New to DIY (PCB DESIGN)

[akamoosh]

Hello all,

As you can tell from the title I am new to DIY audio equipment. I have a question about PCB design. I want to build my own Tube Tech CL1B compressor. I have the actual schematics, but the schematics don't provide the design of the PCB (size, dimensions, spacing, etc). I did some research on if the actual design of the PCB influence the the sound quality of the audio device , and to my knowledge it does. So how would one design a PCB to behave identical to the PCB inside of the audio equipment I want to build, if I don't know exactly how the PCB is in the original? If im going about this all wrong please can you point me into the right direction.

 

[abbey road d enfer]

Hello all,

As you can tell from the title I am new to DIY audio equipment. I have a question about PCB design. I want to build my own Tube Tech CL1B compressor. I have the actual schematics, but the schematics don't provide the design of the PCB (size, dimensions, spacing, etc). I did some research on if the actual design of the PCB influence the the sound quality of the audio device , and to my knowledge it does. So how would one design a PCB to behave identical to the PCB inside of the audio equipment I want to build, if I don't know exactly how the PCB is in the original? If im going about this all wrong please can you point me into the right direction.

Good practice for audio PCB's is making sure the lay out obeys the rule "ground follows signal", minimizing trace length, minimizing capacitive coupling between sensitive nodes, but "vintage" PCB's may not have had the same rules, because production methods had limitations, so you may find that a newly designed PCB may be objectively better than the vintage one, but would not sound exactly the same.
In the 60's and 70's, two-layer PCB was expensive, so many equipment had single-sided with a lot of strap links. You wouldn't do that today, because 2-layer is the same price as single-sided.
Many clones of iconic equipment have PCB's that are very different than the original, and still sound good. The end result depends much more on the quality of components .

 

[rock soderstrom]

I did some research on if the actual design of the PCB influence the the sound quality of the audio device , and to my knowledge it does.

If your PCB design has no major flaws, I doubt that there will be any audible differences. There are some variables that influence the sound result much more.

A good example would be the transformers. The output transformer for example also has a tertiary winding for NFB, where do you get that?

Your project is certainly not impossible but not particularly well suited as a first DIY project.

I strongly advise you to choose a smaller and better documented project to start with, otherwise it can get frustrating very quickly.

Please also consider safety, tube projects have high voltages that can kill you even if the device is switched off!

 

[akamoosh]

Good practice for audio PCB's is making sure the lay out obeys the rule "ground follows signal", minimizing trace length, minimizing capacitive coupling between sensitive nodes, but "vintage" PCB's may not have had the same rules, because production methods had limitations, so you may find that a newly designed PCB may be objectively better than the vintage one, but would not sound exactly the same.
In the 60's and 70's, two-layer PCB was expensive, so many equipment had single-sided with a lot of strap links. You wouldn't do that today, because 2-layer is the same price as single-sided.
Many clones of iconic equipment have PCB's that are very different than the original, and still sound good. The end result depends much more on the quality of components .

Interesting, That's a reliever actually because I thought I couldn't continue the project if I didn't have the exact PCB. Where does the one acquire this knowledge about PCB's? (I know the internet) but can you point me to a legitimate source? Thanks for your time!

 

[akamoosh]

If your PCB design has no major flaws, I doubt that there will be any audible differences. There are some variables that influence the sound result much more.

A good example would be the transformers. The output transformer for example also has a tertiary winding for NFB, where do you get that?

Your project is certainly not impossible but not particularly well suited as a first DIY project.

I strongly advise you to choose a smaller and better documented project to start with, otherwise it can get frustrating very quickly.

Please also consider safety, tube projects have high voltages that can kill you even if the device is switched off!

yes I know this will be challenging. I do have a background as I completed training classes for being an electrician, so some things I understand. Im willing to put the time and money to learning and building this project.

 

[abbey road d enfer]

Where does the one acquire this knowledge about PCB's?

What about 20+ years of practice?

(I know the internet) but can you point me to a legitimate source?

I don't think there is anything like the "Hitchiker's guide to audio PCB's".
Audio PCB's are easy... ... to make wrong.
Easy because there are none of the constraints of high-speed circuitry, like trace impedance and standing-wave reflection.
As I mentioned, the most important thing is to make sure ground follows signal, or in other words how to make sure dirty nodes are not allowed to pollute sensitive ones.
Ground planes are not a necessity for audio. Better have no ground plane than one that inject rectifier current into a mic pramp.
One must consider that every trace is a resistor, particularly the ground traces.

 

[akamoosh]

What about 20+ years of practice?

