richfiles Posted December 10, 2021 Author Share Posted December 10, 2021 (edited) Here it is! The 115vac 400Hz Amplifier. Takes my 10v p-p reference sine wave from my California Instruments 850T Precision Sine Oscillator, and boosts it up to enough voltage and current to drive the Navball! This schematic doesn't show the rectifier or the bulk capacitors that create the +/- High voltage rails, but those portions of the schematic are textbook. Also since added is a resistor and coupling capacitor to a shutdown pin on the primary op-amp. I caught that one. The HV rectification is done at the top. The center area with all the surface mount parts is the primary input. The high and low portions of the circuit are the tri of transistors on the left and right sides. I believe the center transistor may (but I'm uncertain) be involved in foldback of the current if it exceeds over-current limits. I'm not 100% certain on that, but I think that's the gist of it. A huge shoutout to Dr. Rockzo, who decided to clown around with some high voltage analog circuitry after we clowned around about Chrome tabs! I'll try to make sure the files are available, whether he has a repository for the stuff, or whether I set up something. For now, he's happy sharing it. I don't know the exact license, but basically build it freely, no commercial. We've still got a few tweaks... I noticed the orientation of the left set of transistors makes heatsinking awkward. He's flipped them on his file already. I haven't had time yet. I need to do the same, as well as get him my modified files with the shutdown circuit tied to enable. I'm really excited to have this high hurdle knocked out of the way! This simplifies a lot for me, and lets me Finally power up my Navball! Edited December 10, 2021 by richfiles How fitting! A new page on this thread for a new chapter of development! Quote Link to comment Share on other sites More sharing options...
richfiles Posted December 19, 2021 Author Share Posted December 19, 2021 Made a few minor updates to the PC board. flipped the power transistors on the left side, so the heatsink tabs face outwards. Just needed to rearrange a few component locations and trace routes to make that work. Really despising the parts shortage right now... I am very low on hours at work, and I don't know how long it'll be before we get parts again. I'm not sure if I should order these parts right now, given my hours, and thus income right now, yet at the same time, I have nothing to do, plenty of time, and once parts do show up at work, I'll be slammed playing catchup, and have no time for any of this, most likely... It's very frustrating... Quote Link to comment Share on other sites More sharing options...
Lisias Posted December 19, 2021 Share Posted December 19, 2021 (edited) 1 hour ago, richfiles said: I'm not sure if I should order these parts right now, given my hours, and thus income right now, yet at the same time, I have nothing to do, plenty of time, and once parts do show up at work, I'll be slammed playing catchup, and have no time for any of this, most likely... It's very frustrating... Order the parts. I'm on the exactly opposite situation as you: I have a pile of parts on my kitchen waiting for me having time to use them . I have some very interesting toys on that pile for almost an year now. Well, there's a chance I will have some time to spare on this holidays, so it's good to have the parts around. Two years ago I had a whole month to spare - but since I didn't ordered any part (as I didn't knew I would have that time in advance), I spent that time playing videogame and watching animes. So I learnt my lesson: better having the parts hoping for time to use them, than having the time and no parts at all. — — pOST eDIT — — I didn't paid enough attention, and missed the income thingy. Money changes everything. Yet, when I got on the same situation last year I choose to spend some money - and I don't regret, as some of the parts I bought are almost twice the price now. Granted, a few are cheaper… Well… You don't have to buy everything at once. Try to plan ahead what you will do if the time is available and buy only the parts needed for that task - at least, it is what I woul do. Additionally, and this is something I do now and then, you can try to find second hand components - perhaps buying broken equipment to scavenge working parts. You will have to cope with the trash later, so this is not always an option - this is easier when you have partners on the same situation and so you can exchange components you don't need for the ones you do with them. I have a friend on USA that used to go a flea-market that happens regularly on MIT - he told me that he was able to find some good pieces of equipament for pennies, some even working at is. Flea markets can be your solution if you find the time to scavenge working components from such material. Edited December 19, 2021 by Lisias post edit Quote Link to comment Share on other sites More sharing options...
