Hempstead

Simple. I want the outside dimension to be exactly how I designed it. There are basically two ways of doing composites. 1. You make a core, then you do a “lose” lay up. That is, you lay up the fiber on top of the core and dabble the resin on top of the fiber. 2. you do something to compress the fiber/resin to to the form so you a. squeeze out the air pocket and b. you make fiber/resin more compact, which reduces the resin/fiber ratio. You want the resin to fully fill among the fiber, but no more. The more resin you have after that does not contribute to structural strength. Moreover, the more compacted the fiber, the stronger the structure. Vacuum bagging is commonly used in composite construction, particularly in experimental aircrafts. It’s easy to do, and cheap, produces even compression, and can easily be done in garages. Now, since I want outside dimension to be exactly as I designed, #1 approach is out. I could do forged carbon fiber by printing two half of molds and compressed. But, with the stick shape, I might need 4 pieces mold, and will require daft angles to be design in! I need outside mold, yet I don’t like outside mold. Plus, how do I form the inside shape? So, the original idea was to print the outside mold, 2 pieces, the place a structural aluminum tube inside and cast. Two troubles. 1. The epoxy volume is too thick… exothermal reaction, gets it too hot. 2. Compression is a bit difficult to achieve. And how do I get that thing out. I actually destroyed the mold trying to get it out. So, change of plan. How about just cast a thin carbon shell, with a thin disposable mold? Then destroy the mold to get it out is easier and takes less time to print. Then I can use other material to cast in the structural tube and form the inner shape. Say, high density expansion foam. The problem? I need to compress the thin layer of fiber and resin mix in order to get the thin shell. But, then if I use a ballon to blow up from inside, the thin mold would easily be distorted. I could cast some plaster outside of the thin mold to reenforce. Or I could CNC the mold out of two pieces of big chunks of aluminum. But, vacuum bagging will allow me to compress from both inside out, and also from outside in simultaneously so not to distort the thin mold!

Just came back from Asia last weekend... This weekend... dry test for Strechlon 800. Not sure if it worked... The whole idea is to see if Stretchlon 800 can strech up to compress all corners. Seems to when I peeked from one end (D=1"), but I couldn't see all the corners. But, once I cut it open... not sure if it did (might have been it did, but once the vacuum is removed, it sprung back???). The result is inconclusive. I mean, should I make the Stretchlon tube's diameter bigger? How big? Why Stretchlon? I could use an oversize plastic tube... like everybody else does. But doing this has the trouble of some "folds" of the plastic bag would lock in some excess resin causing undesired bumps and ribs. So, if Stetchlon can "stretch" up to all corners, then there should be no locked bumps and ribs of resin, b/c there will be no folds. The alternative is to use a kind of paint on silicon RTV to make a silicon inner sleeve so it's exactly the shape of the cavity, hence no fold whatsoever. Then use this inner sleeve for vacuum bagging. I am trying to avoid this approach, b/c it's more expensive tooling and processes, although I have already ordered and received the said paint on silicon RTV (it's expensive! And think printing a different mold for the silicon sleeve, mixing the resin, degassing... paint.. and correct the fit, etc. etc.). So, if I can avoid another whole different process, and use only regular vacuum bagging with plastic bags, albeit different plastic bags, the better. So, I ordered a bore scope . Now back to waiting for delivery. null

*imitates Nixon's voice * It ain't if I do it!

I guess I will keep trying to contribute to the compounded server load. Is there a script to automate this?

1. It ain't a SPI bus (although it's possible to drive it with a SPI module). 2. It is supposed to be a daisy-chain of 3x cascaded CD4021 chips, i.e. 24bits. 3. The firmware to read it... it will assume #2, and read 24bits. So, if you only have one chip, the firmware is supposed to "still" read 24bits off the chip. So, it depends on how you wire up pin 11 Serial In. Assuming you wire up the chip correctly, and the firmware is reading correctly, then, if you float pin 11, you most likely will get the correct answer for the first 8 bits, and 8th (0-index) bit and beyond are pure garbage. If you ground pin 11, as you are supposed to, then you get 0s from the 8th bit and up. If you wire it up to Vcc, then you get 1s from the 8th bit and up. Read, pp. 34 to 38 of Hempstick User's Guide, https://www.hempstick.org/download/manual/Hempstick-UserGuide.pdf, for the correct pin assignments.

