Magnatone 410 with 6Sj7 pentode input stage. design/build

[tony321owen]

I recently got my hands on some of those old magnatone varisistors and I have a few 6Sj7 tubes, a vibro champ chassis and transformers so I dreamed up this schematic for a pentode input stage paired 6v6 power tube to create this 5 watt single ended amp with true magnatone vibrato! I based the pentode input stage on the silvertone 1301 and most of the rest of the circuit from the magnatone 410.

Now I have to make it work! I will be most grateful to those who can help answer some of my design questions and for those who want to look at my schematic and see that components and circuit design makes sense!

Questions:

1. I would like a low and high impedance input if anyone can steer me towards a logical design...
Also a discussion of input stage grid stopper resistor values and grid leak values would be beneficial! I notice most 12ax7 tubes using lower grid stoppers much lower than 100k and grid leak is usually 1m on grid leak resistors..Any reason for these alternate values for 6Sj7?  Also is the .05 cap before the 6sj7 just to tame bass response?

2. When comparing the the schematics of the 410 and 1301, will the one pentode gain stage of the 1301 provide enough gain to replace the 1st 12ax7 stage of the 410? does my hybrid circuit look like it will work? 1m vs 500k volume pot?

3. designing B+ distribution rail: My transformer spec sheet projects a B+ of 360v when using a 5y3 rectifier @100ma dc load. I wrote some target plate voltages for each gain stage based on schematics I found for similar circuits..
Can anyone walk me through how to choose resistor values to reach my B+ goals at each node? Also does it make sense the way I arranged my power rail with point B node going to the OT and power tube plates with a screen resistor to slightly lower its voltage in relation to the plate..
      The 410 has a 6v6 screen voltage of 50 or so volts lower the the plate where as the 1301 has a screen voltage just slightly higher than the plate.. Can anyone estimate the differences these alternate designs may have on the circuit? Still looking to learn, in practical terms, how the relationship between the screen and plate voltages translate into sound, feel, power and headroom of the amp in both the 6sj7 and 6v6 stages!

Thanks for your knowledge!!!

Below is my first draft! will most likeley be revisions!

-Tony   

[tony321owen]

I was worried that my first draft lacked preamp gain using just one 6sj7(gain factor of 19) at the input stage so I wrote another draft with a 12ax7 stage after the volume control.. this changes my design to more closely match the 410 topography except with a pentode input stage instead of a 12ax7.
It is probably a good thing that I stick closer to the 410 circuit since my knowledge of electrical engineering principles, formulas, mathematics is a bit behind my ability to think up circuit designs that in theory sound interesting but in practicality may fail when I build them! I am probably better off slightly modifying circuits that are proven to work until I learn some more... Hopefully by sticking closer to the 410 schematic I can minimize unexpected design issues..

2nd draft below. Input apreciated!!!   

[PRR]

> 6sj7(gain factor of 19)

That's wired as triode. As *pentode* you can easily get voltage gain near 100. Look on the G.E. datasheet for Amplifier Table. Those suggestions were good enough for Leo Fender. They will get you started.

> ...probably better off slightly modifying circuits that are proven to work until I learn some more...

Yes. But you can "build loose" and modify until happy. As long as you don't put it in the van, you can tack-solder and other "sloppy" construction for easy changes. As each detail comes into focus you then do short leads and correct turret-wrap so it can travel without failures.

I'm not commenting on your proposal because others know this Magna circuit much better than I do.

[bmccowan]

The Fender Champ 5C1 and the Gibson GA-5 Les Paul jr (there are many GA-5 variants) used the 6SJ7 as the preamp in a SE 6V6 amp. Slight differences from the Silvertone. There is a Joe Piazza drawn schematic of the GA-5 LPjr in the Library that is a particularly clean diagram. Any of the 3 should work as the starting point for your project.

[tony321owen]

Thanks for the input guys!

Here is a VERY rough and unfinished layout. Things will certainly be moved to more logical places but for now i just wanted to see a sketch of things all hooked up.

anyone see an issue with having an L shaped preamp design? im thinking about drilling a hole in between the input and 12ax7 tube for the 6sj7..

im also deciding between using a cap can or seperate filter caps localized closer to each stage..

This is the first time where i will attempt a build that is point to point and on turret strips.. hopefully all goes well!

