Qustions about Friedman Dirty Shirley filter cap

[psychepool]

I'm considering make Friedman Twin Sisters.
I heard it's an amplifier made up of two Dirty Shirley preamps, and whether that's true or not, I'm trying to make it that way.

This is the most visible and detailed circuit diagram I've found.

Dirty Shirley Schematic 1.jpg

In the circuit diagram, B+ and B+1 are marked as 473V and 471V.
It may seem like a somewhat high voltage, but it doesn't seem to be a problem.

The problem is that when looking at the actual photo of the amplifier, the filter cap assumed to be used in that position seems to be F&T 33uF+33uF 500V, but it looks too tight to handle the voltage of 473V.
Doesn't it make a big matter?
It is difficult to find a multi-can type electrolytic capacitor that exceeds 500V on the market. Would it be a safe choice to connect capacitors of about 350V in series like below?

cap.jpg

If there is someone who made Dirty Shirley, I would like to know what combination of B+ voltage and capacitor breakdown voltage they made.

Also, when making Twin Sisters, the number of vacuum tubes that must be handled in the preamplifier power supply is doubled compared to the Dirty Shirley circuit. Would it be okay to make the TS like Dirty Shirley circuit with 16/16uF?
Or would it be better to increase the value to 32/32uF?

[two tone]

I built a DS with JTM45 Iron,so my B+ is about 440v . No problem with those 500v caps.
Wouldn‘t try that with 470v on the plates though. 2 Caps in series is the solution, you may try 70mf/350v.
16/16mf is fine for the TS

[psychepool]

I built a DS with JTM45 Iron,so my B+ is about 440v . No problem with those 500v caps.
Wouldn‘t try that with 470v on the plates though. 2 Caps in series is the solution, you may try 70mf/350v.
16/16mf is fine for the TS

Thank you for sharing your experience.
I need to think a little bit about whether to choose to lower the voltage or connect capacitors in series.
I don't know if the voltage chart of the actual DS is like that, but I think it's a little unnecessarily high.

If I choose to series capacitors for high withstand voltage while maintaining the voltage on the chart, would it be better to apply it not only to B+ but also to B+1 (screen grid)?
Or, as in the power supply example posted in the text, if only B+ has sufficient withstand voltage, will 500V withstand voltage be enough for B+1?

[psychepool]

I would like to add just one question.
Dirty Shirley's PI circuit is almost identical to Marshall JTM or Plexi, but the grid resistors are not matched and has different values.
I think this is my first time seeing a circuit with different values here.
There is a double difference of 1M / 470K each. What is the effect if the grids in this pair of PIs have different values?

What difference will there be if this part is just unified to 1M like Plexi or JTM45?

[pdf64]

Steve Grindrod at Marshall used that arrangement back in 1980 with the Park Rock and Marshall Powercell (nicknamed the Baby), I think with the intention of reducing asymmetry in the LTP / power amp when heavily overdriven.
https://el34world.com/charts/Schematics ... 0_rock.pdf
https://el34world.com/charts/Schematics ... w_2150.pdf
I recall reading an interview with him somewhere, maybe this, it's quite entertaining https://guitar.com/features/interviews/ ... -designer/

[martin manning]

I recall reading an interview with him somewhere, maybe this, it's quite entertaining https://guitar.com/features/interviews/ ... -designer/

No mention of PI grid leaks, but that's a good read.

[psychepool]

I recall reading an interview with him somewhere, maybe this, it's quite entertaining https://guitar.com/features/interviews/ ... -designer/

Oh, this is very interesting. There was a hidden hero who changed the times. Thanks for sharing!

[pdf64]

It struck me that he may be largely to blame for the prevalence over the past 3 decades amp design of valve rectifiers being subjected to a hot switching standby.

[martin manning]

It struck me that he may be largely to blame for the prevalence over the past 3 decades amp design of valve rectifiers being subjected to a hot switching standby.

Nobody else there knew what they were doing, remember? ;^) I found it interesting that SG said his tastes in amp tone weren't always what bands or the public wanted.

[Roe]

Steve Grindrod at Marshall used that arrangement back in 1980 with the Park Rock and Marshall Powercell (nicknamed the Baby), I think with the intention of reducing asymmetry in the LTP / power amp when heavily overdriven.
https://el34world.com/charts/Schematics ... 0_rock.pdf
https://el34world.com/charts/Schematics ... w_2150.pdf
I recall reading an interview with him somewhere, maybe this, it's quite entertaining https://guitar.com/features/interviews/ ... -designer/

Yes it reduces redplating on v4-5 at full distortion. But other ways to balance the PI works even better I think, since the 390k changes the response for the worse, reducing punch somewhat

[pdf64]

... other ways to balance the PI works even better I think ...

Please expand on that

[Roe]

... other ways to balance the PI works even better I think ...

Please expand on that

Some things that do help:
- trimpot or plate resistors that are well-balanced (e.g. 91k/100k, https://www.ampbooks.com/mobile/amplifi ... iled-pair/). Obviously, the balance may change when you change PI tube
- increasing tail resistor
- NFB increases the balance untill you get full distortion
Also, matched output tubes matter.

A balanced driver gives a faster, punchy response (but may decrease distortion).