Hemp’s Flight Sim Stuff

As promised in the previous post, here's the video on how to export drill holes and vias together with KiCAD. The unfortunately thing that caused us having to do this is, for some unknown reasons, when plotting DXF, KiCAD will NOT plot vias, but only drill holes. So, the workaround I found is to export as Gerber drill files, and then open it with KiCAD's GerberView, then export it as a KiCAD PCB file... Then, open the new temp pcb file with KiCAD, and then plot it out as DXF. Now, the unfortunate side effect of this new "holes and vias" DXF file is that there is no edge cut boundary, simply because Gerber drill files don't have edge cuts. Trouble is... when you import this DXF into programs like XtoolStudio, they are not capable of accepting local origins like regular CAD programs would do. You can only specify the coordinate of the upper left corner of the bounding box. So, unless you like eye-ball alignment, this is not acceptable. What we need is the exact same...

One of the problems I often encountered was ... the iteration time of PCB manufacturing. I don't design complex circuits, I mostly design circuits for sensors I need. For instance, a tiny circuit for a single Hall Effect sensor... pretty much one single 8 leads Hall Effect IC plus a couple of capacitors, and pull up resisters, at size like 0.5"x0.5". They could be CNC milled.... I achieved CNC milling of such PCBs 20 years ago.... but they are tricky and finicky.... backlash of your CNC router and the diameter of the carbide end mill would be larger than the trace width means you have to tightly control all parameters. Any mistake, and it will cost you USD $20 for a broken tungsten carbide end mill. Board houses.... take minimal 2 weeks. Any design mistake, and as a hobby circuit designer, I make a lot of stupid mistakes even for trivial circuits. Each costs not just money, also costs another 2 to 3 weeks of wait time. I tried many many different ways of getting around th...

Don't get me wrong, I like the electronics! What is it? It's basically an USB composite device, a mouse, keyboard, and joystick 3-in-1 USB composite device. You get to download a program, ElfKey, and graphically program the foot switch to either output a key, mouse button, or a joystick button, or record a macro. Then, you download the "programming" to the USB device. The MCU is most likely located in the oversized USB plug, completely embedded inside the plug. Not that it can prevent me from opening it up, but I am too lazy to even bother with this trivial thing. The one shown below is an upgraded version with so the called electro-optical switch. It's simply a pair of IR emitter/receiver with a couple of passive elements to 1. drive the IR emitter using a simple resistor as a current source, and a pull down resistor, and 2. a diode, for cascading up to 3 such PCBs (3 pedal set), using just 4 wires instead of 5. The later part I really do not like. I mean, all yo...

  In the last AVRPit post, I "claimed" modular design.... but didn't do a good explanation nor demonstration on modular design. Here, in this post I will attempt to demonstrate what I meant by modular design. In the next two screenshots, you will see that the basic frame design is very simple. It's built on top of a horizontal rectangular frame built with 1515 extrusion, sitting on two 6x6" wood posts with caster wheels mounted under the posts. Everything else are mounted on top of this frame, including a sliding car seat (not shown). In fact, the frame is built around the car seat. I chose a car seat that is narrow enough to fit inside an F-16 pit (most car seats are too wide), and then design the frame's width to fit the rail for the seat. Up front mounted a plywood platform and two "gates" built with 1515 extrusion. The wooden platform is designed for rudder pedals, so the rear half is covered with thin aluminum sheet for sliding of the the pilot...

I recently bumped into a problem with Hempstick 2.0 development, and have to put in a "sort of" delay. Shouldn't affect the final result timeline too much... For those who did not read the previous post about the next gen Hempstick, it's a Linux-based panel controller instead of using RPi Pico kind of embedded processor (I might still write that one, if I find the latency of Linux-based Hempstick too slow, and let it handle the low latency "hungry" stick/throttle/rudder, and the 2.0 handles higher latency panels. So, whether there will be Hempstick Pico depends on the results I get from Hempstick 2.0.). This would be a fleet of RPi Zero 2Ws, each has 4 cores, 1GHz, and as much as disk space you want on the SD Cards. It could also be RPi 5 with an NVME drive (but I am not building a Linux distro for RPi5, as there is no point, just use a Ubuntu distro., instead I will build a bunch of Debian .deb files that you can just download and run "apt install *.de...

Nothing much. Just showing off some fringe benefits of building a "pit" with 8020 (or clones) extrusion are. What everybody thought first of using these 8020 extrusion is the benefit of quick construction. It's easy to cut, and can be quickly framed using store bought connectors, and screws. But the other well-known benefit is easy to hang stuff to the frame, all kinds of stuff. The first picture below shows a basic hollowed 1x1" block. This could be used to "screw" in rails. I wouldn't put it to lock in two major structural load bearing frame, but for minor frames like left, right console rails for an F-16. It's plenty strong. Just don't put you ass on it. You basically slid in the "T-plug" from the end of the 10x10 rail, and then screw in the other one. The next picture shows a split basic block (filled in), and with bigger holes for holding cables. You press in 1/2" T-plug anywhere you want. put the cable on it. And you put in ...

