Messages

I've tried couple different ones for laser. The best I've tested so far I've bought from local arts and crafts shop. The brand is "Holi" and they are meant for scale models. The cost is ridiculous (like 5+ euros per paper or something like that).

All the no-name ones from eBay that I've tested so far are really hard to use. Either you need to soak it forever, or the laser "doesn't stick" when you apply clear coat / Envirotex or they rip up super easy and get wrinkled.

If someone know decent one from eBay let me know.

Hah, never seen one of these before. The board looks really good and well organized.

The L is the connection for the switching ground for the led. If its not connected the led wont light up.

I would suggest you to build audio probe: http://www.diystompboxes.com/pedals/debug.html

For me that is the easiest way to debug pedal that has no signal at all. In matter of minutes you will locate which is the amplifier stage that has signal at input but not in the output.

There might also be option called "fit to page". Take that off also.

Hey!

Got it! They had the CAM job ready for Eagle, so all I had to do was export the files and insert them to the controller program which turned them to gcode. Took 24 minutes to make the board and at least I think it turned out pretty darn good!



Sorry about the fingerprints.. 

Okay, I tweaked it bit more. Probably good enough now.



Now I need to start wondering how the CAM Processor works and what all the options do. I asked the guy who has used the CNC machine before about the files, but he just said "yeah, you need Gerber files". 

Does LED2 need to be installed for the circuit to work?

Thats just rate indicator, shouldnt cause any issue if not installed.

Alright, thanks!

I'm still wondering if I should move the trace that goes between the
legs of sustain pot. It could be easily routed under C2 and around the whole pot. What do you guys think?

I moved parts around at left down corner and now I think it's looking better:

btw. remember that madbeans pots are designed for boardmount, ie. on the BACK. if you have them on the component side and want to use board mounts too, you need to turn them

I was thinking I would do them boardmounted at the solder side (like any commercial PCB, the soldering would just be bit different). So are those currently the wrong way around? 

That's true. Probably wont have any kind of solder mask. Should I swap the transistor in Eagle for ones that have more space between legs?

looking good! minor thing, more of a personal perfectionism actually.
you can run the R5 to R3 trace horizontally instead of diagonal.

Good idea, will do that tomorrow. Also should move R2 and R4 bit to left so I can lower Q1 and ground plane fills that "hole".

I've heard that the 90 degree thing is for REALLY REALLY high frequencies, places where straight traces act like antennae and the frequency ends up polluting something else that the trace is pointing at. But that might be BS, you know?

Brian told me that some fab houses might actually have problems doing them properly as well.

Yeah, I asked couple of friends, who work lot with higher frequency electronics, and they both said it won't matter at all with audio frequencies. Well, I guess its still good habit to avoid those.

I updated the PCB with Felix's ideas, did you mean something like this?



Edit. updated image.

check your schematic too. it seems to me that your power section is wrong. 9V goes to D5 (protection diode??), but the other end of the diode is also connected to 9V via a resistor...

Good catch! Yeah, I thought I could do it like this:



Where +9V(T) would only mean the pad and +9V would automatically connect to other +9Vs in schematic, but that doesn't seem to be the case. Now it should be fixed.

the traces from R12 to R5/R16/CLR you can save the turn and run a straight line if you go from R12 to R5 rather than R16.

any reason for the increased leg spacing on C10? also if you move C10 directly under the right pad of D3 and align the lower pads of C9 and R13 with the lower pad of C10 you will make things easier too.
rotate Q2 90° to the right, this will also clean things up. also you can rotate R10 180° and reroute the trace that is now going from its lower pad to D1/D2

Oh yeah, good ideas all. No reason for C10, probably just picked wrong component while drawing the schematic. I'll post updated layout later..

Did some more moving around and increased the isolation bit more. This is surprisingly addictive.. 

John Roberts (one of the moderators there) was an engineer at Peavey and might actually be the guy who's responsible for the annoying PCB sandwich.

I've fixed couple of these and this is probably one of the most bizarre PCB designs I've seen. Works great most of the time, but when it doesn't (for example because of heat from the tubes) its goddamn annoying.