Hempstead

Thanks for your kind words, Weegie. Just wanted to add a little clarification. On page #2, there is the full spec. of the PTFE ring. You are free to make as many as you wish and sell them too. If you do sell them, please let me know, so I can refer to you and get out of this community service. The ring is made out of the most common grade of 1/8" thick PTFE sheet (yes, there are several grades of PTFE, including Teflon AF, an optical clear PTFE). 1/8" is slightly "too thick" and indeed interferes with the most extreme deflection of the stick just a hair. However, I chose this grade of PTFE with the thickest possible thickness so that it can: 1. be stiff enough to wedge itself in position without the aid of glue, 2. have more "meat" for wear, and 3. this grade of PTFE is softer than the ABS (?) of the gimbal, so the PTFE wears but not the gimbal, if any. The next thickness, 1/16", I can get is too thin and provides not enough stiffness so it would require glue or double sided tape. That would have gotten you in the same problem like the original rubber ring -- glue failed and the ring got out of position and got pinched. However, the PTFE I use is soft enough such that after about 1 week of uses, it will form a ring of shallow depression groove due to compression and reduces the effective thickness and you get back the full deflection. The change of effective thickness shouldn't affect the calibration to the degree that you can feel it. But, it wouldn't hurt to re-calibrate often.

Lowest resolution output of my SLA printer. Left, the real one, brand new. Middle, lightly sanded, and glass bead blasted. Right, right out of the printer, alcohol washed. Didn't even bother to go higher resolution. After UV cure, it's hard as nail. SLA printers are still expensive, ~USD $3,000, and the process is still not ideal. But, they are slowly becoming viable.

I am done here. Have fun!

That would be your freedom to do so, man! Then, don't share! It's also my freedom to spend thousnads of USD on authentic parts, then spend countless hours meticulously measuring the authentic parts to construct these 3D models and then give them away free of charge, as I prefer not to turn myself into a minimal wage worker making peanuts. Damn it, it's a hobby! Not a job! Don't turn a hobby you enjoy into a job, because once it's a job, there is no fun in it! You see why I treat the PTFE rings as a community service, and I actually only barely stay in the black? It's a hobby, not a job! If it's a job, I'd have to charge $20 apiece and you would have no choice but to pay up when your $500 joystick broke, or make it yourself, or contact TM to get a relacement rubber ring which is bound to break soon. If it's a hobby, then share! That's my take on it. You are free to disagree! Some would prefer to buy your faithful replica with minimal fuzz, some others would prefer to make them with their own hands, given they could get their hands on the information. Each to their own! I just prefer to give people all the information and let them go "hog wild" with it and then come up with something better than I could ever do! Once people downloaded the 3D models, they don't need me anymore. Call me a Hippie if you like even though I don't wear a flower in my hair in San Francisco, I prefer free flow of information.

Attached is the image of the 22 knobs set Kumrik sent that will be published later. These are indeed F16-centric. However, the point is that he constructed these out of EHC spec sheet, without physical knobs. They are damned good! I am sure there are some errors here and there, but I haven't found them yet. I am pretty sure they will work very well if you were to 3D print them out with SLA 3D printers. The point is that if Kumrik can make these knob 3D models out of the EHC spec sheet, you can do the same to the A-10 knobs too. If you do 3D model them, please consider sharing them!

Kumrik has sent me a bunch of knobs 3D models he constructed from EHC specsheet. They look pretty good! We are planning to put them up on the Hempstick Flight Sim 3D model pages. These are good for 3D printing. If you have SLA printers, they print out beautifully. After a bit sanding and painting, you would be hard pressed to tell them apart! Note that, nobody says you can't mold the print and cast them out of resin (although some photo resin inhibits some RTV). If your printer can print multiple knobs in one go... unless you need a lot of the same knobs, casting isn't exactly time saving (might be cheaper). For instance, my SLA printer's software allows me to lay up multiple different knobs in one go. So, even though it takes about 40 minutes to print. I can get 9 knobs or more in one go, and each can be a different model of knob! That is, all the knobs of a panel in one go. Stayed tuned, the new knobs models should come up in a few days time.