  That is a screenshot of USB Probe on my iMac, showing the Hemptsick USB descriptor... 3 axes, 64 buttons. But it can contain whatever combo you want. It is a live USB device, a Raspberry Pi Zero 2W Linux device, costing USD $15 + a microSD card. It's a 4 core ARM64, 1GHz device. It's a full computer.  The deal, eventually, is... you download a .img file, that is a full Linux OS. Then, you burn it to a microSD card of your choice, with a program like Raspberry Pi Imager. Plug in the microSD card into a Raspberry Pi Zero 2W stick... boot up, and plug in the USB cable to the Zero 2W, and the other end to your gaming rig (iMac, in this test)... and it boots up and shows up on my iMac as a USB joystick. Inside the .img, there are the following, at the moment. a Custom compiled Linux Kernel, v.6.16.3 to be exact, with the RealTime feature turned on, a program, tentatively called hempstick-hid-import, a config file, tentatively called hempstick-hid.conf, a systemd unit file to auto...

They fit. A real Left Aux Console was measured with a caliper and a few other hand tools and 3D modeled, years ago. This year, I acquired the real landing gear module. I didn't even know what those 3 holes were for... but I modeled them anyway. And today, I finally 3D printed it (with some extra excessive stiffening I am regretting now). Now, I know what those 3 holes are for. And they fit perfectly. I think I will just use the real landing gear module, and the real landing gear panel. Simply because I can. ;-)  

7" Raspberry Pi Touch Display 2 on the right Aux Console, fits..... almost perfectly. I had to cut about 2mm into the rim of the aux console on the upper right hand corner. Other than that, it fits right in. 7" is still a tad small for fat fingers, like mine. But, it's workable. Plus, I got a wireless BT/4GHz keyboard touchpad combo thing hooked up to it. Hey, it even comes with a red dot laser pointer. I think a magnetic holder at about where it sits now will do just fine. Of course, in anticipation of my own control stick to be slotted in where the stick is supposed to go, I did added bottom panel, and some ribs inside to reinforce the "slot", and increase the rigidity of that plastic thing. After all, this thing is 3D printed plastic, not cast aluminum. Whether it's strong enough when missile are chasing up my six remains to be seen. I am thinking about putting up a long 12.5" wide screen on top of the glare shield as the 2nd screen. You know, one of...

Haven't posted in over 2 months... I was pulling my hair out in the last two months... and finally achieved two breakthroughs. I will not talk about the first breakthrough, suffice to say it's a software breakthrough. The other breakthrough I will talk about now is I now have a Linux Hempstick distro. for RPi Zero 2W. Yes, you read it right, it's a whole Linux distro. I build myself. It contains the followings. The newest Linux Kernel 6.16.0 (only about 3 weeks old). It has the newly mainlined Real Time module turned on. It has a minimal Linux distribution (as much as I care to install), built into a .img file just like the Raspberry Pi OS. Why in the hell do I bother with this difficult route???!!! Well... originally, it started with an idea of using a Linux on RPi Zero 2W or Pi5, etc... for controlling all the panels in the "pit." One Linux per panel... or whatever. Then... there are a bunch of stuff I want to write and install on each of the instance of OS, pos...

Where indeed to put the keyboard when you build a pit??? I decided to put it right above the F-16 hood. I have a keyboard arm for it installed on the left hand side of the pit on a 19" rack, but ended up not using it because it's too flimsy, and it's quite difficult to type anything longer than 2 characters when the keyboard and the screen are 90 degree from each other. So, it's now used to store the Quest Pro and charger. There are two problems with putting it on top of the hood. It's a bit far, if I am not careful, I would bump my chin on the ICP, and It's 2 1/2" below the top edge of the "hood" making it a bit difficult to type. So, to solve the problem, I need a keyboard riser to boost it up to at least even with the hood. This solves the long term storage problem of the full sized keyboard and it's at least not too difficult to use. But it still doesn't solve the problem of being too far from an inclined seat during a flight. The bes...

This is the new 10 degree F-16 front panel fixing adapter (the black one up front) with quick connect. The bottom is supported by the same rod adapter (blue, with an additional lip) sitting on two solid OD=1" aluminum rod (with a notch to accept the lip). The weight of the front panel is supported by the two solid aluminum rods (slids in and out for supporting different "boards", like driving wheel board, and large keyboard/mouse tray etc. Now, this new quick connect adapter allows me to quickly connect/disconnect the F-16 front panel, and align to the 10 degree correctly and accurately, with the help of two "spacer" rings on the bottom. The top quick connect will force roughy 10 degree when tightened. And the bottom spacer ring will butt against it to make accurate 10 degree. Then, the two aluminum rods are locked in place by two more quick connects to fix the position of the bottom edge. The space between the two quick connect adapters is just perfect for ano...

