptable amount.
There's no hint of red-plating, but it's just the initial voltage testing.  It could be different when pushing the amp with a guitar.

Another thing I have to wonder about, would the plate wattage be different with a 4 ohm load?  The output transformer is wound with 1 output and it has an impedance of about 9600 ohms to 4 ohm load.

[PRR]

You don't want to "load down". Why waste power?

The plate voltage is far too high for the current and the load impedance.

Series resistors!! About 1 ohm in series with the heaters (0.5r in each side). 2k to 4k in series with the plates. Consider a fistful of 1k 5W and use 2,3,4,5 until happy.

[AmberB]

I have 2 stages of power supply filter before the output transformer connection, with a 430 ohm 5 watt resistor between them.  I could change that to a higher resistance to lower the plate voltage a bit.  I can also raise the resistance of the cathode resistor to lower the plate wattage of the 6SN7.
I'll try the 1/2 ohm resistors in the heater line and see how that works.  I have some .47ohm-5 watt resistors I can use for that, although I probably don't need a 5 watt resistor.  Since I already have them, I might as well try them.

[AmberB]

Here's a couple of pictures of the build.
The first is the original solid state amp that I used the chassis of for the build.  It's an old Sears amp.

[AmberB]

Here's the solid state circuit board setup before I took it apart.

[AmberB]

Here's the completed build.  I used a power transformer that I had that was originally mounted through the chassis.  I mounted it with stand-offs on the chassis with a hole through the chassis for the wires.  I put a rubber grommet on the hole, just in case.

[tubeswell]

... how do I add loading to the filament winding to bring the voltage below 7 volts?...

Why not add a pair of 6A diodes in anti-parallel - in series with one side of the heater winding? (Each diode has a 0.6VAC forward voltage drop, so a pair in anti-parallel should drop the VAC to about 6.4VAC)

[bmccowan]

Cool project. Why not search on ebay for a more appropriate used power transformer and save that one for another project? 

[AmberB]

Cool project. Why not search on ebay for a more appropriate used power transformer and save that one for another project?

I've been looking at various PTs on ebay, but I haven't found one that would work and I can afford at this time.  I'm also looking for a PT for a different project...
The reason I decided to use the one that I used is because I wanted the 400 volts on the plates of the 6SN7.  My RCA tube manual rates the 6SN7GTB at 450 volts max (design center values)  410 volts on the plates seems reasonable.  Considering the other uses for the 6SN7, it seems to be a robust tube.
I wanted to try the 6SN7 to see what I could get from it.  I could, at some point in the future, rewire the octal socket for a 6V6 and the plate voltage would be within the area of what Fender puts on the 6V6.

[AmberB]

Why not add a pair of 6A diodes in anti-parallel - in series with one side of the heater winding? (Each diode has a 0.6VAC forward voltage drop, so a pair in anti-parallel should drop the VAC to about 6.4VAC)

I'm not sure how to go about doing that.  When you say "anti parallel", what does that mean?
Does that mean that you put the diodes in parallel with one anode facing each direction?

[AmberB]

So, I followed a couple of suggestions on this amp project, and here are the results.
First, I put a .47 ohm-5 watt resistor in series with each side of the heater winding between the transformer and the tube heaters.  That dropped the heater voltage to 6.4 volts AC.  The thing that concerns me now is that those 5 watt resistors get hot to the touch.  This makes me wonder if I should use a pair of .33 ohm-5 watt resistors and see how that works.

The second thing, I changed the cathode resistor on the 6SN7 from a 470 ohm to a 1Kohm resistor.  The plates, the grids, and the cathodes are all wired in parallel.  The result was that it raised the plate to ground voltage from 410 volts to 433 volts.  the plate to cathode voltage went from 395 volts to 418 volts.  The cathode to ground voltage went from 13 volts to 15.7 volts.  My calculations come up with 6.56 watts of plate dissipation.  That falls within the 7 watts max with both plates driven.  The voltage on the plate does seem a bit high, but still within the tube's rating of 450 volts.
The heater winding does not have a center tap, so I have a 100 ohm resistor from each of the heater connection pins of the octal socket to the cathode connection for an above ground reference.  That seems to work well with single ended output amps.