[bmccowan]

No problem with an L shaped layout - it was done a lot in older amps. Not so much later on, but I don't know that the changes were for any performance reasons. I like F&T cap cans and their axial caps. Space and convenience being the only drivers of choice. F&T has some double axials now - good space savers.
I think controlling gain will be your struggle with this amp, but you'll get there.
And, I know this is throwing in an unwanted wrench, but I would try the 12A_7 V1 and the 6Sj7 V2.

[tony321owen]

Thanks for the tips! Is there a reason to put the pentode after the 12a_7?

Does it look like too much gain may be an issue?

Thanks again!

[bmccowan]

Using the pentode as V2 is just a choice - I've done both with both 5879s and 6SJ7s. As the second gain stage it seems easier to get the tone one wants while controlling noise/microphonics. Merlin's chapter on small signal pentodes has good info on this. http://www.valvewizard.co.uk/pentode.html  I recommend the book which has additional info on the subject.
As PRR said, you can get a lot of gain out of a pentode, when its wired as a pentode. Multiple gain stages mean experimenting to control the gain. But experimenting is the point, eh? Unless following a totally proven schematic, I do just what PRR suggests, making temporary changes until getting to some place I like, and then finalizing the build. In circuits along the lines of what you intend, I end up experimenting with different value voltage divider, or grid leak to tame gain.
As a trial and error builder, I'm not the one to do the math on your circuit. Others on this forum are the experts on that. PRR is one of them. And Tubenit has a lot of designs with the 5879 pentode - he helped me with a "lite" version of the HoSo circuit.

[tony321owen]

I decided to use that extra triode on the 12ax7 to turn this into a 2 channel amp. I wanted to keep the traditional 12ax7 input design and circuitry as an option in case the 6sj7 channel experiment doesn't end up sounding good with the rest of the circuit. I know I will probably have to dump some gain from the 6sj7. Im thinking a voltage divider or a split plate-load resistor could help if need be.

The schematic below(draft 3) shows how I join channels with two 220k resistors before the tone controls. Can anyone tell me if my circuit at this junction looks all right?

Also I am not sure about the input circuits. Does it look right to you that on the magnatone 410
schematic there is no grid leak resistor at the input of the amp?

I moved the 6SJ7 grid stopper on the other side of the .005 cap so I could attach it right to the pin. Does this look like a worthy circuit change?

Chassis drilled and will begin building as soon as I iron out my design changes.

I also attached the Magnatone 410 schematic I am basing my design off of.

 

[sluckey]

The schematic below(draft 3) shows how I join channels with two 220k resistors before the tone controls. Can anyone tell me if my circuit at this junction looks all right?

That should work.

Does it look right to you that on the magnatone 410 schematic there is no grid leak resistor at the input of the amp?

The 410 was an entry level amp (cheap). It relies on the guitar circuitry to provide the grid leak resistor when using the high input. When using the low input the 47K resistor on the high jack becomes the grid leak. This works but is not ideal. Spend the extra money and put a 1M resistor across the high jack.

I moved the 6SJ7 grid stopper on the other side of the .005 cap so I could attach it right to the pin. Does this look like a worthy circuit change?

That''ll work. I would probably just remove that .005 cap.

I believe the original schematic has an error on it. Surely that 470K resistor on Node D (B+ dropping string) is wrong. In the similar M2 schematic that resistor is 47K (I changed to an 8.2K in my M2.). And that .047µF filter cap for node D is not nearly enough. I would change that to 22µF or even 47µF. I replaced the .1µF in my M2 with a 47µF to prevent motorboating at high volume settings.

[tony321owen]

Awesome, I thought that .047 looked suspicious at point D.

I will make the changes you mentioned.

Sluckey, I was looking at your schematic for your M2 build and see that the circuits are quite similar!

Can I assume since the M2 came out later than the 410 that the vibrato circuit was improved upon with those extra caps that are in between the cathode/ grid leak network of the 12Au7 stage and the intensity pot?
I do want to add a foot switch control so maybe I should just use the M2 circuit instead.

Also the addition of the Vib/trem switch you used looks interesting as well! Ive never seen bias very trem on the output tube of a single ended amp so perhaps theres a way to inject it on the cathode of the 12ax7 stage after the volume knob like on a vibro-champ?
I know the vibro champ uses a cathode follower on its oscillator so maybe this wont work after all... Does it seem possible with a one triode oscillator design?

Thanks for your input!