I don't know if it's available in UK. I use SuperLube, a PTFE-based synthetic grease.

Can't say it in public. Otherwise, I could get sued for defamation, partucularlly when CougarWorld, the site I posted the idea, is down, i.e. can't even defend myself. I live in the land of the free, got to watch out for law suits that could ruin you fonancially.

I agree that's indeed the case. I figure it's because the posts and holes are not perfectly aligned. So, if you rotate the platform relative to the posts, i.e. make the holes mate w/ differrent posts, then you get tighter or loser mating, resulting in more or less dead center. Of course, I could be totally wrong on this reasoning/hypothesis. But i think sharing the reasoning behind why I think so is more important than the end result. If the hypothesis is correct, then it stands to reason that there is one direction (1 of 4 possible permutation) that is tighter than the other 3! Therefore, you should try to rotate the platform and get the tightest one!

I posted the following findings long time ago on WarthogWorld before it went down. The dead center on Warthog is only dead center for the spring force, not the sensor measurement. The sensor will still accurately pick up the stick movement inside the dead center zone, as Shadowlin pointed out. However, I think the dead center is mainly caused by the 4x holes for the 4x posts being tapered, not because the holes are larger than the posts (the holes on the white plastic being bigger than the post obviously would make that worse, but even if it's not, you would still have dead center). If you look carefully, you'd find that the hole diameter is slightly smaller than the bottom, i.e. the cylindrical wall of the hole is not exactly cylindrical. I didn't notice that before I went on to mill a "perfect" spring platform out of 6061-T6 aluminum, see the attached picture. I thought the same thing, that the holes are not accurate, so I set out to CNC mill this thing out of a solid chunk of aluminum, and accurately drill and ream the 4x holes to be exactly 5.01mm, to tightly fit the 5mm diameter posts. It took hours to accurately CNC mill this thing front and back out of a solid chunk. The result was a "spectacular" failure -- the precision reamed holes bind the 4x post as soon as the stick is moved, i.e. the stick now does not move at all. Upon close inspection, I find that as soon as the stick is tilted, the rim tilts up on one side, contacting the platform at one point (through the PTFE/rubber ring). Then, naturally, the platform wants to tilt too. The 4x posts are there to "guide" the platform up and down and prevent tilting. But, if the 4x holes are as tight as I made them, each post will contact its corresponding hole by two points only, one on the top, and one on the bottom. This causes great pressure between the post and the hole, thus binding! I am quite sure TM found this problem during prototyping and they solved the problem by tapering the holes, making the contact area bigger so no binding. But the consequence is the slight dead center of spring force -- the platform has to tilt slightly before moving up and down. I am designing a new gimbal mechanism that has no such dead center of spring force, and might be able to adopt this mechanism to Warthog. However, this time around, I might want to patent the idea if it works well. So, some commercial company can't just take the idea and run with it and beat me to the punch without even acknowledging the idea's community origin, like last time. But even if I patent it, I will let the non-commercial use by the community use it for free. Don't hold your breath on it though. It's not urgent and I have so many other things to work on. It will take months (oh, it's already been months since ... ;-)

Again, please DO NOT superglue or epoxy the old ring or any ring in. This ring is consumable. You permanently glue it in, then how are you going to replace it when it wears out? Clean up all the grease and double side tape it back. Or buy a PTFE ring from me, which requires no gluing. Just slide the PTFE ring in place, grease up, reassemble and go. I view the PTFE ring thing as a community service, so I am only charging USD $8, s/h included. Just enough so I don't lose money on it.

