Out with it

Well, finally.

An output module. It takes four inputs, attenuates them, pans them (manually) in the stereo field, and has a stereo pair of line outputs and a stereo headphone jack.

This thing has been such a pain in the ass for such a long time. Design changes, mechanical design stupidities, PCB layout errors, parts availability, soldering problems, unsoldering problems, you name it, if it didn’t involve catching fire it probably happened.

Originally I was going to build the Barton Stereo Outs. Then I decided there were changes I wanted to make to the design. Most notably, I didn’t like the headphone amplifier and wanted to replace it with an LM4808. SMD, but only one part and not that hard to solder. I’d done similar size SMD chips before, I did an LM4808 on a breakout board for breadboarding, it would not be a problem.

It was a problem.

I chose to use 60 mm sliders, but that left room only for 9 mm rotary pots for panning and headphone volume. And those all are dual gang pots. You know where you can get dual gang 9 mm board mounted vertical pots? Thonk (in England), Pusherman (in England), and possibly Asgard. They’re not that cheap and postage is not that cheap. Had I used 45 mm sliders I could’ve used 16 mm rotaries from Tayda. But noooo.

I assembled the thing, tested it, it didn’t work. But for some dumb reason I’d designed it as three PCBS: Rotary pots, sliders, and jacks plus main circuit. Why I chose to put the sliders on a different level from the jacks I don’t know; they’re not as high, if you want them up against the back of the front panel they have to be soldered on a different level, but if you have them up against the back of the front panel the shafts stick out too far anyway. On the DC and Audio Mixers I had them on the same level as the jacks and that was fine.

But I didn’t do that here, and while they are on a different level they’re not much different, so I couldn’t connect the boards with headers plus sockets, so I chose a different (stupid) way that involved soldering the boards together — with the back side of the main circuit permanently underneath the sliders PCB. Impossible to diagnose and fix without unsoldering that connection.

I found part of the problem was the LM4808 soldering so I redid that. Twice, I think. Another problem was a 17 pin header/socket linking the main board to the pots board, of which I’d actually soldered 16 pins. The one I missed was ground. That caused trouble.

Eventually I got everything working except for one thing, and I re-connected the sliders PCB with the main PCB, this time hacking up a connection with a ribbon cable wedged between the two. The one thing not working was I’d gotten the pan pots connected backwards. <Takes bow.> Yes, you had to turn the pots left to pan right and vice versa.

So I got a corrected pots board made. Not wanting to throw away those expensive dual gang pots I tried unsoldering them, successfully… for four of the five. Unfortunately I had zero spares, so it was back to England again for more. They arrived yesterday, I assembled it today.

It’s kludged together in a horrible way due to said mechanical design stupidities; if I hadn’t gotten this one working I would have entirely redesigned it before trying to build another. But it’s working, it’s together, it’s in the case. And yes, I misspelled “PAIN” on the front panel.

Schematics and BOM are in this repo, but PCB design files and Gerbers are not. Make better ones yourself.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s