richfiles Posted December 22, 2021 Author Share Posted December 22, 2021 (edited) I have officially ordered the board(s). I'm getting 5 identical boards, just cause that's how the manufacturer does things. Total cost was only about $22, and that was with the faster shipping. Even if I manage to order parts tomorrow, I'll still possibly get the boards before the parts... possibly. More than likely, I won't have time tomorrow, so the board has a solid head start. Unfortunately, me incessantly working on other people's projects has screwed me over... The board design took into consideration parts that were in stock at the time of the design. Because I have myself tied up in half a dozen projects that aren't even my own, my board revisions took some time, and getting the board ordered took some time... It's also been enough time that one of the two main power transistors is out of stock with a month+ lead time... I hope... I can't win... At least I actually got the board design finalized and ordered. Since I'm already kinda in a board designing mood, I think I wanna try to come up with a board design for those Russian bar graph LEDs I picked up a while back. Huh... I thought I posted about that, but i didn't see it. Anyway, I had debated for a long time if I could justify getting the LEDs... I just went ahead and got them several months ago. Anyway, the big issue with the boards, is that they don't have very much space to sit, and there are a LOT of pins I need to break out, which makes through hole mounting iffy at best.. Furthermore, orienting the board endwise, while it could work, also likely means slots in the aluminum panel, on three of 4 sides of the Navball, and I want the maximum physical support in that area, so I wanna minimize the size of any slots I cut into the surrounding aluminum. I pretty much expect to be multiplexing them, but the red and green LEDs have to be multiplexed separately. The reds are common cathode, and the greens common anode. No idea why they made the parts that way, but that's how they are. The Rate of Rotation Meters will consist of three bar graphs, one for each axis of rotation. Each LED module is 5mm long, and has five 0.5mm x 1.6mm LED segments. Each bar will use 10 of the green modules, and 6 of the red modules, for a total of 80 segments in an 80mm (3.15 inch) bar. If space or electrical complexity becomes an issue, i might be willing to go down to 9 or 8 of the green modules. I'd rather not though. i could cut one red module of either end too, and have less red portions of the bar graph. honestly, depending on the complexity of it, I might look into a 4 layer board, if that helps route all the connections. Routing is such an issue, because of the limited space and absolute number of connections to be made. I feel like I should make a small addendum... These might look like standard DIP style chip packages, but they are very much smaller than the standard DIP package. A standard DIP has 0.1 inch spaced leads. this has 0.049 spacing, converted to the same units... as small as the part is, close enough to 0.05, or half the pin pitch. Edited December 22, 2021 by richfiles Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 6, 2022 Author Share Posted January 6, 2022 (edited) Now THAT is a fine sight! This has been trapped in my mind for a very long time now, and it is good to see it in a physical, tangible form! For the time being, I have the 4 power transistors tacked in place by only a single lead. I wanted to leave them repositionable. I still need to find a heatsink that will be suitable for this, and have yet to finalize the orientation of the transistors. A happy surprise! I can't believe how closely this lines up with the transformer dimensions. I could probably make an enclosure to contain the high voltage bits, and it'll all fit nicely in a simple, uniformly sized box! Definitely a neat little bonus! ... Now for some bad news... Here is a shot of the front and back sides of the PC board, and the mostly completed unit... Notice the location of the negative stripe on the Electrolytic Capacitors, and the (+) marking on the PC board. (-) faces the rear, and (+) faces forward. Notice the (+) marking for the Bridge Rectifier on the board, in the rearmost through hole? Now look to the left side of the image, to the back side of the board... The traces for the bridge rectifier and the capacitor banks are all parallel. There is definitely a case of crossed polarities! That's BAD! I took a look at the schematic, but it correctly shows the (+) of the Bridge Rectifier, all the (+) leads of the Capacitors, and (V+) all properly connected. Likewise, it also correctly shows the (-) of the Bridge Rectifier, all the (-) leads of the Capacitors, and (V-) all properly connected as well. This was definitely the schematic linked to the board layout, and yet the board layout CLEARLY showed the (+) and (-) of the Bridge Rectifier flipped... When the files for this design were originally sent to me, I tried to open them with an older version of KiCad, and the software was unable to open the files. I proceeded to install a current version on my tablet, but when I opened the files, I had all sorts of issues with broken library links. Furthermore, the 3D models were not associated with the part footprints at all! And it's here that I think the mistake occurred... I had to reallocate all the parts with their footprints and 3D models, and I suspect I either grabbed the wrong Bridge Rectifier, or flipped it. That is the good news though... The board is fine... Just labeled wrong. The solution to fix the problem is to simply flip the Bridge Rectifier! That's it! I'm just very glad I spotted the error! I'll just leave this example of what happens to capacitors when they are powered from the wrong polarity... As a side note, I may have found a suitable means to control the LEDs for my rate of rotation meters... The PCA9685. The outputs are persistent, so you can set an output LED or meter, and forget it, as it retains the output till the output value is updated. Because the outputs are also PWM, this chip can not only drive varying brightness LEDs, but also control analog meters! The PCA9685 is a 16 Channel, 12 bit, PWM Driver, with a highly addressable I2C Interface. Unfortunately the chips are also "semi-unobtanium". Unfortunately, most stocks are depleted for the chip alone. Fortunately, there are still many circuit boards available using the chip. Fortunately, you can still order servo controller boards using these chips, but it's one big board plus an inflated price, per chip. The boards are also not ideal for me, as I still need to design a compact board to drive the LEDs in the bezel that will surround the Navball. I'd have to desolder the chips, and at that point, I may as well just buy the one weird package in stock and do a board around that. 16 channels is a fair bit, but it'd still take Edited January 6, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
