Gadroc

Got the first CMSP panel assembled using the new EOS LCD Board. I have begun working on the firmware. First step is to get the EOS Library supporting the board which is under way. Once this is complete I will add a GUI in Helios to configure and bind it. Currently the library supports both driving the board directly via USB to serial cable as well as through an EOS Bus interface. The initial test firmware allows for control of the panel backlights. Over the next few days I will be adding support for polling buttons and brightness knob as well as setting text on the display. The protocol is already in place so it should be pretty quick going. Here you can see the panel backlight on and off. The backlight has 255 steps of dimming. The backlight is less even in real life. Picture makes it look better. Here is the messy back end. I need to cut the controller mount plate and do some cable management, but those will wait till the firmware is settled. The buttons in that unit are not good and I have begun cutting a new batch. First step for the buttons is to cut out a flange out of 1/16th clear acrylic and a cap out of 3/16th clear acrylic. Next I glue the cap onto flange with the cap centered in the flange. The flange sticks out about 1/16th of an inch on all sides. This prevents the button from popping out the front. Next I create a sheet of painters tape sticky side up. This masks the back of the button for light transfer and is easy to pick up and move. I will then paint three coats of white, followed by a coat of black to prevent light transfer and then three coats of gray. Once those dry overnight tonight I'll etch in the symbols leaving just the white paint. I'll then hit them with a dull clear coat.

Circuit boards came in today. I have assembled the first one. Of course I found a few mistakes, but so far I can correct them all with wiring thankfully.

Very cool. Excellent job. That has just got to be blast flying in that pit!

Well that looks like a performance jump enough to make me move off my 5870. I just hope they delver some eyefinity 6 versions as well. I need six monitors to drive my pit.

I was using the birch plywood from Lowes. Based on their specs online it's .703. I used a cut depth of .73 on it when setting up my tool paths. This probably varies by area as well. .75 wide notches would be to wide for the plywood I was using there would be significant "wobble" in the joints. Everyone should verify their wood before cutting.

RWR export is done not in Helios. If your using Loz profile you'll have to re-do the lua files he included, assuming they still work. The new install / patch would overwrite those lua files.

True you don't need to control the back light from the computer but dimming LEDs is not as simple as just plugging them into a dimmer like incandescent lights. You will need a PWM circuit and several transistors to supply enough current. Keep in mind you'll need twelve or more LEDs per board to get a good backlight depending on how your panels are constructed.

No you can't just connect them to one port. You could get fancy and do multiple, but you'll have to be careful and do some reading up on how to limit current to each LED individually. Also the current settings are global on the phidgets board so it would require you to put a series resistor on all LEDs connected to that board as well.

HOP, The most meaningful differences between the ACES II in an A-10 and F-16 are the seat angle and the ejection handle. In the F-16 the seat is at a 30 degree recline to help with the G-Force load on the pilot. The A-10 is only 15 degrees. If the A-10 is at 9G there is probably not much the pilot could do anyways ;). Second the F-16 has a side stick controller so the ejection handle is locate between your legs and the ejection seat has knee guards. In the A-10 the knee guards are the ejection handles since it has a center stick and the center ejection handle couldn't be used. I'm sure others could point out several more detailed differences, but if your planning on building those plans your probably not worried about that level of detail.

Awesome! Thanks again for sharing your incredible work. I plan on just having a fixed seat position. 99% of the time it's just me flying the pit and I'm not growing anymore (at least not in the directions the seat height adjust helps with).

Yea. The display data is not available in from LUA. Believe me if I knew how to get it from there I wouldn't be writing image processing code :)

Looks good. Unfortunately I don't have space for the projector setup, well not at least if I want to stay married. I already get complaints about how much of the basement I take up. How's the right side comming? Any chance of it getting posted this week? I'd love to be able to get it cut out while I'm on vacation the rest of the month.

