Layout design and positioning

[AleS]

A quick couple of questions, to which I've not been able to find an answer through searches.

Is the positioning of each preamp and the PI within the layout crucial?

The original 18w had the PI in the middle of the two (ditto the valve), while the Modern Classic moves the PI next to the power section. Since the MC is effectively an improvement to the layout (and in a way design in its differences), is this swap an improvement and if so why?

The PI's B+ supply should be grounded with the power section, but that happens off to the side anyway so the proximity to the grounding isn't necessary. Only advantage I can otherwise see is perhaps a shortening of the signal wires, but could this not be mitigated by using shielded wire from pins 1 and 6 of the PI ECC83, aka V2 in the original layout?

Secondly, can B+ wires (including what would often be called C, D, E etc, ie the filtered supply for each stage) cause any noise? Many layouts (including the MC, but also all Ceriatone Matchless layouts for example) have filtering all done together on the side with the supplies then shared by underboard wires. Is this safe noise wise or is there anything to keep in mind when doing this?

Thank you very much in advance!

[Bieworm]

Well history has proven there are many ways to lay out an amp. But …
There are a lot of things that the modern classic has that are closer to common sense and good practice.
Putting the PI next to the power tubes not only makes more sense in the logic of the circuit, but also shortens wire runs, has less wires crossing each other, makes the grounding layout easier to separate preamp from power amp,…
As for the filter caps: it’s just good practice to place the filter cap close to the part of the circuit it’s serving. Also avoiding long runs of B+ wires.
In short, the MC is just more logic and good practice all over
And thank you Josh for agreeing on that with me and having redrawn the classic into the modern classic

[JMPGuitars]

Bieworm has the general idea correct. However, it's also a best practice to keep the filtering cap next to the final cathode cap for a given stage.

Also, wires should generally not be run under the board. It's that way in the documents sometimes to make seeing things easier...like the notes on the layout say.

[AleS]

Thank you both for your input, it's truly appreciated. It seems to confirm what I suspected - things *can* work in many ways, but some are more trouble-proof than others, so why not stick to those...!

Reason I asked is because I'm trying to design a layout for a build which starts as a lite II variation, but has Ch1 as the classic Volume and Tone channel (with the triodes in parallel) and Ch2 as the Vox style EF86 one.

To have more control over voltages (especially on the EF86, where it seems a bit of experimenting pays) I'd like to stick to the original Vox (then stolen by Matchless) power supply/filtering network of parallel 22k/22uF network for each stage (Ch1 Pre, Ch2 Pre, PI and PA), which is a couple more than the standard 18w.

Would the attached (which of course is missing a lot of connections and is very much a WIP, but should provide a good example of grounding and component placing) be a good, clean way of going about it?

Many thanks once again!

[JMPGuitars]

You should start with a schematic before a layout.

You have your PI and preamp grounded together (bad idea), and it would be better to place the filter caps after the cathode caps for each stage.

You should also only use between 10k to 33k for your grid stopper resistors.

the grounding thread in my signature.

Thanks,
Josh

[AleS]

Thank you Josh. I do have a schematic I'm working off of, just wanted to run a few eyeballs past the layout I'm working on - just as well I did!

That was an old one but even back then I'd have known the PI filtering should go to the power ground, thanks to your very helpful thread and valve wizard's website.

In terms of filter caps placement after the cathode bypass caps would it be a case of having them on the right hand side to put it crudely?

On the point of grid stoppers - would lowering the value result in a lower HF rolloff point to mitigate potential buzziness/high freq overload especially with distortion? As I'm not working towards high gain I stuck to the original AC15 values (again, also copied by Matchless) but can see the HW version of the Vox uses 10k...

[JMPGuitars]

That's probably fine.

Merlin answers your grid stopper question: https://www.valvewizard.co.uk/gridstopper.html

Phil had additional thoughts on the subject, but I don't remember right now.