y2kiah

Environment Panel As you can see in the picture, the two round dial instruments will be routed directly into the panel's light plate. Micro servos will sit directly behind the back plate to turn the dials. The raised bezel is a two-piece, the clear cover sandwiched between them. The instrument face plates will be done with a PC printer. The two rotary switches at the bottom of the panel have engraved discs that I will fabricate and glue to the bottom of each knob. The last issue is the toggle switch at the bottom of the panel. The 4 position switch required would be very difficult to find and probably very expensive unless found used. Instead, I'll plan to use a simple pushbutton which toggles each mode (off, auto, man cold, man hot). I'll put an LED next to Auto, Cold and Hot labels on the panel to indicate which mode is active. The whole panel is pretty much for show, it won't be used much when flying in DCS Warthog, even if the switches are interfaced. Though, with the new smart bomb capabilities of the A-10C, flying at higher altitudes will become more common, so maybe I can play pretend with these controls :)

EWMU, Electric The EWMU panel, still haven't found the perfect display for it yet. Basically I'm looking for a bright 16x2 character LCD that fits the space. Also, the text color has to be bright (like green, white, blue or yellow) with a dark background. It's easy to find the reverse with dark text and bright background, but that just won't do. Some 14-seg LEDs or dot matrix LEDs would work too, but those tend to be expensive. Any suggestions? Next - the electric pwr panel. This was just a mirror image of the Aux Lighting panel, with different cutouts of course. And finally, another peak at the whole pit. This is being built piece by piece, I'm not just throwing polygons together here. This way, when it comes time to build I will have an exact parts manifest and cut schedule in the model, and I will avoid nasty little fitting problems that pop up during construction.

No, not in both dimensions. The A-10C MFDs are 5"x5", and the 8" display is ~ 4.76"x6.38". If you were to use the Thrustmaster MFDs, they are smaller so the screen would probably fill those up. edit: oh you said couger MFDs... sorry in that case I don't really know

I bought 4 feet of 3/8" round aluminum bar stock from onlinemetalsupply for something like $4. I should be able to get about 120 screws fabricated out of it. After I cut the bar into 1/4" lengths, I'll put a hole through each one and countersink for a 1/4" or #10 screw, whichever fits better. It will be a LOT of tedious work, but SO much cheaper than paying $.72 per screw. I have a feeling this will be one of the last things I will do for the pit.

Just got a circuit built with a MAX7219 connected to my mega, allows up to 64 individual LED outputs using only 3 pins! The ICs can be cascaded for even more outputs! I'm using it with 7-segs right now, for the comm and USB panels. I will also drive all indicator lamps in the pit with this method. I'll post a vid early next week when I'm able to.

That's a very cool board, I like that it's a HID device. I might pick one up to check it out, maybe use it for my keyboard/trackball panel. The reasons I don't choose USB for general pit IO are many. First, the number of available USB ports is limiting. You can get a USB hub, and for HID devices it may work fine, but you're going to have a LOT of cables to plug and I think you'd need a lot of hubs, maybe even have to daisy chain hubs. Second, using HID for any kind of output would require at least one more layer of software, and more difficult on the programming side than TCP/IP or serial. Whereas with TCP you can open a socket right from a scripting language like Lua and send information to/from the microcontroller directly, with a usb hid you probably can't (unless a library exists that I don't know about). The sim's ability to use OS recognized hardware for input events is great, but you're going to have a LOT of devices, and there aren't any similar output events with this "front door" method. That's why for general pit I/O, I think the back door is a better solution. Sweet, I might take you up on that some time. Sounds like it would cut through 1/16" aluminum plate like butter!

looking good! which ACES II kit did you use, or did you diy? What was your choice of wood for the seat? it looks nice edit: on second glance... looks like MDF maybe?

I was planning to make them myself. If so, no problem I'll make you a set. Looks like Alex_rcpilot might have an option we may all be able to take advantage of? We'll see what he can set up. Nice work on the CDU! Lucky man, I imagine there must be a factory on every corner in China, like Starbucks here. We..... don't really have factories here anymore :-) We gave them all away. Tell me more about the button pad idea. Is it kind of like this one? That pad has LEDs in the center of each pad, sounds like a perfect way to backlight the caps if we could get a PCB designed using the same method! I am thinking TFT screen also, running the visuals of the screen on a PC. For me, I ONLY consider this an option if it is not the main sim PC running the peripheral visuals. A second PC on the LAN would have to run the visuals, in a BSVP-like fashion. I'm also thinking about using VFD character screens instead, ever since it was suggested earlier in this thread for the EMWU panel. That might be a little trickier to find the right part, or fit multiple parts into the given space. The LCD/VGA route would certainly be easier and give more control over the screen output to make it look genuine. On the other hand, making a self-contained unit with its own microcontroller, rendering its own visuals, and hooked up via ethernet only (not relying on any connection to PC directly) seems very desirable to me. The frames around keys look raised to me in a couple pictures I have of the unit. They don't look raised in DCS because they are just textured on. I just raised them to the same level as the key tops (I think 1/16"). My plan is to use polystrene plastic strips from the local hobby shop and glue to the acrylic plate. Once painted, they will blend perfectly.