I think that to lower the B+ to the output transformer, and thus the power tube plates, I would need to put a bigger resistor in between the first and second filter caps.  I get a roughly 10 volt drop with the 430 ohm resistor that I'm using now.

[kagliostro]

I confirm your choiche for the tube conversion

this chassis was clearly calling for what you did 



BTW, Tubeswell council about the use of a pair of diodes in antiparallel is a good council


--

EDIT:

When you say "anti parallel", what does that mean?

read the complete article here http://www.valvewizard.co.uk/heater.html


Franco

[sluckey]

a pair of 6A diodes in anti-parallel

I was taught to call that "back to back" diodes.

[SILVERGUN]

a pair of 6A diodes in anti-parallel

I was taught to call that "back to back" diodes.

I would say it's way more like 69 

[sluckey]

That does sound more appropriate. But when I learned about back-to-back, I was still working on second base.   

[PRR]

...the original solid state amp that I used the chassis of for the build.  It's an old Sears amp.

I'd say it was 'my' Kent, some years later. Same cheese-board construction and gloppy glue. Even same braces. Speaker larger and ceramic but otherwise too similar to mention.

[mresistor]

gloppy

You are the second person I have known to ever use this word..      I guess it's a combination of the words glob and sloppy.

[bmccowan]

Or its an example of onomatopoeia - when you drop some on the floor it makes a noise that sounds like "glop."
I've waited 5 years for a chance to use the word onomatopoeia on the forum.

[dwinstonwood]

I've waited 5 years for a chance to use the word onomatopoeia on the forum.

But, it's odd how much onomatopoeia lives inside tube amps: sizzle, pop, thump, hiss, hum, scratchy, buzz, flub, squeal, and so on...

[AmberB]

A bit more part swapping, and the results.
I don't have any high amperage diodes for the heater line right now, so I changed out the .47ohm-5 watt resistors in the heater line for .33ohm-5 watt resistors.  The heater voltage to the tubes is now 6.7 volts, and the resistors only get a little warm.  A good compromise for now.
I changed the 470 ohm resistor in the power supply line to a 1Kohm 5 watt resistor.  it dropped the voltage to the plates of the 6SN7 a bit more to about 430 volts.  I recalculated the plate current and it came out to about 6.2 watts.  I think I'm good there.  I probably could replace the cathode resistor with a bit lower value, but I don't have any power resistors between 470 ohms and 1Kohms.  I'd like to try a value somewhere around 750 ohms to see what happens, but I'll have to put a resistor like that on my next parts order.  I could probably use a 2 watt resistor there, but I'll see if I can get a 3 watt resistor.  The price difference isn't worth worrying about.

Onomatopoeia, I don't that word means what I think it means...

[PRR]

....I wanted the 400 volts on the plates of the 6SN7.  My RCA tube manual rates the 6SN7GTB at 450 volts max....

The pipe used to bring me heating gas is rated 150psi. It actually runs at 11psi or less. You don't have to work things to their bursting-point.

The 450V rating is for switching, bang-bang sweeping of a CRT. Not smooth steady audio.

The higher the voltage the lower the current and the much-higher the optimum load impedance. High impedance audio windings are rare because they are very hard to do well. IMHO you will be better at 250V than up above 300V.

[AmberB]

....I wanted the 400 volts on the plates of the 6SN7.  My RCA tube manual rates the 6SN7GTB at 450 volts max....

The pipe used to bring me heating gas is rated 150psi. It actually runs at 11psi or less. You don't have to work things to their bursting-point.

The 450V rating is for switching, bang-bang sweeping of a CRT. Not smooth steady audio.

The higher the voltage the lower the current and the much-higher the optimum load impedance. High impedance audio windings are rare because they are very hard to do well. IMHO you will be better at 250V than up above 300V.

If I use an 8 ohm speaker on this little amp, the impedance of the OT would be somewhere around 19Kohms.  It's about 9600 ohms with a 4 ohm speaker.  I assume that would be a better loading for the parallel 6SN7 at 400 volts.
About the only way I could knock the voltage down to 300 volts on this amp would be to replace the PT.  I would have to buy another PT to do that...

[PRR]

...About the only way I could knock the voltage down to 300 volts on this amp would be to replace the PT.

4k, 10W.