Wanna bet whether A P P L E will focus on health craze like with the watch? It seems to be their corporate crowd.

nullMine is bookended by two marble plates. Don't buy them from kitchen supplies stores like I did... very expensive. Anyway... I used other methods like a long stainless steel rod to butt against two front legs of the table. But this 2x marble plates method is much better. 1. I get to position it anywhere I like, instead of where the table legs are. 2. The rear plate serves as a glide surface for my heels without wearing two tracks on my carpet, like most planes and cars do. Plus, the marble surface has just the right surface roughness for bare feet to glide on. Bottom line is... just some plates heavy enough, with the surface texture you like.

Oh... BTW... forgot to mention... this "disposable" mold is an experimental concept.... The idea is that this serves 3 purposes. 1. It's an experimental step to tease out the carbon fiber/epoxy molding techniques and design details, for instance the first molding kinda-success/kinda-failure led me to the two caps. 2. It doubles as an one-off manufacturing process most of do. I mean, 30 hours printing for a mold that you can only use once is definitely not for production. But it serves most home projects well... particularly it frees us from the worries about over hangs, draft angles etc. etc. Very liberating for design and manufacturing process. 3. Once #1 and #2 are completed.... I might consider using the one-off mold to make another mold for using high-density expansion foam to make the one-off mold (as well as using it as filler between the thin carbon fiber outer shell and the internal structural tube (or just the thin shell!). I may or may not go to step #3... But, it's been in the back of my mind during the design process. Just found out... I have always used Smooth-On X3D 3D print epoxy to smooth out the FDM mold etc..... to great successes.... I just found out... it's a completely redundant step to do in this particular process.... in fact, it's down right detrimental. Don't do it! Lesson learned.

The two mold end caps. They serve for two purposes. 1. to clamp and align the two mold halves, not for "capping" the resin/carbon fiber mix. The mold is only for outer shell molding using vacuum to compress fiber/epoxy to the outer shell... the two ends... no. But 2. to create a space and shape to prevent the outer bag from getting sucked into the middle and pushes the fiber/epoxy mix inward and causing some parts of the article starve of fiber/epoxy mix. Hence the two tubes on each end. They are there to create the extra space for the outer bag to get sucked in without reaching the fiber/epoxy mix. Also, they are there for the inner film tube to anchor and seal with the outer bag. In the end, I might forgo the metal clips and just use some twine to tie up the two halves. I mean, all I needed is a temporary tie until the atmo compresses down. What I need is more of alignment than pressure down. I am hoping the two end caps are reusable. We shall see.

null All printed. Still needs some end "caps" to prevent vacuum bags to be sucked in and distort the tow/epoxy mix. White one is PLA filament. The gray one is Phrozen 8K resin (not that it matters).

null

Thin shell disposable mold (half), printed on my Phrozen Sonic Mighty 8K... fits just barely in the Mighty 8K so I don't have to pull out my Mega 8K printer. I hate using the Mega 8K... just think how big your ISO alcohol tank has to be, and how man gallons of alcohol you gonna need to fill it! Using an 8K printer is overkill. a 4K or even 2K printer would do just fine, resolution wise. B/C, no matter what you do, this thing has to be heavily sanded multiple times. However, my Sonic Mighty 4K is just not big enough. Whatever, 8K it is.... It took about 3 hrs 48 minutes to print.... the same one being printed on my Ultimaker 3 filament printer will take total of estimated 13 hrs 50 minutes.

nullBeginning of 1mm shell mold. Still need to add reinforcement ribs, and some extensions to help prevent the unintended vacuum compression of both ends. I might consider using 1 of my 4 SLA printers for it for two reasons. 1. SLA prints hell faster than filament printers. 2. It's a disposable mold... SLA prints from the generic resin is fragile... but maybe it's strong enough for one casting. If so, then it's fragility might be an asset for de-molding, i.e. destroy the mold. I might have to worry about SLA resin inhibits curing of epoxy... A lot of PLA release film and/or wax perhaps? null null

