#Breaking-Taps #video about how he got silicon #electronics fabricated (through #WaferSpace #MPW, not a multi-project chip). He settled on a MOV machine (transport-triggered architecture) called BTTAP, designed it in the Spade #HDL, synthesized it with Yosys and ABC, and simulated it for SKY130. Includes animations of some analog simulations. He tested his RTL (?) in cocotb instead of Verilator and viewed his waveforms with Surfer instead of gtkwave. Then he used LibreLane, which wafer.space provides as a #Nix Flake, to get from a netlist to GDSII. Finally wafer.space got it fabbed through GlobalFoundries’ GF180 process, which I didn’t know had an open-source PDK, in only 3-4 months. Wafer.space got some of the chips packaged as COB with transparent epoxy. He got about 1000 chips. From looking at MPW pricing in the past, I’d estimated that this cost him about US$2500, but I think it was actually US$5000. wafer.space GF180MCU run 3 is in December, early-bird price for a quarter-size chip (1.94 × 2.53 mm) of only US$2000.
on 02026-09-22#Breaking-Taps is fabricating his own 20MHz transport-triggered microprocessor at GlobalFoundries in 180nm using Tim Ansell’s new "wafer.space" #MPW service. Also he’s doing a cubesat and has sold off most of his shop. Also there's a coarser-grained alternative to FPGAs called a "CGRA" which is similar to #Groq’s LPUs.
on 02026-03-12#video #toread about voice coil #robotics #hardware #mechanisms from #Breaking-Taps. Inspired by birds. Using Dyneema cable actuation. Glass-filled PTFE bushings, carbon-fiber bones pressed into holes in fittings. He actually gave up on the rolling-contact cable-ligature joints and switched to conventional sleeve-bearing pin hinges. Using voice-coil actuators.
on 02025-10-14#video by #Breaking-Taps on #manufacturing optically-clear epoxy copies of glass lenses by casting in silicone molds (Smooth-on Mold Max XLS); he gets 8.90nm RMS surface roughness on his copy, worse than the 1.74nm RMS surface roughness on the glass original, according to his AFM, plenty good for #optics. It worked easily for 10mm lenses, but it was difficult to get the process to work with 100mm lenses, with hazy surfaces (due to mold release), deviations from sphericality, etc. He says you can’t polish, lap, or grind plastic, which I don’t think is true of all plastics. A reader says, “Hey, I just want you to know that you can actually polish a plastic lens. I’m a lens maker in Canada and everything we do is plastic. We don’t always do PC but we do a ton of other materials. Everything gets polished.
we use a liquid polish and some abrasive pads, and I’m not exactly certain what kind of polish you could use for it, but it’s definitely something we do.”, and another comments, “I worked for years in a surface lab lapping and polishing plastic lenses all day. It’s absolutely practical”. A third says, “YES you can polish plastic as good as you can polish glass! You just need to use finer grit paper or less abbrasive [sic] materials and also you get the best result if you do wet sanding and then wet polish with VERY high grit sandpaper and oil or water. I do it myself so YES you can!” He sputtered metal (60nm silver) onto the glass as an alternative mold release that wouldn’t create micron-scale roughness. Smooth-on Accel-T tin-cure accelerator additive caused more shrinkage during curing of the silicone, ruining the figure of his mold. Relatedly, he says you want the resin to cure as slowly as possible, because fast-cure silicones shrink more. Mold Max XLS he says is 1000ppm shrinkage, EpoxaCast 690 is 2000ppm.
on 02025-09-28#video #toread by #Breaking-Taps on laser scanning #microscopy
on 02025-06-22#Breaking-Taps (Zachary Tong) #video on liquid-metal ion thrusters for #space
on 02025-03-06#Breaking-Taps #video about looking at #precision flat surfaces with atomic force #microscopy. His new gauge block has 12 nanometer RMS surface roughness with a total range of almost ±50nm over the 10μm square field of his AFM. A first-surface mirror is 3nm RMS with 36nm range. A λ/20 fused silica mirror blank (λ = 632.8nm) is 2nm RMS, 39nm range. A cheap glass microscope slide is 1.9nm RMS, 62nm range, mostly due to a few big peaks. A silicon wafer is 1.6nm RMS, 17nm range. And a piece of mica (!!!) is 530 picometers RMS, 8nm range.
on 02024-09-17#Breaking-Taps #video #toread about DIY laser #photolithography
on 02024-09-17#video #toread about #3D-printing #optics components (not lenses, prisms, and mirrors tho) #Breaking-Taps
on 02024-09-17#Breaking-Taps #video about #MEMS accelerometers. He uses a lot of electron #microscopy to show the #flexures and other #mechanisms, then makes macroscopic replicas with #3D-printing. I don’t understand the whole flexure layout, but I appreciate the effort. He says the “gyros” use the Coriolis force generated from resonant vibration, which changes their resonant frequency. I don’t know. #toread
on 02024-09-05#Breaking-Taps #video on high-entropy alloys #toread
on 02024-09-02#Breaking-Taps #video on doping silicon with lasers. With phosphoric acid. Explains how his IR fiber laser penetrates too deeply to be ideal and how hot-probe tests find positive (Seebeck?) voltage for N-type doping, as produced by phosphorus, and negative voltage for P-type doping. But he hasn’t been able to fabricate a PN junction on his P-type wafer yet. So apparently to avoid Schottky contacts you have to use “titanium and aluminum” (titanium-coated aluminum?) for the N-type probe and just plain aluminum for the P-type probe (or a highly-doped N-type region), both annealed at 480°.
on 02024-09-02#video on #manufacturing #optics by #machining pure copper (to eliminate inclusions enriched in other elements that can cause tearout) with a monocrystalline diamond cutter. Lots of footage from atomic force #microscopy and scanning electron microscopy. To eliminate 50μm steps between adjacent passes of the diamond milling cutter, he switched to 0rpm, just using his milling machine as a sort of shaper, and got the surface roughness down to about 13nm. #Breaking-Taps
on 02024-08-28#BreakingTaps #video on small rolling-contact joints when ball bearings won’t fit. He holds the joint together with Dyneema tendons under tension (“Hercules” brand PE braided fishing line), rolling through the joint like a Jacob’s Ladder toy (magic tablets, Chinese blocks, Rubik’s Magic). He’s left channels in between the rolling contact surfaces to pass the fishing line through. So essentially the joint is sewn together.
on 02024-05-15#video by #Breaking-Taps about a homemade #vacuum box, made from a used eBay Laco Technologies box, so not really very homemade at all. Explains roughing rotary vane pumps vs. his 50krpm turbomolecular pump that gets down to 2.4 nanobar, oil mist eliminators, coaxial foreline traps to eliminate backstreaming of oil, ISO 100 flanges, KF16 and KF40 and KF50 flanges, conflat (CF) flanges with copper gaskets, KF/NW centering rings with green Viton O-rings, etc. This is the first time I remember seeing a detailed explanation of the KF/NW flanges, and I really like the design.
on 02024-04-28#Breaking-Taps ultralight metallic microlattices. #video Somewhat disappointing because he didn’t manage to do the self-propagating lightguides thing or the electroless plating thing.
on 02024-04-03