Lisias Posted January 6, 2022 Share Posted January 6, 2022 46 minutes ago, richfiles said: I'm just very glad I spotted the error! I'll just leave this example of what happens to capacitors when they are powered from the wrong polarity... When I was younger, there's was a place called SENAC where professional classes were available. It's a mixed government/private company dedicated to teaching things needed by the local Commerce. One of that classes was electronics, and visiting the place I noted the roof of that room was littered with small roles. I asked why, and before my friend (the instructor) could answer me something blew up behind me, on a student's workbench. Very didactic. Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 6, 2022 Author Share Posted January 6, 2022 (edited) 4 hours ago, Lisias said: When I was younger, there's was a place called SENAC where professional classes were available. It's a mixed government/private company dedicated to teaching things needed by the local Commerce. One of that classes was electronics, and visiting the place I noted the roof of that room was littered with small roles. I asked why, and before my friend (the instructor) could answer me something blew up behind me, on a student's workbench. Very didactic. I recall one of my classes at college, an AV repair class (I feel that this dates me... ), and I just so happened to have a friend come over to toss ideas for a troubleshooting session at me, and I was focused on him more than my soldering. He went back to his spot across the room, and continued working on his device, and I powered mine up, and POP!!! - - - w h i z - - - TINK - - - w h i z - - - OWW!!! My buddy had just gotten a capacitor can ricochet off his cranium. The capacitor, in the most perfect case of RNG ever, had managed to target the person who had distracted me from noticing I'd soldered it in backwards! It sought out it's vengeance! Edited January 6, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
AtomicTech Posted January 6, 2022 Share Posted January 6, 2022 Just now, richfiles said: I recall one of my classes at college, an AV repair class (I feel that this dates me... ), and I just to happened to have a friend come over to toss ideas for a troubleshooting session at me, and i was focused on him more than my soldering. He went back to his spot across the room, and continued working on his device, and I powered mine up, and POP!!! - - - w h i z - - - TINK - - - w h i z - - - OWW!!! My buddy had just gotten a capacitor can ricochet off his cranium. The capacitor, in the most perfect case of RNG ever, had managed to target the person who had distracted me from noticing I'd soldered it in backwards! It sought out it's vengeance! Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 9, 2022 Author Share Posted January 9, 2022 (edited) Much like the Harris Stereo 5 chassis fit was meant to be, so too, this was meant to be... This is the CPU/GPU heatsink from an iMac G5. Ancient tech, but this heatsink is built quite nicely! Had to be... The G5 was the toastiest of the Power PC CPUs! So, I woke up, and it was COLD here... It was -14°F/-25°C and windy out there! Yuck! I knew I had an old heatsink laying around, and I REALLY did not want to drive out into that weather to go rummage around in my storage unit... After checking a few spots around the apartment, I remembered I had the old G5 heatsink around... I had it stored 6 inches away from the California Intruments 850T Precision Oscillator that will be DRIVING this board! My next shock was the fact that a board designed by a stranger just happened to have mounting holes perfectly lined up with he studs on this heatsink! At least the width, anyway... I'd have to drill two more holes, but that's fine, cause both those studs line up with empty board space. Truly, this was meant to be! Wild! Man... Even the fins on this heatsink are made from copper! This thing REALLY is primo! It's got 3 heat pipes as well. The plan is to mount the transistors on the bottom of the board, and fold the leads 90°, so the thermal tabs face the heatsink. I'll be using silicone pads between the board and the transistors to compress them into the heatsink. I'll be using a thermal ceramic plate as an electrical insulator, so the transistors don't short to the heatsink. I ordered M2.5 standoffs to raise the board up, so there's room for the transistors underneath. When the board is screwed down, the silicone will compress the transistors, thermal compound, and ceramics into the heatsink. I've desoldered the bridge rectifier, and marked the transistors and removed them (they were only lightly soldered by one pin, to keep them in order)... There is one problem though... The board was designed with the transistors fanning outward, not inward. This heatsink is ideally fit for folding the transistors inward, but for the thermal tabs to face the right way, it flips the leads. Given that the boards were made so quickly, and were cheap, and I need to mod them anyway to work... I may just simply run a Rev 2.0 board. I gotta wait for silicone and standoffs and screws in the mail anyway... Why not the board 2. ===== EDIT ===== It has occurred to me, that all I need to do is desolder the through hole components, solder them to the other side of the board, and then just flip the whole board over, and the transistors will be correct. Rather than have two differing variations of the board for people's differing heatsink mounting methods, i'm gonna simply add the additional 2 mounting holes, and add silkscreened component labels to both sides of the board, so a person can build the board one way or flipped, depending on their desired heatsink configuration. Edited January 9, 2022 by richfiles LOL! These are such nice heatsinks, I bought another pair of 'em off the 'bay for 'bout tree fiddy Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 9, 2022 Author Share Posted January 9, 2022 (edited) So, I was looking for ways to control the bar graph LEDs for my Rotation Rate meters, and I'm learning that the old school analog LM3914 isn't actively produced anymore, and is getting a little pricy, and I don't trust Chinese sources