X1Plus Expander fellows,

They say that "no news is good news" and gosh, these days, when I’ve been reading the news, it really feels like it. But that said, I’m given to believe that crowdfunding backers do like to hear news about the projects they’ve plunked down money on and it’s been over a month since I last wrote you all.

If you don’t have time for the rest of it, the short answer is: I’m basically on track to get you Expanders, and roughly when I expected to, also! And if you want the long version, well, read on.

In Which Boards Keep Rolling In

A few weeks ago, another box arrived from King Credie — this time, with production samples of the Expander modules! I panelized them with KiKit to get the right balance of the different designs on a panel to match what y’all ordered and, in total, 33 modules arrived on my desk. I was a little nervous about this and, in particular, I was nervous about how the V-cuts would split (do the board houses know something I don’t when they design panels?), but to my relief, they all looked good, and, more importantly, they worked right on the first try. It was a pretty big relief also that the new Neopixel-compatible JST PH connector on the LED level shifter board had the right pinout, and I hadn’t, like, swapped it upside down or something.

Anyway, I hand-tested each one in an Expander board with some haphazardly-thrown-together scripts and the yield was 100% (hooray!). I programmed all their EEPROMs, printed labels for them, and then stuffed them under my desk and wired money over to King Credie for them to go ahead and make the 637 modules that I owe Crowd Supply.

In Which Cables Land with a Thunk

In other exciting DHL arrival news, there was a solid WHAM of a box on my front doorstep that was absolutely packed to the gills with cable harnesses! Oseisy sent me 480 of them to get started with and then will send me the rest in a few weeks. It is so cool to see volumes of things rolling in. I pulled one out of the packaging, and tested it in situ on my printer. It fits nicely!

In Which I Get the Shaft (Twice)

You may recall that just before I went on vacation recently, I got the pogo pin test jig up and running for the Expander mainboard. I thought I was so cool for designing with pogo pins! Unfortunately, the idea of holding the board in a plane using 3D printed posts really did not work all that well — it did not slide well on the posts, and also, there was enough play that it didn’t reliably come down exactly parallel with the test plane. This was something I thought I’d just have to be careful about, right up until I wiggled a board on my desk such that the ground pogo pin came disconnected before the +24V pogo pin came disconnected, at which point I blew up the board.

And the test jig.

And a port on my USB hub.

Oops.

I decided this clearly would not do, so I ordered some metal alignment dowel shafts from Aliexpress and redesigned the entire jig around those, hoping to improve my tolerances there. A week or so later, they arrived, and, reader, you may not be surprised to hear that, roughly, I had exactly the same problems with the board binding in the jig as I did before.

I asked a mechanical engineer friend of mine, and he educated my small electrical engineering brain about the concept of linear bearings. Oh, right! Anyway, another Aliexpress order later and this time, while I was on vacation, I really got the shaft: three 8 mm diameter bearing steel shafts, and some linear bearings, and some clamps to hold it all together. Heck yeah. (My neighbors brought it inside for me.)

I assembled it all, with some different pogo pins, and this one works pretty well. I’ve been using it regularly to test boards and work on the software. And as an added bonus, it just looks so much more pro, and really, that’s the important thing here.

In Which There Are More Tariffs

This seems unnecessary to spend too too much oxygen on, but it is in fact the case that much of this is 25% more expensive, and my margins on this are somewhat smaller now. I built some margin in for "things going wrong", and I guess I had thought that would be something like "something went wrong and I needed to do another run of samples," instead of "geopolitical upheaval in imports", but this general class of thing is what I had budgeted for, at least.

In Which Boxes Will Have Boxes

I don’t want to spill too much about this, so I won’t say all that much yet… but, I just got draft art for the boxes and I’m hoping to have those out to manufacturing in a week or so. If it goes the way I want, I hope you will be truly delighted when you open the box from Mouser.

In Which There’s Still a Decent Amount of Work To Do

Things are going pretty well! But there is still a decent amount of creative work to do. I am going to be busy for the next month, I’ll tell you that. Roughly, here are a handful of things I am tracking:

• I need to finalize test software for the mainboards. There are 564 of them to make, and that is a lot: if I can get each one down to 30 seconds, that's still nearly five hours (and I think that, realistically, packing them all will take much longer than that). The test software works well and is fast, but the thing that will make or break me here is the ergonomics of driving the tool to do it.
• I just spent a weekend in CAD on the printable case and, also, the preprinted case. Production samples are going out for SLS now. I think the lead times on those are pretty short, so hopefully what comes back is right and I'll be able to just turn it right around and run off 314 of them.
• I need to design a test jig for the modules, so that once those come back, I can start ripping through testing and programming serial numbers into those. Not too tricky, just effort. At least I know how to do it from the first one!
• There's a small amount of work left to do on the X1Plus-side software to drive these things.

But, all that said, the end is definitely in sight. It really seems like this is all going to work out!

Talk soon,
joshua