KY58 and CDU-900 I made the CDU light plate 3/16" thick, so I will have to bind 2 layers of acrylic together. This is so the circuit board holding all of the tactile buttons can be embedded within the light plate. The buttons on the far right and left of the panel would cause it not to fit between the rails. There are two other alternatives if embedding the circuit doesn't work out. 1) I can cut the rails in those particular spots to allow to panel to fit, or 2) I can move the buttons inward until they wouldn't cause a problem. Also, back lighting the buttons will be impossible, so I was thinking about side lighting them, and running EL wire through the hollowed out light plate to achieve that. Anybody think of a better way?

Yeah the screenshot I took shows a 1/8" rim which might be a bit too much to use those controls. I'll try to get a little closer to the surface when machining to help that, maybe leave 1/16" thick? I'm also thinking about throwing together a custom design for the IFF panel where the top half is basically replaced by a trackball and keyboard number keys to be used for communication and mouse control in DCS. That way, my hand won't have to move far from the throttle for communication, and my right hand won't have to leave the stick. Just a thought at this point, I think it would actually be better if there was some way to communicate right from the hotas. I'll be using arduino for my pit, with the wiznet ethernet shield. I'm also thinking about using eagle to design an input matrix shield for the arduino that will help me save some pins. The software that I'm currently writing will allow you to designate pins for certain types of controls and pass those commands to the card via ethernet (with the wiznet shield) or serial. Then, the card will start reporting events for the enabled pins based on the type of control you say is attached. I'll package the software up as an open source arduino library, so it can be used by any interested pit builders. Keep your progress pictures coming too Gus, your work is an inspiration for sure. Ah I see how my picture makes it confusing. I actually only separate into 2 layers for the model, because I have to cut front and back of the light panel differently, and designing 2 layers makes it easier for me to export for tool pathing. I still plan on one piece of 1/4" acrylic for the light plates, and flip it over for the backside cuts. The EL sheet will sit between the light plate and metal back plate, and the screws to hold the light plate on will go all the way through. Do you have any more leads for making the panels? You got me thinking about that option and it's worth looking into I think, especially if they can do engraving with laser. My tests engraving with my machine turned out adequate, but it could certainly be better if done with professional equipment. If I look very close I can see some squiggles in the lettering, probably from vibration in the machine - but from an arms length away, it looks good enough, for me at least.

Yep, and the mode 1 digits are also 0 to 7.

IFF Panel You may be wondering why I didn't include slots for the thumb wheels. That's because I bought these BCD switches instead. You also may be wondering why I'm bothering to make the IFF panel at all, because apparently it will not be simulated in DCS. I'm going to use it mainly for the civilian modes (3 A and C), because in FSX (and maybe XPlane) I will need some way to select the transponder code. I can also use the switches for another purpose, such as toggling miscellaneous simulator options.

I was thinking about just stenciling them on. Would have to create the stencil first... maybe use some thick photo paper and cut with a razor. How were you thinking of doing it?

Thanks Alex, I use Sketchup. The yellow stripes took a lot of clicking, that's for sure... it didn't help that I ended up doing it all twice, but that's another story. I used intersecting lines, erasing parts outside of the stripes and painting the regions inside. It didn't take as long as it may appear though - Sketchup has a nifty keyboard shortcut for repeating an action X times. So, I would place a region then copy-move the lines x 20 to fill up a stripe in one shot. I've used other modeling tools like Blender, Milkshape, and some other freebees which are more powerful than Sketchup, but I tend to get bogged down in the polygon-oriented process of those tools.