that much, due to the counterfeit problem. It seems that people like to repurpose the very popular PCA9685 16 channel PWM 12-bit servo controller to drive LEDs instead of servos, and it works well, however, thanks to the parts shortage, that particular part is kinda hard to come by right now, except pre-built on Chinese sourced servo controller boards. To use the chip, I'd have to buy 15 boards, and desolder the surface mount chips, and reuse them that way. Less than ideal... Eventually I did find some in stock, but still, I wanted to see if there were better options out there. Honestly, my coming across these other chips was purely accidental, but apparently the same company makes a number of dedicated LED controller chips... PCA9532https://www.nxp.com/docs/en/data-sheet/PCA9532.pdf The PCA9532 is the budget LED controller. It has less features, but no fewer ports, and honestly, is probably simpler to control. I2C bus 16 outputs 8-bit (256 brightness steps) GPIO capable (unused outputs can be read as inputs 8 chips can be addressed per I2C bus (8x16 = 128 LEDs) PCA9955https://www.nxp.com/docs/en/data-sheet/PCA9955B.pdf I2C bus 16 outputs 8-bit (256 brightness steps) Advanced blinking controls Three quinary address pins allow 125 devices to be addressed (125x16 = 2000 LEDs... Woah...) PCA9959https://www.nxp.com/docs/en/data-sheet/PCA9959.pdf SPI bus 24 outputs 6-bit (64 brightness steps) Advanced blinking/fade controls SPI daisy chains data output via a series of shift registers, so no addresses. All three of these chips require a common anode, so they could only directly drive the red LED segments of my meters. TO drive the green segments, I'd have to invert each output using a transistor... That would be a LOT of transistors, but I haven't found a common cathode variant of these yet. Still, considering how many annunciators I'll be having on this thing, It might be worth it to use these types of chips, specifically because of the set and forget nature. Once set, an LED remains lit according to the characteristics programmed into the chip. This frees the microcontroller(s) from having to hold digital I/O high or low, or to keep up a PWM output. I figured I could share these with others. Maybe someone else can use them to light up their LEDs, or maybe someone knows a common cathode variant of this style part. I'm not nearly close to being through this companies entire offerings, but I've not had luck yet. Honestly, I'd love to know if anyone has had experience with these chips, how they liked using and controlling them, etc. Edited January 9, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 17, 2022 Author Share Posted January 17, 2022 (edited) And it's fully assembled! The silicone pads and the standoffs arrived, so I did the final assembly today. The silicone pads simply provide some compression to hold the transistors against the heatsink. The big pair of transistors use just one pad. The bridge rectifier used two pads. I had to cut pads to make a thin second layer to match the correct height of the smaller pair of transistors. Once everything was physically assembled, I soldered the transistors to the board and conformal coated it to protect it. What remains now is to do some basic testing to ensure that it works properly, and I can then move on to the next stage of the build. As I said in an earlier post, I'm in a board design mood, so designing the boards for the rate of rotation meters and for the synchro emulators should be next up. I may do both at once and order them together, or do one at a time... I'll feel that one out. The LED decoders for the DSKY are super simple... I'm not even sure if it's worth the cost of ordering boards. Might do protoboards for those, might not. I'm also getting pretty antsy about the Vertical Velocity meter. That's been just needing a diffuser plate and scales to be printed. It's disturbingly close to done, and I kinda wanna just get it checked off the list. Really isn't any good reason to put it off. The tape meter is functionally complete, but I still need to print the tape. That's going to be an exercise in tedium... To do that, I need to make a jig that can hold the tape in exact increments of each number line, and find an appropriate flexible ink. I need some manner of stamp holder on a hinge, that will apply even, level, uniform pressure on the tape. I'm also gonna have to practice plenty... I ONLY have 2 tapes... If I ruin them, I can't replace them! The final individual assembly for the instrument panel is the Delta V carriage meter, the type of part best done with an LCD I am choosing to do it the hard way! All in the pursuit of that sweet, sweet Apolloesque electro-mechanical glory! I have a few different concepts for how to do it, with none set in stone. A LOT rides on the available space left in the enclosure. The way I'm doing things, I know I will have some space left, but due to the variability and uncertainty of how it'll all come together... It may not actually be much space. That determines how wide I can go, whether it's limited to 4 or 8 digits, etc. What I have decided, is that printer parts are my best bet for a compact carriage mechanism. The Brother word processor I found ages ago has some nice parts from the printer that I think I can use. I also have the perfect cable for powering and transmitting date to the smart displays. Really flexible stuff. Once that's all done, it pretty much becomes time to wire and mount all the bits together inside the enclosure. The control panel is mostly just physical construction... Drilling, filing, etc, finished up with soldering the bits together. Ultimately, the biggest hurdle for me will be coding the software t actually run everything. Getting things finished and assembled, and having all the logical I/O mapped out will greatly aide in the learning process. I've also encountered many people who have expressed interest in the coding side of things, so I hope I'll have guidance as I learn. I'd like to finish this update with a quick thank you... I noticed, while editing my top post today, that I had the Thread of the Month tag. Sure enough, I was selected! This has been a long, slow process, but It's getting there! I just wanted to thank all of you... Those who enjoy the build process, those who have pitched in with advice, ideas, or help, thanks! I also wanna say it's been my pleasure to share in the actual process of creating all the various bits of this project. Edited January 17, 2022 by richfiles Thanks everyone! Quote Link to comment Share on other sites More sharing options...
