See if you can tweak some alignment. Make it look a tad prettier
From experience, sometimes placing electrolytic caps too close to other components has led to complications.
What trace widths are you using? I tend to go parallel to Bean's PCB layout tutorials and go with 18 mils for voltage rails and 10 mils for signal traces. Sorry I'm doing it in freedumb units, but I haven't switched over to metric for this yet.
Some of those comments might not work depending on your footprints; e.g. (using Bean's libraries) I'm used to 1/4W resistors matching up to one rank of a, 8-PDIP package.
Those are just knee-jerk thoughts, and I'm a bit tired, so
@rwl Looking good in general. Mostly aesthetic preferences, though I am no EE!
I would look at some of the placements around the IO and pots. For example, in the compressor, there are high standing capacitors around the Makeup and Attack pots. That may be a little annoying to solder.
Similar with the IO pads at the bottom. There's a cap right next to it which limits the angles you can solder from. Maybe not an issue for you, but that sort of thing has annoyed me enough from time to time.
If you're doing dylan159 designs, remember the license that he's released them under. You should be making any works derived from his designs open source and available. Whether you actually do or not is your choice of course. I have seen others declining to honour this agreement, and it runs against the spirit of what is being done in the community.
@jessenator - You know, it didn't occur to me until you marked up the layout that it would be a lot easier to keep things aligned by drawing some helper lines in an unused layer in the editor itself!
I'll keep this in mind. I don't mind having something tall to one side of the pots, but I do really hate if the pads are tightly enclosed on two sides.
Yes, my intent was to share them. I'd like to build up a collection of pcbs on github similar to yours, once I've validated the layouts work. I've had most of your Dylan159 layouts printed and really like all of them (except the Clown Centurion), so I figured I would complete the collection. I also find Dylan's schematics very easy to read.
---
For reference, after the feedback, this is what I ended up with, and am going to get fabricated.
Compressor:
And Phaser:
---
I might throw out a few cases I encounter, where I'm not sure about whether there's any functional difference. Here's a starter, the blue trace is coming from below the screenshot.
A. Wrap around (at minimimum clearance):
B. Wrap around (add padding):
C. Cross over:
D. Makes no difference.
Thoughts? I've tended to opt for B if there's space on the board since it feels "safest" (farther from other components), but it's also the longest trace.
In a case like this I would modify the schematic so the blue trace connects to lug 3 and the red trace connects to lug 2.
As far as routing the traces as-is, I would look at the surrounding traces and determine which of the two in question (red/blue) would benefit more from the clearance.
I suppose that guidance only applies for when lug 2 and only one of the other lugs are populated, since basically the only important thing is that the two traces should be connected somehow when the switch is flipped
I know you wanted feedback, so/though I'm going to suggest having a switch to cancel the feedback... (I vaguely recall this is a popular mod, to disconnect the feedback path).
Just in case you haven't read it, do read "The technology of Phase Shifters and Flangers" by RG KEEN.
You said you sent these off to be fabricated, but did you check to make sure your DRC passed after you revised the boards?
Those 10K resistors (vertical oriented next to pins 1-4 of the TL074) may have their pads merged (not sure if this was intentional). Additionally, the trace leaving the vertical 10KΩ on the left has zero clearance to the uppermost horizontally oriented 10K resistor.
I don't claim to be an expert on these things, but why cram so much into such a tiny space? You have the real estate to give your traces and pads a little more breathing room.
@Brett - here's what it looks like close up - it seems to meet the exclusion requirements, it's just hard to see in the zoomed out screenshot.
Your broader point is correct, I could make better use of the space... I got tired of moving the larger blocks of components around every time I shuffled the layout, then having to reroute all the connecting traces to see if it works out. I see now KiCad offers a "drag" functionality (hotkey D, can't have traces selected) that rubber-bands the traces outside the selection, but it's still pretty janky.
That looks really neat. I wish I could let go of the symmetry and arrange boards this way. What's the circuit? I like your staggered pots in the top row, I'm doing similar on my designs.
That looks really neat. I wish I could let go of the symmetry and arrange boards this way. What's the circuit? I like your staggered pots in the top row, I'm doing similar on my designs.
Thanks! It's a jfet preamp into a Si FF output stage using low gain soviet transistors with a foot-switchable dirty boost in front. Toggles are for the coupling cap into the FF stage and a low pass on the output. Waiting on enclosures to arrive from Tayda them I'm going to send some out to a group of testers and start selling them if the response is good, otherwise back to the drawing board.
I haven't got experience in it, but I saw some comments from old hands over at FSB talking about isolating the LFO in one part of the board and using a separate ground plane, only joining where the negative input comes into the board.
Exactly. I usually separate the audio/LFO ground planes and power traces. In extreme situations I will hand route the ground paths to control current flow and divert noise away from the audio path. I do the same with any digital components.
Amazing, thanks. Your ground plane shaping there is blowing my mind, I'm still relying on "snap to board outline" I haven't needed to do any separations of ground planes yet, but I'm working on an Ultravibe and a modded Phase 90 that I'm guessing will need more careful planning.
In the meantime, here's a reasonably symmetrical true bypass Klone:
The second part of my Mayer-inspired Christmas gift for my daughter. It's heavily based on the Merman/Sea Monk. I used my symmetry cheats to note the board design's month and year.
You layout work is always tops, @Robert! I looks like you're connecting power and ground to the audio portion of the circuit on the backside of the board. If you're willing to share (and I understand if you aren't), are you also isolating the copper pour on the back of the board like you did on the front? How did you handle joining the ground and power planes?
I haven't needed to do any separations of ground planes yet, but I'm working on an Ultravibe and a modded Phase 90 that I'm guessing will need more careful planning.
