LA2A PCB

[Deepdark]

Hi all

here is a first attempt for a la2a pcb, true to the original schematic. i would like to have your opinion about the design, some advice, things i should change, or whatever else. i don,t run the heaters yet, i don't know if i could run in some hum / oscillation trouble if i run it on the board. i made it on a 2 layer board. top layer is all the audio and bottom, the gnd plane and ht. i made 4 gnd plane ( input, output, audio and psu).

thanks

 

[Deepdark]

I'll rework a bit the the 12ax7 at the bottom left side. For the rest, i tried to keep the traces as straight and short as possible.

 

[snaper]

AS a self etcher, would be great to done this in single sided, so "cheap-guys" like me could etch at home

 

[Deepdark]

Yeah, but the goal here is to make a kit once tested

 

[r2d2]

Yeah, but the goal here is to make a kit once tested

I Agree 
r

 

[Deepdark]

I would like to know if running heaters on boards is safe or if i run the risk of instability (hum, oscillation, etc). Also, i did run my ht trace, from the psu to the anode at 0,050  and all the other races at 0,030. It's probably a little overkill. What's your thought?

 

[dustbro]

I run all my heaters on the top layer, along with my ground plane. It has worked well for 15+ designs at this point.  Low noise.  No hum

 

[Deepdark]

ok. so you're signal traces are all on the bottom layer. Is there a special reason for that for it's just the way you always did it? Do you ground place both side or just top side? and your heaters, do you run them straight, from tube to tube or your run them along the border of the pcb?

 

[keefaz]

I would like to know if running heaters on boards is safe or if i run the risk of instability (hum, oscillation, etc). Also, i did run my ht trace, from the psu to the anode at 0,050  and all the other races at 0,030. It's probably a little overkill. What's your thought?

How much for clearance between tracks?

 

[Deepdark]

don't remember how much but I tried to run traces no that close of each others so thsi shouldn't be a problem.  Do you have some recommandation about spacing between tracks?
the test pdf of this thread isn't accurate. I redraw it a lot and the design change.

 

[zamproject]

i made 4 gnd plane ( input, output, audio and psu).

hello

In fact you have 5, your "audio" plane is cut in half !

Zam

 

[Deepdark]

Hi sam,

Yes, you're right, but since I modified all the layout, this one isn't applicable

 

[joaquins]

C7D and C7C should be close to where they are connected rather than in the PCB ground plane, this will make a shorter current path and reduce the ground plane cuts. The trace from C7B could be almost entirely routed on the other layer to avoid cutting the ground plane as well. I think you should try better component placement, then is much easier to get a better layout, you have a few traces going all around places that I don't think they need to, try to place the parts as close to where they need to be connected so you keep traces short. Trace impedance is not a big deal in tube stuff but they can pick up trash and the capacitance may also be a problem. I don't know the clearance you are using but maybe increasing it a bit would be safe, specially in the HV parts of the circuit, as PS and anodes.

JS

 

[Deepdark]

Thanks Joaquin. Actually, I already moved a lot of components, shorten the runs, etc. C7C and C7D are close to where they connect, and the psu is on an external pcb (I like working that way). I even change the psu a little. I draw a RC filter network with 3 1k5 résistors and 3 47uf capacitors. This should stabilise even more than the original design.

Actually, when I first draw it I was inspired by the way Drip did manage their power caps, all side by side at one end of the pcb. But I finally decided to move C7C and C7D closer to their respective location.