Staticalliam7 Posted January 17, 2022 Share Posted January 17, 2022 That's one nice looking navball - surprised it wasn't more expensive Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 20, 2022 Author Share Posted January 20, 2022 Ah, me too! Still ate ramen for a month to afford it! I've mentioned before, it was originally part of an Israeli F-4 Phantom flight trainer. I think people were reluctant to buy it, due to it being international shipping, possibly. I spent $70 something in shipping costs alone... Honestly, it was well worth it though. I've never seen one go for as cheap in the years since I originally snagged mine. I took the gamble, mainly because it was pulled form a working simulator, simply because it was being retired... The F-4 is a very old jet, after all! Honestly, three axis navballs are just very rare on ebay. They don't pop up very often, and if they are flight rated, they'll generally be over a grand. Mine was rated for simulator use only, so it sold for way cheaper. It's amazing how many 2 axis balls are out there. They really do clutter up the search results, and it makes finding a good 3 axis ball very tedious. 2 axis balls are less useful for space applications, where orientation can easily be situated in all three axes. The two axis units also can be misleading too... If they utilize an integrated gyroscope that is vacuum or electrically driven, then the indicator is 100% useless. If it is a remote sending unit... that is a remotely oriented gyroscope sends signals to control the two axes, then it can still potentially be used... Displaying pitch and roll is typical for commercial and small aircraft, along with having a heading indicator display yaw separately. For space applications, one could have toggles to toggle yaw and pitch, or yaw and roll on a pair of such indicators, but it's universally less intuitive for space operations than a unified three axis navball. That's why I've been so determined to get it up and functioning! Quote Link to comment Share on other sites More sharing options...
richfiles Posted January 22, 2022 Author Share Posted January 22, 2022 (edited) Hmm... an idea... Adjusting tongue to most optimal angle for creative thought... So, the Navball is obviously the star of this show, and has been highlighted lately by my working on the power supply for it, and soon the controller for it. For the longest time, mounting it, along with what sits around it had always been in flux until very recently. I'm now quite sold on the current position and the current order of instruments, with the current bezel configuration including three rate of rotation meters and the velocity readout. One issue has been the fact that due to the angle of the front instrument panel, the rear of the Navball housing, and by extension, the tape meter, was likely going to stick out below the plane of the bottom panel, and require a "pan" to enclose. I was fine with this fact. One area of concern was what cutting slots for the rate of rotation bargraph PC boards would do to the overall strength of the Instrument panel. The Navball is the heaviest of all my individual instruments, and was designed to be supported from the front bezel. I was trying to find ways not to slot the panel surrounding where the Navball would mount, but now I think I might have an entirely different strategy for mounting. If I cut a "U" shaped channel in the instrument panel plate surrounding the Navball, and bend it inward, the Navball will have a flatter facing front (which doesn't matter much as the whole ball is round), and this will leave a full gap both at the bottom and on either side of the Navball. If I use a pair of brackets to secure the bottom of the Navball to the front panel to preserve the gap at the bottom while reintroducing rigidity, then I'll have a gap all the way around to support rate meters on both sides, and below the Navball... I can create a bezel that has a much shallower angle, and let the bottom rate of rotation meter sit at an angle that transitions well between the horizontal control panel and the near vertical instrument panel. It buys me vertical room, which makes mounting the velocity readout easier. It also allows a much more shallow mounting angle, and could help reduce the depth of the pan required to close off the hardware on the bottom panel. Reworking the tape meter only involves re-drilling 6 holes, where the belt frame mounts to the main housing of the tape feed assembly. Those holes simply determine the angle the two assemblies sit against one another, at the point where they both attach together. Hopefully you can imagine it... No time to do a sketch up of it. At the moment, I have circuit boards to solder for work again (Thank goodness)! I've not had regular work since October, thanks to the stupid parts shortage, and the bank account has gotten pretty low. I'll be doing more brainstorming than building for the next two weeks, at least. Man, I'm just glad to have work again! Edited January 22, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
richfiles Posted March 14, 2022 Author Share Posted March 14, 2022 (edited) Small update, one I am not happy to make... The Radar Altimeter Tape Meter experienced some unexpected lithobraking, and is now non-functional. Shirt sleeve caught a corner and yeeted it right off the shelf where I had it stored. The mechanism is completely bound up, and worse... One of the gearmotor leads appears jammed half WAY into the motor. I get nothing but high current draw when trying to power it. It's completely busted right now. The good news, is most of the warping appears to be the screws and standoffs holding the sandwiched layers of the mechanism together. If I take it all apart, pound the aluminum sheets flat again, and replace any bent hardware, it should be unbound again. As for the motor, rather than order a replacement from China, which will take forever, I'll just disassemble the gearbox from the motor, and as long as no teeth are broken, I should be able to replace the motor from one of the other ratio gearmotors I ordered (remember that I ordered the wrong ratio motor, initially, so I have a spare motor). Even if it is likely recoverable, it still sucks. I had only a limited supply of those standoffs, leftover from my old job, and if I have to replace warped ones, it could cost a bit to order them, as the local hardware store has a limited stock. That, and It's always annoying to have something that's otherwise finished, be set back so severely. Grrr... I'm mad... Edited March 14, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
richfiles Posted March 15, 2022 Author Share Posted March 15, 2022 The Kraken really seems to have done a number on this piece... Man, it is SO warped... Honestly, I don't think there's a single straight plate left in the whole entire mechanism. I don't think I can straighten the plates out well enough to not still cause binding in the bearings... I hate to say this, but I think I may have to start entirely from scratch with these plates... I think I'm looking at scrap metal here... Quote Link to comment Share on other sites More sharing options...