Kind of works. There are some "forging" defects. So, next forging will correct that. However, the biggest problem is there is pretty much no drafting angles... so it's very difficult to extract it from the mold. I pretty much destroyed the mold (20 hrs printing each half). I might be able to cast one more out of it... but... that's not good. So, now the idea is... how about just offset the shell out, say 1mm or 2mm, and add some reinforcement ribs... that's the "disposable" one-time use mold! After casting/forging, just destroy the mold! Doh! Also, as you can see, the seam has some red release film got pinched.... these are some casting skillz I have to learn and perfect, etc. etc. But, in general, the carbon fiber prototype is heading toward the right direction. I might even consider printing a mold for casting the disposable mold with high density foam (quite difficult to buy very high density ones, it's a 16 lbs density 2-parts expansion foam that can give you a lot of details and a lot of strength). nullnull

I am changing the mold and construction method. The basic idea of forged carbon fiber with a Breakaway core construction is sound... but there are technical difficulties. The most important part is that there is various thickness on the shell, vastly different thickness; some parts are over 10mm thick, some barely 1mm thick. It's very difficult to get the right amount of Carbon Fiber tow at the right place and have even compression. So, it hit me... why don't I just keep the mold and do a thin even thickness Carbon Fiber tow shell on the mold... and vacuum bag it to create a compression force? The rest of the cavity... I can either pour in cheaper glass fiber epoxy mix or, hell, high density expansion foam. So, I am changing the construction method. I am going to use a thin shell forge carbon fiber construction method. I am sure somebody else invented it before I came up with it. Basically, I am going to put a thin shell of Carbon Fiber Epoxy mix on the two halves, just like in hand lay up. Close the two halves... put in the release film, absorbent fabric, and a plastic tube in the middle and close the two halves. Then, I will vacuum bag it. This will give me a thin shell of "forged carbon fiber" look. Then, I can put in the Breakaway core and "pour"/tamp whatever space left between the core and the shell. Perhaps I will even use a very dense expansion foam I have at hand (16 lb... your regular structure foam is about 2lb.). If I use the expansion foam... then, the Breakaway part would only be the "head" space. The tube part could just be a carbon fiber tube and have it embedded right in there. That is, if the 16lb expansion foam is structurally strong enough (it should... I guess...). Therefore, I chopped up the mold to facilitate the thin shell construction. Like I said many times before in other places... my F16 3D models are just models.. they are not design. This here is the design/construction process you have to go through after the 3D model was constructed... many times.

I decided to leave the bottom open... this way, once the two halves of mold are closed, I can pour in more epoxy/carbon fiber mixture from the bottom and use a cylinder to apply pressure.

I have always set the device resolutions on Quest 2/pro to max. Then, in DCS I set the PD to 1.0. This way, I can clearly read the Course number on the upper right corner of HSI in F16, without leaning forward, essential in flying a DME arc. The other way of doing it is to set the PD in DCS high, say 1.3, then set Quest Pro’s device resolutions to 1.0x. Try both and see which one you like better. I like setting Quest’s resolution to max better. However, you have to remember to dial this Quest device resolution down before you play MSFS 2020, otherwise you get a slide show, 5fps.

Congrats!

Instructions? No. But try this thread, There Sokol provided the replacement part # for the mini push button inside. They are quite cheap. So likely you will have to buy a bunch to justify the s/h charges. These will require some slight modifications, snipping of the leads, to fit. Opening up the throttle us easy. It’s opening up the HAT switch assembly without breaking the soldered wires that’s a bit tricky. To open the throttle, just split the two throttles. In between, on the right throttle, there are 3 screws to unscrew. Then on the right hand side of the right throttle, there is one more screw. Unscrew all 4 and you have it. The main wires going from the throttle base to the right throttle us a bit just short so, it easily come off during disassembly, and us tricky to get plugged back in during reassembly… so, watch out. Those soldering of wires to the HAT switches are very fragile, easily broken during disassembly. Quite a lot if the wires rely on hot glue for structure integrity. Breaking the hot glue blobs for disassembly often inadvertently break the solder joints. So, I would advise marking them with tapes so you can solder them back to the right points in case you break the soldering. So, make sure you have temperature controlled soldering iron and soldering skill, and a pair of tweezers before you open it up.