There are reasons why most of the modern aviation controls have "standardized" yaw control with foot pedals. Didn't used to be at the dawn of aviation. The Wright brothers used their hips for yaw. The main reason I think is human physiology. Like MetalWood said, twist action of your wrist is just not as precise as your foot work. And more importantly, it's almost impossible to move in the pitch and roll without affecting yaw if you have a twist stick. That is not exactly a desirable trait when you are trying to fine tune your aim at a target. It's particularly bad when you fly a chopper and trying to aim some rocket pods at ground targets. Also, when under stress (say bandit at six with radar lock on you), humans tend to use actions that are more natural to their physiology. With a twist stick, you tend to turn left more (for right handers) than right, because it's a lot harder to twist your wrist right than left (try it if you don't believe me, and then center mount the twist stick and try it again). I tested this theory when I was teaching a buddy to dog fight; it worked like a charm -- he "always" turned left when I got the radar lock on him (he's right hander with a twist stick!); and I would be waiting there for the kill. Then only tell him when he got frustrated. Now, he would never forget that lesson -- don't use a twist stick! ;-) I am not saying twist stick is such a big handicap. There are people who are very good with twist sticks. But in the air I will exploit any weakness. If you hand it to me, I will certainly take it! And, thank you very much for being so nice in a dog fight!

It is, unless you can find a ready made pot with long step cut shaft with threads on it and have tools to cut metal. Or you can use something like this, PTR902-2015F-C104 on Digikey. Since this is controlling light brightness, it needs to be logarithmic taper.

Here's how the original's Flood light works see the picture, you can do the same, similar to DM says for the Flood light. The original flood light electronic is a variable AC transformer with a off position built in but that's irrelevant to us. In the picture, from the left to the right. 1. Pot (made in Mexico. ;-), 2. a aluminum disk with a drilled hole and a pin to create the click at "off" position (no electronics here). 3. an aluminum disk with a flat notch and a microswitch to trigger the off signal, 4. another aluminum disk with a flat notch and a microswitch to trigger the T Storm signal.

I'd still be interested in seeing why yours wear out too! How about this... send me a PM with your name and shipping address. I will send you one for free. Let's see if we can find what's causing the wear.

You wear out the PTFE ring I made? That would be the first. I would love to find out the condition that wears it down. If you could, pictures please? .

Very much agree on that point. Look, it's a home flight simulator. The more details the better, but there is a point of diminished return. Even if you are building a real replica that can take off, there is still a point of diminished return. I ain't going to spend 10 years collecting all the authentic measurements. That's just me. But to some, that's their "hobby", it's their pride and joy collecting this stuff (I collect the useless state quarters. ;-). Great, each to his/her own. If you would share that info collected over the 10 years period, I'd greatly appreciate for saving me a lot of time! Hey, if you publish a book, I'd even be glad to pay for it! That point of diminished return is different for everybody. To me, it's not worth my time chasing down the last detail. I am not even interested in building the whole pit. I only want some panels that matters to me in the air. Lighting control panel is NOT one of them, but that's the only one I can get on eBay and it serves as a good sample to see how real panels are made and emulate "as close as I CARE." Note that, NOT as close as I CAN. I am more interested in the real function behind them that gives me an edge in the air. That's why I wrote Hempstick because after I got the panel to my satisfaction, I couldn't find a suitable controller to drive it, not even for a freaking simple lighting control panel that does not give me any edge in the air! All the commercial controllers I could find are not extendable, and the extendable ones are Arduino crap; none of them meet my criteria (I care more here, not there!). And, except one or two controllers, all use extremely outdated MCUs. Without good controllers behind these... to me, it's just a good light show; useless to me. But to others, that might be good enough. Hey, everybody values different things... If Arduino is good enough for you, it's good enough.

As promised, the lighting control panel SolidWorks files can be downloaded in the following URL, http://www.hempstick.org/download/warthog/Warthog-LightControlPanel.zip This is measured directly from a real A-10 simulator panel. I bought the whole thing from eBay, which included the "electronics" and the mill-spec connecto and housing. It uses variable power transformer instead of pots (there is one that actually uses a pot) The US military has long standardized on mm as the length unit (standardized is a relative term... a lot of their screws and bolts are still imperial unit ones). But, somehow, the designers are still used to Imperial measurement and they have to work with raw stock that are in imperial units. So, they sometimes have weird mm length that only make sense when you convert them to inches. I have to work with such mixed bag of measurements as well as I live in the US... It's very tough for me to get any raw stock that are in metric (reads expensive). So, I always draw such things in metric units, but uses imperial unit stock. A good example is the thickness of the panel. it's ... 1/4" (the real one is 1/4" thick too!).

