Designing Own Layouts - Please Judge My Work

[karis12]

I've been doing DIY pedals for close to 5 years now, but up until now had worked only on relatively simple builds. Due to limited knowledge regarding the whole engineering aspect of this hobby in the first place, I've mostly worked with perforated board with individually isolated copper holes.

As my concern is often to make the circuits as compact as possible, I often end up with layouts looking like this:

Silicon Fuzz Face (with BC548 transistors)


Paul Cochrane Timmy

So far, the only issue I've come across with is my first layout of the RunOffGroove Thor, which had a feedback problem that seems to be affected by the bright switch and gain knob. I am suspecting it may either be due to the positions of the FETs, or the fact that I didn't check the Vgs or Idss of each FET before building. Here's what it looked like:


I redid my layout, and also measured the FETs I used the next time around, even tweaking the values of the source resistors on some positions, and came up with this, and this time, it worked pretty well:


What do you guys think?

[GibsonGM]

I see nothing at all wrong with your layouts, Karis.  Those are much neater than the circuits I throw together! 

[karis12]

I see nothing at all wrong with your layouts, Karis.  Those are much neater than the circuits I throw together!

Thank you GibsonGM! My reason for starting this thread is to see if anyone could tell what potential issues my layouts might cause me down the line (tight spacing of components, standing resistor bodies, etc.) as I'd like to try my hand designing high gain/more complicated circuits next.

[EBK]

I'm going to chime in just to say that I love the challenge of making layouts as small as possible too, but I would suggest a slight change.  Instead, figure how small you can make something, then come up with what size enclosure you want to use, and then design the board to fit the enclosure.  The reason I suggest this is that it could allow you to (utilizing otherwise wasted space) incorporate some better layout practices to reduce noise or more thoughtfully route your off-board connections.  I'm not saying there is necessarily anything wrong with what you have shown though.

[GiovannyS10]

Hyia Karis!
Your layouts seems me very good, want you do some for me? haha. Really, i know how hard is do some layouts, but yours seems me very nice don't worry about it.
A layout is not like a puzzle, but is like an equation... You have a lot of ways to find the same result. Finish a layout is good, turn it small is a challenge. I am in one right now, to adapt my actual layouts for 1590lb enclosures, let's go to see if i will can. 

[rutabaga bob]

I like trying to figure out layouts...well, sometimes!  I've done a lot of layouts that have never gotten built - nor are they likely to.

Sometimes I have what I think is a nice, useable layout, and it does work, but the assembling of it turns out a little (or a lot!) more aggravating than I expected.  I see a lot of 'standy-uppy' resistors in your photos...while I like the compactness these can give, I always seem to have an awful lot of trouble keeping them in position 'til they can be soldered.

Cheers!
Larry

[duck_arse]

a way to practice is to get a bunch of 24x10 perfs, or the 28x6, and then lay out a circuit to fit, sticking to a set of rules. like 1/8W resistors all bent for 4 spots, or 1/4W's all bent for 5 spots, and all resistors layed flat. and no flying links.

but there is nothing wrong with your layouts. your layouts - your layout rules (ok?)

[EBK]

Am I the only one who doesn't think resistors standing up is a bad thing?  That op amp feedback resistor between two adjacent pins will always stand up in my layouts.

[rutabaga bob]

I don't think they are a bad thing...I just have more aggravation with them than laying them flat.

[GibsonGM]

Am I the only one who doesn't think resistors standing up is a bad thing?  That op amp feedback resistor between two adjacent pins will always stand up in my layouts.

At the frequencies we deal with...20Hz-20kHz....I cannot imagine a scenario where standing-up resistors will make a noticeable difference to circuit response.     And lead length simply isn't as critical as if we were messing with GHz, microwave stuff!   

When we get into high values of amplification  (think fuzz face, distortions and so on), using shielded in/out wires, wire routing, and thinking about RF getting and being amplified do come into play.    So, trying to make short runs to ground, setting up a star grounding plan, avoiding crossing wires/components, and if necessary to do so to do it at right angles...those rules start to have more effect.   Once in a while, it becomes less about 'neat' and more about 'what is going where'.   Figuring this out just takes a lot of builds, mistakes, and debugging

LFO wires can be notorious for coupling into audio paths/high impedance inputs....that stuff.

[duck_arse]

I didn't say they was bad. I used to include standing two spot, and standing three spot resistors in my vero rules. but then I moved on to something more challenging, all flats.

[edit :] sat? that makes no sense.

[EBK]

I don't think they are a bad thing...I just have more aggravation with them than laying them flat.

Sometimes I'll tack one end of a standing resistor (or other hard-to-solder part) with an intentionally bad solder joint just to wiggle it into the position I want it.  I then correctly solder the other lead before coming back to fix the temporary joint.

[GibsonGM]

I didn't sat they was bad. I used to include standing two spot, and standing three spot resistors in my vero rules. but then I moved on to something more challenging, all flats.

No, no one thinks you did, bub!    It's all in the technique....I bend the leads over opposite each other on the copper side, 90 degrees.  That holds them...only 1 at a time, tho.

[ElectricDruid]

Am I the only one who doesn't think resistors standing up is a bad thing?

No, definitely not. You only have to look at the number of commercial pedals that do/did it to see that it's fine.

My own method these days is to develop a circuit on a stripboard prototype, which is usually massive and very spaced out so I have lots of room to add things as I go along and adjust stuff which isn't right. Example:



Once I like it, I design an actual PCB layout. So these are not issues I have to deal with, thankfully.

Tom

[EBK]

Here's an example of what I meant by designing the layout with the enclosure in mind.  This is my compressor pedal, with 3 boards in a 1590BS (same footprint as a 1590B, but thicker like a 125B).



The large board is the compressor itself.
Underneath that, wrapping around the stomp switch is a board with an Arduino Pro Trinket microcontroller that sends data to the bicolor LED bargraph driver board, which is partially visible beneath.  That thing I'm lifting up in the second pic is a ground plane made from copper shielding tape sandwiched between layers of gaffers tape, by the way (essential to block digital chatter from the analog board).

[karis12]

Thanks for the comments and tips everyone! I'm preparing to build a fetzer valve version of a high-gain amplifier, and your examples will definitely come in handy when I begin prototyping.

[nachus001]

Here's an example of what I meant by designing the layout with the enclosure in mind.  This is my compressor pedal, with 3 boards in a 1590BS (same footprint as a 1590B, but thicker like a 125B).



The large board is the compressor itself.
Underneath that, wrapping around the stomp switch is a board with an Arduino Pro Trinket microcontroller that sends data to the bicolor LED bargraph driver board, which is partially visible beneath.  That thing I'm lifting up in the second pic is a ground plane made from copper shielding tape sandwiched between layers of gaffers tape, by the way (essential to block digital chatter from the analog board).

Hello EBK

would you tell me what's the paint in this box? Brand/type will be welcomed

regards
Nachus

[EBK]

The box is a Hammond 1590BSBK.  All the Hammond "BK" boxes are textured black.  The bezel around the bar graph has the same texture because I made it from an old 3.5" floppy disk (the window in the bezel is made from the vinyl sleeve the disk was in, by the way).

[EBK]

I should add that Hammond boxes are powder coated rather than painted.

[nachus001]

Okey

thank you

Nachus