Yes... You are lucky that the Coolie switch is a standalone part that you can buy from TM, unlike the MIC/SpdBrake/Boat/China switches which form a 4-in-one PCB so you must buy the whole thing. Now, since your symptom said the whole Coolie switch stopped functioning... but not other switches, then most likely cause is either the Coolie switch plug got loosen or the power or ground wire is cut (say solder joint broke). See, the Coolie switch is simply constructed out of 4x mini-buttons in one housing. To the MCU they are 4x independent leads. The chance all 4x independent input broke simultaneous is very low. So, likely something common to the 4x inputs broke... hence my guess of either the plug is loose or pwr/gnd wire to the Coolie broke. I'd suggest that you open it up and take a look.

nullBeen experimenting with Carbon Fiber Composites.... Failed miserably several times in making the bend tube... but I am getting there... still a tiny amount of wrinkle. The next picture is... a mold for forged Carbon Fiber construction. WIP... still need to make the "compression" portion and hardware, like screws or clamps) to press down (the forged part of the method). Note the top and bottom "extra plates" are simply there to fix the location of the core, which a common forged carbon fiber part does not have. The idea is to print the core out of Ultimaker Breakaway... then once the part is de-molded, just dig out the core. At this stage, I can easily construct two separate forged halves. Then epoxy or screw them together. However, I prefer to experiment and see if I can make one complete part at once using this construction method.

I stick with Ultimaker filaments, because: 1. I have an Ultimaker 3 printer. All the settings, profiles are fine tuned/verified by Ultimaker on each new version of Cura. Not much tweaking is needed from me. So, I can upgrade to new version of Cura and be relatively sure that my setup is still good without have to tweak it again, and again, and again. My purpose is not playing with 3D printing. It’s a tool for me. I am trying not to fall into that old tinkerers’ trap — working on improving your tools more than you work on what you bought the tools for to begin with. 2. Buying other filaments, I do have a few, is a chore, and a time sink. You often have to experiment quite a bit to get it right for your setup. 3. Yes, Ultimaker filaments are more expensive than others’, but I get consistent quality I can rely on. Peace of mind, you know. I always have a spool of PLA and Breakaway loaded. I can just fire off a print job remotely any time I want from my desktop, and collect the print a few hours later upstairs. I am not recommending Ultimaker filaments. What I am recommending is avoiding the Tinkerers’ Tool Improvement Trap, which I am guilty of in many occasions. Set up one configuration that works for your printer (requires a lot of tweaking) and stick to it. And, if there is one kind of filaments I recommend against, it’s the no name cheap filaments on Amazon. Consistent quality is not a thing for those. Even if you get one filament from some XiLeader working for you, what about the next batch…? The seller might not even exist anymore, let alone consistent quality! Back to tweaking again? Ultimaker has been there as my supplier for filaments for the last 7 years. Will yours be there for you in the next 7? The more time you spend on tweaking your setup to reach a stable configuration you can rely on, the more important this question is to you regarding your selection of filaments. Whatever brand of filament you choose, make sure it’s one that you can rely on for years to come. How many years? Up to you. And I am praying the “merger” of MakerBot and Ultimaker won’t mess up my next 7!

First, make sure no controller other than the twist axis of TM160000 is the only one assigned to the rudder function. I do mean all controllers, including your mouse. DCS has a old “legacy” function of assigning any controller you plug in that has x, y, or z axes to raw, pitch, and yaw automatically “for you.” You plug in a new controller, it will automatically assign it for you again on all your profiles for all planes! These unwanted extra bindings could “show” up at weird time, with “undefined” behaviors.