With the R1 plan http://hogpits.net/downloads/summary/6-reactorones-a-10-cockpit-panels/19--r1-a-10-panels-.html. It's a good starting point. If you would provide a set of authentic measurements from the "real" thing, that'd be great. Otherwise, I will draw them to the way I like based on the R1 plan. I don't know R1's source data anyway. I displayed R1 plan's lighting panel on my screen, and then put the real panel on it. The location of dials and switches are a bit off, but not too badly off. My lighting panel, however, is from the real thing (even though mine is from a mil simulator, it seems to use the same thing as the real production planes). I do use non-authentic font, but like a former colleague who used to work in the game industry often say -- if it looks good enough, it's good enough. 100% authentic is never my goal.

I have one, the lighting control panel in SolidWorks. It's measured from a real A10 simulator's lighting control unit I bought off eBay. I'd be happy to put it out in the public domain. I will put it up on the Hempstick site this weekend.

Arduino Due/X does have an atmega programmed for Arduino debugging/programming on board. So, if you use it with Arduino IDE, you don't need to buy a separate programmer/debugger. It's just that we can't use this "debugger/programmer" with Atmel Studio, Atmel doesn't support this one. But Arduino folx are graceful enough to wire out the JTAG debugging pins for other debuggers. This is what ATSAM-ICE and Atmel ICE debuggers support. In other words, Arduino Due/X board can be use for both Arduino purpose and anything else, including Atmel Studio for Hempstick, unlike its older brothers. I suspect that there will not be any clone for the XPro boards, because Atmel does not sell the EDBG chip to anybody. Moreover, the cloners have very little incentives to do the XPro's small market. Also, note that Arduino Due/X's JTAG debug header is 10pin 0.5" pitch.... ATSAM-ICE does not come with 0.5" pitch header. You'd have to buy an Olimex adapter and wire for it. Atmel ICE, however is native 10 pin 0.5" pitch. And Due/X's layout has a little problem -- the 10pin 0.5" header won't go in. You'd have to cut a little plastic corner to get it in. No big deal.

AFAIK, F16 FLCS's stick electronics is similar to Cougar and Warthog, just a bit older and slower version. Hempstick should work right out of the box. And you get 14bit instead of Mega's 10 bit. Also, you get 1ms USB report rate, i.e extremely low latency. That is, you turn a game port stick into a state of the art joystick. Just follow the instructions for Cougar and it should work.

It's a simple why question. You read too much into it. Open exchange of ideas is great. That's why I ask why. Because I don't get it. Exchange of idea is not just about the what part, the more important aspect, IMHO, is about the why, i.e. the reasonings and insight behind it. For instance, remember I asked why you changed from NeoPixel to matrix LEDs? Is there some trouble using NeoPixel that I did not know about that I would learn a lesson from you? Or is it some trouble with using Arduino to drive NeoPixel, that I would not have to worry about? Why would you choose a string based protocol that is not exact the strong suit of embedded processors to parse. Is there a limitation in the DCS that you must use a string? I don't know the answers to those, so I asked why. As simple as that. I did say "That would have the same learning value, yet produce something more useful." Meaning, I do agree with you about the learning value of what you posted. But you seemed to have read the opposite. "more useful" -- meaning, what you posted is useful, but you could have done even better if you have made the board do things those two other boards I linked to could not. I don't know the answer to that, so I asked. As simple as that.

Why didn't you choose to make a board that is something you need and nobody else already made much better ones? That would have the same learning value, yet produce something more useful.

Why didn't you choose to make a board that is something you need and nobody else already made much better ones? That would have the same learning value, yet produce something more useful.