3D printing a programming jig and embedding pogo pins | Hacker News Reader

3D printing a programming jig and embedding pogo pins

conorpp.com

30 pointsecesena18 comments

Cool post, but an expensive SLA Printer is not required for this, neither is Fusion 360.

An OSS toolchain would be a reprap FDM Printer (like a Prusa), OpenSCAD for the design, so modules like the pogo cutout and the base structure can be easily and quickly reused in other projects, and then a Slicer program like Slic3r to turn the model into g-code.

Nothing against SLA: especially now that Prusa has the SL1 coming to market with good features at a sub-$1,500 price point, things are about to get very interesting.

I think regular FDM printers may have problems with the fine detail required for some pogo pins. I know mine can't quite reliably get holes <2mm in diameter, for instance. I guess you could use a smaller diameter nozzle.

> mine can't quite reliably get holes <2mm in diameter

A great trick for small holes on a FDM printer is to not print them circular but square or triangular. You'd be surprised how small you can get fairly accurate holes.

I've always printed on size (so usually undersized because of fdm overflow) and then drilled out with the correct size.

You could also do this with a couple chunks of wood and a drill press.

It'd be difficult to get all the holes lined up correctly

lazy way to solve this is to tape a 1:1 scale printout of the pattern to the workpiece and use a center punch to mark the hole positions. you could also do this with a ruler and a pencil.

I'm not at all trying to say that you shouldn't use certain tools, especially if you have them set up. Do whatever is most expedient. If that means a $5,000 SLA printer, fine (lucky you!). If you're an Open Source advocate and there's a good FDM printer, great, use that. But none of it is _essential_ to this particular project.

This isn't an indictment aimed at the article's author, but when I see someone go "you don't need to use that fancy printer to make this simple object, you can use this slightly less fancy one," it makes me thing that techy people (including myself) are losing touch with how objects are traditionally made.

I agree, you could do this using more traditional methods, I just think leveraging a nice printing process could save some time/effort. Some other challenges that printing solves:

1. you can set the width of the jig to match your pogo pin spring "active region" easily.

2. You can make cutouts for the board and other features (USB-C + USB-C connectors in my case).

I admit, a Form 2 is an expensive perk. But there are much more affordable routes: https://dirtypcbs.com/store/print3d https://dirtypcbs.com/store/lasercut

I.e. when you order PCBs, you can order a $5 SLA jig :)

I'm starting to put support for Tag Connect in my PCBs. This is a really simple and elegant way to pogo into a board without having to build a clumsy fixture.

http://www.tag-connect.com/

The cost to add support on a board is literally zero.

Thanks for sharing this. Would love to see a more compacy version of this but either way, it doesn't seem like a really elegant solution.

Pogo figures are never "elegant" by any means.

Oddly, TagConnect with the locking holes is slightly bigger than 4 standard .100 pins.

TagConnect is really only useful when you want to only program the board, and not do anything like debugging.

Tried it, not a fan. I use the 10 pin ARM JTAG pinout on all my boards with a 0.050" 2x5 ribbon header. Works great, very small.

http://infocenter.arm.com/help/topic/com.arm.doc.faqs/attach...

Love that connector, I’d like to see something roughly the same physical size that can do SWD/JTAG, SPI, UART, and I2C.

Even splitting that across two connectors would be good, maybe SWD+UART AND SPI+I2C

I've had good luck with making the mechanical parts of the jig from PCBs. It's just so cheap to get small quantities of boards made now, and you're laying out the main PCB anyways...

A neat OSS toolchain for this, is KiCAD for the PCB, and FreeCAD for any 3D mechanical bits required. There's rapidly-improving integration between the two, so for instance the PCB can be brought in to the 3D space.

Enjoyed the post. We have made a ton of flashing and testing jigs for my company (flair.co) and I have 3d printed a few but in general, laser cut acrylic seems to be the most popular route unless you need something more than a 2D profile. My favorite strategy is to design them in Fusion360 with the assumption they will be laser cut and then if I don't have access to a laser cutter and/or material, I can still 3D print. Designing for the lost common denominator makes this easy/fast/cheap. Pretty much most anything basic on a laser cutter can be made with a 3D printer if the size isn't too big.

Being able to use clear acrylic is great because you can see through your jig to make sure it's lined up

On the silkscreen, it appears that TX (transmit) and RX (receive) are designated backwards.

Reply on news.ycombinator.com