Actually the problem is bigger than that. First that display does not just display flares remaining. It is also the display used when programming the pattern. So depending on how the switches are set it displays the pattern, or the flares remaining. While Terror's math seems doable outside the sim, it's not going to work in all instances. Mainly because you also have no way of knowing the current program. You can in theory always hit the reset button at the beginning of a mission to reset it to a know state, but if you forget there is no way to get it reset latter in the mission (flare counts will be off). Second if the mission is for some reason starts with out a "pristine" config the logic won't work.

I usually don't announce things like this. I have to split my free time across Helios dev, building my own pit and actually getting to fly my toys. Similar to ED done when it's done I don't want people to have the wrong expectations. I'll post more once I have it fully working reliably.

I have a plan for extracting that data via Helios, but it will have some limitations and has not been tested yet. I plan on exporting that data as listed here and doing some OCR on screen shots to convert it to 7-Segment displays. I will be working on the techniques as soon as I get the PCB boards in for my CMSP screens.

Good catch. There is actually a cheaper version which would work. http://www.oreillyauto.com/site/c/detail/CLH0/5011.oap?ck=Search_5011_-1_2103&keyword=5011

Let me know if you have any questions as you start building out some of the physical pit. I'm also located in Columbus.

All parts cut and glued. I've received the first of four part orders. Should get switches tomorrow. PCB will be end of next week or beginning of the next hopefully.

For those cutting out y2kiah's plan here's what I did. First I imported y2kiah's dxf files into CorelDraw. I then grouped all shapes together for a given part (drill holes, inner cuts, etc) and rearranged all the parts to fit onto 3'x4' and 2'x4' panels. I was able to fit the full left side on two 4x8 sheets cut down into 4 3'x4' and 2 2'x4'. I then went in and adjusted notch joints for the thickness of my plywood (.703 instead of .75). I then separated drill holes, inside cuts, pocket cuts and outside cuts into separate layers. Then I exported each sheet to DXF. I then imported those DXF files into PartWorks (VCarve Pro branded for and shipped with ShopBots). I set up each sheet as two jobs. First I ran the drill holes using a 1/8" drill bit using peck drilling plunging .38" at a time. The second job was using a straight dual flute 1/4" bit. I ran it at 19,000 RPM and 2.5 ips feed rate with a pass depth of .25". I cut pockets, inside and the outside cuts in that order. I did start cleaning up parts today and assembling the console. First since I used a 1/4" bit I had to use a chisel to square up the notches. If I had used a 1/8" bit this would probably have not been necessary. I also took a pass over all the parts with sandpaper to clear up all the fuzzed edges. I also had to sand down the tabs from the CNC cuts. Next drilled out countersinks for all the screw holes to assemble. I then attached all of the brackets to their respective ribs. Lastly I used a square and assembled the rear section. To do this I first dry fit everything and then used the square to square up the front panel and screwed it to the support struts. Next still using the square I fastened the rear panel. I then just screwed in all the ribs. Here is the assembled rear section with the top sill and throttle mount just sitting on it.

How does the throttle stand get attached? I see pre-drill holes on the rib it's at and the base of it, but I don't have a matching piece.

The orange seals are honeywell mil-spec switches, in particular the are TL series switches. Thanks to deadman I got them at online components which is the cheapest I've found them. In fact they are the same price as the "cheap" non-milspec switches I was going to use. I really need to spend some time documenting a few things for everyone. I've got several switch lists spread across my desk, as well as a few more panels. I will be posting all of my source files and parts lists for each panel.

Add 3000 to the button number. (Ex: Button_4 = 3004, Button_1 = 3001, Button_10 = 3010) Also you have to get the device first. local ahcp = GetDevice(7) ahcp:performClickableAction(3004, 1)

Also what size angle did you spec out for the panel mounts.

Yes. The I reset the notches to .72 wide on the side struts and ribs. I forgot to resize the ones for the support that rests in the notches on top inside of the console and it's loose but I'll put some shims in it.

I got the plans cut out today. Using .02" tolerance on the notches worked perfect.