richfiles Posted March 15, 2022 Author Share Posted March 15, 2022 (edited) No, no, no, no, no... Oh man... It's worse than I thought. The electrical tab on the motor easily popped out, letting the motor turn again. That actually unbound the motor, but it still won't run smooth unless all the hardware is loosened, because of how warped everything is... Now that the motor spins... I can see that repair was pointless, the shaft is bent, and it feeds into the tightest bearing stack of the whole assembly. Motor might spin, but the gearbox output shaft is unusable. As annoying as this is, its just a shipment from China to replace it... The real issue though, is the PRIMARY REEL DRUM has a deformed tooth, and it now snags the tape itself. This is literally the worst possible scenario... This is actual damage to the tape feed mechanism itself. I can try to file it, but if that fails to fix the issue, I'm not really sure what to do. I might be forced to part out my backup unit. This is literally the worst case outcome. It's nearly a complete rebuild from scratch at this point, if I end up having to do that! I'm so mad right now... I've been busy with work, so I haven't had time to sit down and work with this. I wanted to get back to my vertical velocity meter and actually finish that, since it's so close to done (which I will probably do before this, cause I'm so frustrated and dang it, I need that dopamine high of actually completing something, vs being sent back to stage 0)... This is yet another setback to starting the Synchro Emulator DAC boards or starting the Delta V Carriage Meter mechanism... I just... I haven't felt this defeated by a setback in a long time... I put 7 months into building this assembly, and it took just a second to destroy it. I hate this so much... Edited March 15, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
Lisias Posted April 3, 2022 Share Posted April 3, 2022 Hi, @richfiles I think you shouldn't forget to add something like this on your project!!! Spoiler Cheers! Quote Link to comment Share on other sites More sharing options...
richfiles Posted April 21, 2022 Author Share Posted April 21, 2022 On 4/3/2022 at 6:46 PM, Lisias said: I think you shouldn't forget to add something like this on your project!!! Heh... I think I'll pass on that... I did help a friend install a not quite so large massager chair motor onto a car seat pulled from a car I scrapped We mounted the seat to a frame, along with a racing wheel and pedals, and wired it into his PS1 controller, way back in the late 90s. So, despite parts shortages, shipping delays, and logistics nightmares, my replacement motor for the tape meter just arrived, against all odds... Unfortunately, I've simply been dealing with way too much to even have a chance to look at it... Work is tied up in a product that doesn't have the best return on my time, and we are suffering from parts shortages on better paying work. Unfortunately, this means I'm struggling to just pay bills, much less blow money on parts. Work on the controller has been at a stand still. There was a recent family loss, along with planning for a family visit, so I've been helping my mother out with things too. Rearranged and moved things around at her place to set up her sewing room in a different space, to free up a bedroom for her guest. Honestly, I'm very satisfied with the setup. She likes it, and even her cat likes the new changes. Amazing, right! I'm happy things are coming together elsewhere, but it is unrelated to the controller, unless I ask her to sew me an instrument panel cover! For me... Things have been slow, and they've been a struggle. If things don't pick up, I will definitely need to find a second job, just to get by. In the meanwhile, I'm sitting here frustrated that things are still not getting done. I'm still around, haven't gone anywhere. I'm just so tied up and so broke right now. Quote Link to comment Share on other sites More sharing options...
richfiles Posted July 2, 2022 Author Share Posted July 2, 2022 (edited) Still alive, dead broke. Just still incredibly frustrated at everything going on right now... Oh yeah, and my computer got fried when the above incident occurred. I now am still alive, dead broke, and without my main computer!?!? I just... I can't even right now. At this point, I don't even know what even even is. My brain just completely blanks when trying think how to move forward again... How does this even happen!? God, I hate being broke... I hate my computer and my projects breaking. Edited August 3, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
Jack Mcslay Posted July 3, 2022 Share Posted July 3, 2022 (edited) What exactly was that incident? I'm confused how a piece of equipment apparently getting hit by a truck relates to your computer being fried. Also here's a little idea, why not track down a ThrustMaster Millennium 3D Inceptor for the right joystick? It's a 3-axis Space Shuttle joystick replica, making it much more immersive and functional than a pair of industrial joystick bases Here's the space shuttle cockpit for comparison http://www.collectspace.com/images/news-062611a/015.jpg Edited July 3, 2022 by Jack Mcslay changing embed image to link Quote Link to comment Share on other sites More sharing options...