I don't think there is anything like the "Hitchiker's guide to audio PCB's".
Audio PCB's are easy... ... to make wrong.
Easy because there are none of the constraints of high-speed circuitry, like trace impedance and standing-wave reflection.
As I mentioned, the most important thing is to make sure ground follows signal, or in other words how to make sure dirty nodes are not allowed to pollute sensitive ones.
Ground planes are not a necessity for audio. Better have no ground plane than one that inject rectifier current into a mic pramp.
One must consider that every trace is a resistor, particularly the ground traces.

ok understood. thanks for the advice. This will take a lot of work. But i'm willing to do it. I guess I got to just start and whenever an obstacle presents itself, then I'll learn from it. but again thanks!

 

[gyraf]

Actually, CL1B is still in production - I'm sure it will be easier to just buy one from John. And we don't really support copying of current manufactured designs.

As for cloning the unit, if you don't have access to the original transformers (and photocell?) it will be very hard to make it sound and act like the real one. Those are OEM parts, you won't find them in the wild other than as original spare parts.

You are not the first one starting this thread - try searching the forum for CL1B

/Jakob E.

 

[hghadia]

Hello all,

As you can tell from the title I am new to DIY audio equipment. I have a question about PCB design. I want to build my own Tube Tech CL1B compressor. I have the actual schematics, but the schematics don't provide the design of the PCB (size, dimensions, spacing, etc). I did some research on if the actual design of the PCB influence the the sound quality of the audio device , and to my knowledge it does. So how would one design a PCB to behave identical to the PCB inside of the audio equipment I want to build, if I don't know exactly how the PCB is in the original? If im going about this all wrong please can you point me into the right direction.

To replicate the performance of the original PCB without detailed specifications, focus on replicating the circuit layout and component placement as closely as possible based on the schematics. Pay attention to signal paths, grounding, and component types. Additionally, consider consulting with experienced DIY audio enthusiasts or forums for insights and advice on optimizing your PCB design for sound quality.

 

[abbey road d enfer]

The main thing that affects performance in a PCB layout is the circulation of currents, particularly in the ground.
Although the different ground points in the schematic are apparently identical, in practice they are not, because of the resistance of the tracks.
OPtimizing the circulation of currents on a PCB is a task that demands expertise and analysis.
A number of commercial products are not exempt of problems in this respect.

 

[jokeramik]

PCB design is a complex task due to the interaction (electrostatic and magnetic) between the individual components. For example: avoid right-angled bends in conductor tracks and make them rounded. This avoids peaks in electrostatic fields and their interactions with one another.
Peter Baxandall and Quad probably created around 100 test prints and test setups until a legendary power amplifier delivered the performance that was required. Do you have equipment to measure changes? The following recommendation for PCB design, among others: Prof. Dr. -Ing. habil. Werner Krause “Device Construction” Verlag Technik Berlin and others.
Best regards!

 

[Cqwet Dbdfte]

Today 4-layer boards are not much more than 2-layer, check out JLC, Pcb GOGO, and others.
I use ground plane polygonal pours on all layers, RF circuits or not. Second layer is ground only,, no signals. As I use mostly SMT parts, I try to keep those on one side, and thru hole on the other, which gives room for SMT parts under thru hole parts.
Place your main signal circuits and SMT's on '"top" , and control and data lines on layer 3 and bottom. With lots of traces, run them in one direction per layer,, like horizontal on one and vertical on another, or you may run out of space.
I use massive amounts of ground stitching vias along signal lines, while making sure other vias do not interrupt ground continuity.
Name all your nets, helps a great deal when it gets messy.
I use aluminum 6-32 standoffs to mount boards, and I use grounded vias for these, and mount boards on aluminum surfaces.
While this is RF centric, it does not hurt for audio, which can become RF sensitive, and some parts may have frequency response up in the UHF region, requiring wide band test equipment to debug, and lead to "sonic" conundrums.
Put part designators where you can read them after installing parts if possible.
When connecting traces to thru hole pins never connect on inner layers. If you have to, make a short surface trace then a via. May help if you need to isolate a pin.
Assign board connector pins according to shortest traces, not numbers in a spread sheet, if possible.
Round traces with a radius at least 3 times the trace width to reduce impedance discontinuity. I also make all my rectangular pads have rounded corners. I have gone thru every footprint and rounded each pad a bit. This also helps cleaning solder paste stencils (use electropolished if possible.)
Use a separate program to view you Gerber files, this will reduce clutter and make it easier to find errors, which the ERC may not find, like part label in the wrong spot etc.
"Gerbv" works OK.
One sided aluminum PCBs have become very low cost and can be used for heat spreading and for mechanical tasks , even front panels, with nice silk screened labels etc.
For foot prints requiring a ground pad I always include ground vias in the footprint file.
Larger SMT pads I break up the solder paste into "windows" to make the paste spread more evenly.
You can order arrays of boards for not much more and they will make the arrays for you, V-cuts are OK, may need a touch of a file to knock off edges.
There are many Utube tutorials that are worth looking are worth looking at.