More rocket questions: 12ax7 in parallel

[benoit]

I have sources of fairly together rocket info.

1. Paul Ruby's layout and gutshots of Diana near the bottom of the first page of this thread: http://ampgarage.com/forum/viewtopic.ph ... ght=rocket

2. Mark Abbott's drawings in the files section (schematic sans powersupply and partial layout)

3. Matt Taylor's rocket schem which is attached somewhere in this thread:
http://ampgarage.com/forum/viewtopic.php?t=3230

None of these sources offer a complete schematic and layout. Matt and Mark's schematics are reasonably close, a few different values, Matt included a hi-cut. Paul's has some different components, a bright switch, and the cut. I could just work off his layout except I want something closer to the (or maybe I should say 'an' original).

The biggest difference I can see is that paul wired his first 12ax7 one side in parallel with the other (correct me if I'm wrong). What exactly does this do?

[riscado]

more gain, not a lot like say a cascade, also no noise increase... it's a nice trick, much like crossing channels on a marshall, but hardwired instead!

[rfgordon]

There are several benefits from paralleling the triodes, as has been mentioned, such as better noise rejection. Also, the output impedance is lower, so the gain produced is preserved better as it moves to the next load. The other cool thing is that, according to some charts I saw (but can't remember where--darn that middle aged brain!) parallel triodes produce a larger second harmonic content.

You can also do cool things like voice each side differently, then bring them together.

[paulruby]

It's not really a material difference. So don't sweat it. My opinion is that the extra triode is sitting there, so use it to reduce hiss a little... Either way is fine.

[mirlo02]

Hi guys,my first post here,sorry,
my english is very bad.How to bias 2 triode in paralell,the same values with 1 triode? THANKS

[JimiB]

you half the value you would use for a single since it is shared.
ex: if you want 1.5K r for a single triode than a two shared triodes would use a 750ohm r to "bias" the same - hence the 820ohm cathode r seen in many marshals and fenders.

[jaysg]

If you're not tied into a pre-punched chassis, you use two inputs wired so that the Hi does parallel and the Lo does a single triode (the 6,7,8 section). Reference the Fender 5e8-a or the standard 18Watt.com 1974 schematic.

[keithrick]

I noticed on Paul Ruby's layout that the bypass cap is 47uf.

So do you need double the bypass cap as well to retain the same freguency boost?

[Andyjoe]

On the 5E8 and 18 watt wouldn't you need a y cord to get those trodes
in paralell.

Andy

[jaysg]

On the 5E8 and 18 watt wouldn't you need a y cord to get those triodes in parallel.

No and yes respectively, if you're talking real Marshall. That's not what we did with the 18Watt.com version.

[Roe]

I noticed on Paul Ruby's layout that the bypass cap is 47uf.

So do you need double the bypass cap as well to retain the same freguency boost?

I have the same question

[paulruby]

I noticed on Paul Ruby's layout that the bypass cap is 47uf.

So do you need double the bypass cap as well to retain the same freguency boost?

I have the same question

Yes, to make it percect...

Half the cathode resistance.
Half the plate resistance.
Double the bypass cap.

[gearhead]

Hold a sec, could you make that switchable with a DPDT? Ape the cathode res/bypass cap on one and plate resistor on the other.

Would just have to find one to take the plate B+ voltage.(?)

[riscado]

Yes you can, I've done that with some amps, including my 5E3, which I later returned to stock because I love just as is... The switch will take the voltage/current no problem, you should work a schematic in which you always have the cathode referenced to ground otherwise you'll have some "pop".

[gearhead]

If I'm drawing up correctly, then you're not substituting one component for another, but adding in parallel. in both cases.

What's the effect on the normally-unused triode half if you have the heater going, but no plate current flowing?