richfiles Posted August 3, 2022 Author Share Posted August 3, 2022 (edited) I'm actually not a fan of traditional joysticks, in the sense of wrist operated control. I'm more of a fingertip control type. For me, the industrial joysticks will be far more comfortable to use for long play sessions, and more precise. I've recently considered that adding a gel pad to the bottom edge might also greatly enhance the comfort level. It's my one big "I don't need this to be 1:1 with NASA" contention. Regarding the computer dying... that's the main power transformer for my apartment. It was not hit by a truck... The foundation deteriorated from poor maintenance, and the ground literally swallowed up the transformer. When it collapsed, it cause a massive power surge. In the month that's passed since, I'm finding not only was my computer damaged, but other things, such as lights. Half the lights in my apartment have either failed, or are flickering and appear ready to fail. That surge was DEVASTATING to my electronics... Edited August 3, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
richfiles Posted September 11, 2022 Author Share Posted September 11, 2022 (edited) I should have done this YEARS ago... So, ten years ago, when I moved into my current apartment, I set up my workbench, and discovered it was WAY too low. Rather than take it apart to extend the legs, like I should have, I just took two different chairs, cobbled them together, and made a really low chair. It worked, for nearly 10 years... Thing is, the workbench was still way too low. I've always had issues with my back sitting at this thing. It was only 25 inches (63.5 cm) off the floor... The reason that this workbench was originally built to be so low, is It was originally used with a "proto-gamerchair" that I made in the late 1990s. Back when I was in college, I ended up scrapping a car, and saved the passenger seat out of it, and made a rather comfy computer chair from it. Had it for years, and I built my computer desk around that chair. I later also built this workbench around the same chair. After a decade or so, I removed the old seat, which had then worn out pretty badly, and replaced it with a newer car seat. When I moved to the apartment a decade ago, my computer desk and workbench were no longer adjacent in the same room... I could no longer just roll the chair between the two, and this required me to set up a different seating arrangement... It was not ideal... This year, I scrapped that car seat, and replaced the aging cushion of the replacement chair for the workbench. I've still not resolved the seating situation for my computer desk... Not that it exactly matters right now, with my computer still being dead... Thing is, I've not gotten much of anything done at my workbench lately, and I finally realized what the issue was... The new cushion... it was just enough added height, that the formerly borderline too low workbench was now harmfully too low! Every time I sat down to work, I'd be just in pain even 15 minutes in. Once I realized how bad this was, I knew what needed to be done... I jacked up my workbench, and installed leg extensions! The workbench now sits about 29 inches (73.6 cm) off the floor. It doesn't sound like much, but it really does make a huge difference! I borrowed a jack from a my father, and from a buddy of mine, and combined with my own car jack, I was able to lift the entire workbench, as it was, with all the equipment and tools left in place. I put the two smaller jacks on one side, and the larger one on the other side. With three jacks, I had three points of lift, and the workbench remained stable as could be throughout the entire process! It simply could not have gone smoother than it did, and I was so dumb for not just doing this years ago! My back already feels so much better! You can see the original height in this shot. I actually am tempted to try and get a new pair of car seats (probably find something from a junk yard someday)... Truth of the matter is, car seats are designed to be comfortable, even when sitting in them for long periods of time, and are made durable enough to last years. It's why I used my weird 1990s "gamerchair" for so long. I'm seeing a lot of comments that salvaged seats can be as little as $25-$100, and considering what commercial gamerchairs cost... and comparing the quality, this actually works well for me. I feel like if I find the right seat, I could really fine tune the overall ergonomics of this thing, and improve my comfort for long term build sessions. In addition to my workbench, I also have a desk in my workshop. This desk has been the universal "I'll deal with it later" space... And it has become terribly cluttered. Funny story... There was a box that sat on that desk since the first year I moved here... It was pushed back into the rear corner, and just sorta... took space. I finally cleared it out. Apparently it contained exactly one non local 1999-2000 phone book... That was it... I couldn't even be mad, cause it was so absurdly silly! One thing that has sat on the desk for years, pushed off to the side, was my Twentieth Anniversary Mac. It still boots up even! It's definitely in need of a recapping however, as the audio has developed a hum. Currently, it's configured with a PowerPC G3 CPU upgrade, 128 MB RAM, and the full TV/FM tuner/capture card. These old Macs had speakers supplied by BOSE, and sounded pretty decent. I mainly used to use this for playing music. I think it'd be fun restoring it... Dropping in an SSD and a network and USB card would make a world of difference. Networking can possibly be done with a USB network adapter, if I can't find the old Comm Slot II Ethernet card that this thing takes. I could also cannibalize the Comm Slot II card from another machine I have, if it comes down to it. More importantly, I am clearing off the desk because I want a large open space to actually work on large things, such as the cases for my two PC builds, vintage calculators or computers, and of course, the Kerbal Instrument panel itself. The workbench was built at a time when my primary work was making cable harnesses and soldering circuit boards. Hobby wise, I was 99% working on tiny robots and again... more circuit boards. As a result, I designed the workbench to hold electronic