It's very simple. It's built out of 80/20 aluminum extrusions, specifically 1515 and 1010, i.e. 1.5x1.5", and 1x1". The main idea is simple. It's a rectangular 1515 frame, center to center exactly the center to center width of a narrow sliding racing seat I bought from Amazon. The center to center is about 14.5". Some of the regular wider racing chairs on Amazon would be too wide and would eat into F-16's side panel and control stick space. There are two 4x4" wood post bought from HomeDepot for widening the "stance" and mounting 4x office chair casters. Otherwise this thing would easily flip over when you lean to one side or the other during dog fights (I can't help it!). That was the main load bearing structure. Then, I put two 1515 gates up front for mounting the center console and other stuff.... other stuff means retractable and/or replaceable trays for keyboard, mouse, etc. They must be lockable for they don't move, preferably ...

Refinements of the Mini-Hal Joystick. The basic mechanical design prototype was completed. But it's not done yet. Just received the PCBs and have to now consider the details of the wiring, assembly etc. I do have some idea about how to wire them, with electro-magnet wires, or 30+ gauge wires (that's why the 4-point star shape of the inner components. Now I also have to consider how to easily assemble and make this stuff... for you.... and me. There are already a lot of design details considered for you, and me, to make them at home, preferably with just a 3D printer and some hand tools, avoiding using CNC mills at all costs.... For instance, one of the problems that I have to solve is how to make the latex "spring", easily. Take a look at the first picture. That's the set of jigs to make the latex membrane "spring." The basic problem is that I have to 3D print 2x concentric rings (one OD=1/4", ID=2mm", and one OD=19.31mm, ID=10.34mm) and tie a ...

The mechanical prototype of the Mini-Hall Joystick is completed. It's now mostly 3D printed. I have not beautified it with all kinds of methods to make it look less 3D printed yet. I am still waiting for a new version of PCB, the analog version with MLX90333, plus the optical sensor. I totally forgot to design and order this one but only ordered the planned more advanced version with MLX90363 (fully digital SPI version). Note that in the video, the click click sound is not the activation of the button press. It's the central detent. I did not expect it to be this loud and distinct. It only sounded that loud because of the echo of my vaulted ceiling dinning room. Oh well, now I definitively will know the center! This version is vulnerable to UV damages. I will have to find a way to "cover" the vulnerable part. I think I already have the solution, no worries. The only parts that are not 3D printed are: a small piece of latex sheet, a bronze sleeve bearing, a neo magnet,...

This is a very satisfying button click test for the new button mechanism for the mini-Hall joystick (no gimbal). Most of the parts are 3D printed. Inside, there is an off-the-shelf self-lubricating bronze sleeve bearing for smooth button presses. The force required is 1.8kgf. You can see in the video that such large activation force is not only for the satisfying click, it is also necessary to prevent accidental activation. It's fatal if you accidentally activate the boost button (I often assign it on the F-16 mini-stick button press, and it's a bit iffy with the TM F-16 TQS Viper's puny activation force) inside a space station in Elite Dangerous. I could have done 0.9kgf with just 1x dome switch, but I felt it's too easy to accidentally activate it so I doubled it with 2x switches. This new mechanism (not shown) is much straight forward than the previous working design, and is only possible with the newly released stainless steel dome switch with back lighting (I am no...

After months of keeping the finished two-stage trigger mechanism on my work desk as a fidget toy.... I broke it. It's not as scientific, nor as gimmicky, as how Ikea showed you how they tested their  POÄNG  chairs in store. But I figured, if I can't even pass this simple durability test, forget scientific methods. It... still works... it's just that the 2nd stage force is no longer as strong as new. It's still very distinct, but, I can feel that it's breaking slowly. So, now I have two choices, assuming the main design does not change. Go Hard Go Flexible. This merits a bit of explanation. The original Otto two-stage trigger uses plastic, mostly plastics of various kinds... I am NOT sorry to break all your all-metal mythologies... My guess is that they are made of Delrin, aka Acetal. Delrin is a very slick substance, not as slick as Teflon, but very close. But... it's very hard, as far as plastic can. It's almost as hard as aluminum. So, you can easily machi...

  Optimizing the design.... or more appropriately, optimizing the manufacturing process in design. In the picture, the above is a one-piece clone "piston" of the 2 stage trigger of an F-16. It's lathed on a mini-lathe, out of an OD=8mm brass. First, it has to be turned down to OD=7.5mm, then, the middle section has to be turned down to 4.5mm, then one end has to be drilled with an D=1/8" hole. This whole thing has approximately the same dimensions as the real one. The blunt nose section's precise shape is not important. The original is a circular dome (plastic injection molded part). But, mine is... well, filed on a lathe. The curve? I have no idea... well, about there. But... if you ain't got a lathe.... that is going to be very difficult to make. Just drilling a D=1/8" on an D=4mm hole is going to make you sweat. Any deviation on the center finding (or even wandering of the drill bit) will make it very obvious on these small diameter, thin walled part....