Three more panels down on the left side

Hi Alex, I think you have a good idea with the 20x2 VFD. It would be a lot cheaper than getting the dot matrix type LEDs, and will still look very good IMO. Photo-Etch makes the panel - at least it appears they do for simulation. Pics from their site: and another panel they make appears to use the same type of display, shown lit up here:

From official sources, all the panels are 5.75" base plate width, and the light plate I reduce by .05" on all four sides. For length, it is always a multiple of .375" (also the diameter of a DZUS screw. So the quickest way I determine a panel length is to eyeball how many DZUS would fit in the empty spaces. For some longer panels, this can be difficult, so I try to gather as many pictures from different sources as I can, and use the pixels to measure. Panel length = Height(px) / Width(px) * 5.75" Then round that to the nearest multiple of .375 and I usually have it. Sometimes, if the perspective on the source photo was bad, it will end up being short or long by 1 unit, but that becomes apparent when laying out the panel. This reminds me of something that bugged me for a while. I couldn't get my panel layout to match up with the DCS screen shots. I had the hardest time figuring out the length of the VHF Comm panel. It's so short that you would think it should be easy, but I was getting big discrepancies between source photos. I finally decided that the screen shots from DCS A-10C were the source of the error, I'm 99% sure that ED made those panels 1 unit too long. Because of this small error, it leads to a compounding error on the left console with the way panels align as you move backwards - so by the time you get to the last panel, it doesn't match up with its neighbor the way it does in the DCS screen shots. It's not a problem, just something I struggled through at first. If you haven't seen it yet, http://www.strandedduckling.com/ is a great source. http://www.aircraftresourcecenter.com/ also has some good photos of the A-10 cockpit.

Thanks, it's Sketchup actually. I export parts as DXF right into cambam, and then onto the CNC. The height of the console is actually an unknown at this point - the height I'm using for the model is just temporary until I figure out how thick my base will be, and how high the ACES II seat will sit, etc. For the time being, the flat inner panels sit 12" off of the ground. The rearmost rib reaches a max height of 21". The throttle panel slopes about 5 degrees. Mine actually slopes at 4.8 degrees because the TM HOTAS Warthog unit includes the LASTE panel which normally wouldn't be sloped, causing me to extend the sloped region to include that panel. The Fuel panel slopes up from that another 12.5 degrees, so 17.3 deg from level in my case. I was basically shooting for something close to 18 degrees, and this is how it worked out after adjusting for the throttle panel. The outside panels do slope up at exactly 18 degrees. These are just my measurements taken from blurry photos, so they are not super accurate. The basic rib shape is shown below. They get shorter and slope in more as you move forward. Again, still a WIP.

I gave this one some special attention, for your viewing pleasure... Also, another sneak preview of the pit design. The left side is starting to fill up. I've ordered all the metal sheet, angle, and plate that I'll need for this console. The ribs will be cut from ply or MDF. I'll be able to finish the design and start construction as soon as I get dimensions of the TM HOTAS Warthog throttle unit. Hopefully I won't have to wait for its release to get that info... anyway, enjoy

sorry can't answer the original question, but wanted to let anyone know who may be interested - I will have a complete solution available in the near future for those who want to use Arduino for their pit. Switches, buttons, rotaries, encoders, servos, steppers, 7-segs / LEDs... all of them will be implemented. I'll start a thread here when I'm ready to post the download.

Another panel: I will use the HDSP-F503 Green 7-Segment LED display with a MAX7219, which conveniently will allow me to multiplex all 8 7 segs on this panel using only 3 microcontroller pins. I will use rotary encoders for setting frequency, and maybe try my hand at a CNC'd PCB to mount everything on.

Hi Alex, he means this kind of switch You simply need to add your pit components to your home network, no need for more than one ethernet port on your computer. Since you have two ports on your PC, your pit network could be completely separate, and your Windows machine could belong to both networks simultaneously.

Your pit is awesome! I happen to like your MFD's, but agree that having some EFIS style gauges in there would look better. If you're serious about getting into graphics programming to do something similar to BSVP, I can suggest several paths for you. First, head over to gamedev.net to start gaining the knowledge. Second, you have several options on where to start. You could use a managed language like VB.net or C# with a graphics library like XNA framework, SlimDX, or even WPF. Another managed language, or so-called scripting language, with a common graphics library is Python. Your other option is to start at a lower level in an unmanaged environment with the C++ language. A little steeper learning curve is involved, but with free libraries such as SDL, Ogre3D, Unity (the list goes on and on) to choose from, it takes a lot of the pain away and partially bridges the gap IMO. The last option would be to work at the lowest practical level, and use the OpenGL or Direct3D libraries directly to create your own simple graphics engine. There are also many high level networking libraries available to code your communication with export.lua. Get yourself Visual C++ 2010 Express Edition, C# Express, or Python to start fooling around if you haven't already.

Note to self, do NOT use heat gun to speed up paint drying on panels. Will cause panels to warp!

have you tried updating the shared texture less often? Call LoUpdatedSharedTexture("mfd1"); every tenth frame or something, instead of everyframe? You would sacrifice frame rate of the exported image for frame rate of the main simulation. This could possibly cause a stutter in the simulation, but may not be noticeable, and may improve things overall. This is just a theory, I have not tried it myself.