Bassman aa864 PCB redesign

[Davebassman]

Hi,

I want to build a Bassman AA864. This is actually the second one I'm building; the first one was built using an eyelet board, trying to keep it as close to the original as possible. For this one, I want to use more affordable components and a PCB. However, I have doubts regarding the B+ line. Is there any problem if this line is also on the same PCB, or is it better to use external wiring for this line?

The track width is 2.5 mm, and there is roughly 1.5 to 4 mm of spacing between tracks. I want to build it in the style of SINMARC, where the copper tracks are on the top side, and the components are soldered directly onto this side.


Any advice will be welcome.

Thanks.

[loudthud]

Where are connections to the tube sockets, pots and ground ?

[Davebassman]

I haven't drawn it yet. When I have the complete layout, I will upload it. For now, I’m interested in knowing if it’s a good idea to have the B+ line on the same PCB or if it’s better to use external wiring for this line.

But you can get an idea—on the PCB, you'll see V1_1, etc., which refer to the tube sockets.

[loudthud]

Where are the filter caps being located ? It's more important to get the ground right on those caps, than any wire you might save snaking a B+ trace around. It doesn't make much sense to snake a B+ trace around the board if you have to run wires for the filter caps off board. Any wire you run off the board is a potential failure point. I use an eyelet (small ones) on any off board wires, and most of my boards are solid state.

See post 32 in this thread: https://music-electronics-forum.com/...r-almost/page3

[xtian]

I've used PCBs from a variety of trusted makers, and never had an issue with B+ on the board.

[Chuck H]

Indeed to all above. A big change for Marshall amps was when they switched to using PCB's ahead of Fender. And they DID include B+ on the board. They did it well enough in those earlier years that it was never an issue. With adequate trace size and layout consideration I don't see how it would be a problem. You might even look at how Marshall did it in those early PCB amps for some ideas of how to succeed in the effort.

[Pixel]

The Randall RD5C I recently recapped had the SS diodes and caps for B+ on the main PCB though it did go back and forth via faston connectors to a separate PCB for the power tube. AC heater for the 6V6GT on the standalone PCB and DC heater for the 12AX7s on the main board. They left enough copper on the PCBs so I could drill more holes to fit a couple caps that had slightly bigger footprints than the originals. I do wish they had left a bit more space between components for cooling though.

I don't understand why one would choose to put through hole components on the copper side. I only do that for SMDs or to fix boo boos.

[loudthud]

I don't understand why one would choose to put through hole components on the copper side. I only do that for SMDs or to fix boo boos.

It makes it possible to remove and/or change the component without having pull the board.

[Pixel]

It makes it possible to remove and/or change the component without having pull the board.

Cool. That makes sense.

Looking at the SINMARC pcbs has me thinking I will one-off a caveman style pcb for a 5F2A using electrician's tape for resist. I have LOTS of pcb material. I'll still go through hole though.

[Davebassman]

Hi,

This is the layout. Obviously, the mechanisms and transformers are missing, but at least you can get an idea of how it will be assembled. I have doubts about the GND scheme and, as I mentioned in the previous post, whether it’s an issue that the B+ trace runs on the same board and passes under a component (I’ve uploaded an image). Keep in mind that there will be no solder mask, and the copper traces will be on the top side. Any improvements, corrections, or suggestions will be more than welcome.

Thanks.