test equipment on a middle an upper shelf, and the main shelf has extra lighting, but also only has 7.25 inches (18.5 cm) of vertical clearance beyond the first 12 inches (30 cm) of bench depth (the bench has double that depth, but the second shelf severely limits the available vertical height beyond the first 12 inches of surface depth). This means that as I've shifted to wanting to work on larger things, I don't actually have a good space to do so on. The side wings of the workbench are open, but they are not that large either. Clearing this desk will make a world of difference for my ability to do large projects. This was the other project occupying my time... I wanted a high resolution ultrawide monitor that I could slide and tilt and rotate to either be right in front of my face or out of the way. When I need to have schematics, diagrams, blueprints, parts lists, etc. right in my face while I work, without blocking the actual workbench below, I can slide it right in front of me. Other times, I want to be able to slide the whole thing off to the side and out of my way, for when I maybe want to have youtube videos or some kind of background content going, while I focus on my work. The monitor is made from a pair of iPad Retina LCD modules, and a pair of HDMI to LCD converter boards. They are mounted behind a polycarbonate panel in a simple frame made from U-channel aluminum. The total resolution is 4096x1536 (two 2048x1536 panels), and requires dual HDMI signals to drive. The tablet is my only working PC right now, and it only has one HDMI out, which is why only one side is lit at the moment. The overall display width is about 16 inches (40.6 cm), and 6 inches (15.25 cm) tall. The bezel adds an inch to both dimensions, so the overall dimensions are 17 inches (43.2 cm) wide and 7 inches (17.75 cm) tall. These dimensions are very compatible with my workbench, and permit me to see the work area without interference. I did add a ventilation grill later on, because I noticed some heat being trapped by the enclosure. It wasn't extreme, but it's still good to vent it. I picked up a cable chain and installed a pair of long HDMI cables and a cable to carry power. I also picked up a 12v switching power supply, which will be set up on the bench to provide power to both LCD modules in the monitor, as well as LED lighting upgrades that I plan to install. The old florescent lighting causes visible banding on the rolling shutter of my phone's camera, and it makes taking pictures of my work kind of a pain. Since the LEDs will be DC, there should be no flicker and thus no more banding in photos. I haven't installed this yet. The slide rail and the tilt/pivot were made from a closet door slide and a GoPro bicycle handlebar mount. Pretty simple, but it works! I found a local-ish supplier for Teflon washers, and used a combination of fender washers and the Teflon washers to create the slides. The screws securing the slide rails are flush, except for the two on the end, which act as stops. Honestly, it works quite well! When I install the power supply and the cable chain, I'm gonna pull this all apart and hard wire the power across both controller boards. That will permit me to plug in only one power connector and have both LCD modules power up. I'd also like to make an aluminum port plate to cover the bottom. I may paint the frame as well. The overall construction is just layers of plastic cut out from an old Gamestop poster hanger. All the cuts were just done with an X-Acto knife and a metal ruler. I just layered the different layers together, and held it all together with screws. This part was fun to cut out. I wanted to enclose the boards and the button panels, but leave space for the ports and connections. The area with the buttons needed to be thinner, so I cut out the diffuser from an old LCD TV to close off those spaces and cover the circuit boards, while leaving the buttons accessible. I used a hole punch (the kind used to punch holes in paper) to make the button holes. Added benefit, is the LEDs on the button boards shine through the diffuser. Whatever works, right! I needed to raise the back panel by two layers to clear the tallest parts on the converter PC boards. This project turned out exceptionally well! Honestly, I'm hoping that with all the improvements to my workspace, I can get back to actually making progress. Cash is still critically low, but I hope to at least try to work with what I have on hand, at least for the time being. I still have made no attempt to repair the broken Radar Altimeter, aside from just ordering the replacement motor. I have not finished the Vertical Velocity Meter, nor started on the ΔV Carriage Meter either... I haven't even started on the DAC boards, considering my PC is currently dead. I have to get some stuff done for work first, but then I want to focus on finishing setting up the desk workspace. My first priority will be to get my main and my workshop PCs built. New chairs are also a high priority. After that, I hope to get back onto the Kerbal Instrument panel, now with the added benefit of SPACE to work with! Edited September 11, 2022 by richfiles Quote Link to comment Share on other sites More sharing options...
The Barryfish Posted October 9, 2022 Share Posted October 9, 2022 (edited) I've been working on a scale replica of the DSKY display. The one shown here (1:1 scale) still needs a little work (like the segment miter) but is pretty good (shown cut into a photo pasted on cardboard) It's driven with a BT816 GPU from an ESP-32 and is running about 32FPS Of course the color and details can be changed to any RGB888 color I'm using a 5" Daylight readable IPS TFT display with a 24bit parallel interface. It's a tiny bit taller than the actual DSKY bezel but should fit behind the front casting with modification. The next step is to CNC the front casting from billet AL. Comments please... Edited October 9, 2022 by The Barryfish Quote Link to comment Share on other sites More sharing options...
The Barryfish Posted October 9, 2022 Share Posted October 9, 2022 Here is the board. The copper tape and pink heatsink silicone are smushed onto the backlight driver as it needs a better PCB layout for heat dissipation. "mundane details..." Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.