CRJ Simulator Rehab – Part 3

Once again in our ongoing segment on how to bring a trainer back to life... we begin where we left off... with nothing in the simulator (buttons, switches and the like) except the glass working....

Step 7 - Phidget and Project Magenta Land

So here we were and not a switch or button to be working... I knew that I had all these phidgets interfaced to the buttons, switches, the yoke and the pedals and that they all talked to the phidgets thanks to a small piece of software that Rick wrote (thanks Rick - you are the bomb).

So looking through the folders on the systems, I keep finding this "phidget.txt" file... so I open it. OMG! It has a list of phidget serial numbers to their related I/O channel to the system variable for FSUIPC. OK... so now the fun begins...

I find every last phidget.txt file that I can lay my hands on... there is a decade worth of backups in the hard drives... and merge them... so that I can figure out who went where and why.

Digging and digging.... got what seems to be the correct phidget serial number attached to the correct system variable. Made one complete file. Put file in the folders where they should be. And... nothing.

So digging some more into backup and reading the Project Magenta / FSUIPC manuals, I find a program called pmSystems... it runs the logic for the aircraft... hummm... not one computer runs this software. OK. Let's copy to the Prepar3d computer and see what we get....

OMG!!! I can turn on the aircraft...

from the overhead panel...

not all of it works...sigh

Step 8 - Fixing the Damaged Aircraft Logic

So this is where we separate the men from the boys (or women from the girls as the case may be)... what is working with the logic. The pmSystems software is kinda cool in the fact that you can see if a switch moves, and what system variable it is attached to, but that is about it. Go back to FSUIPC and find out that they gave Project Magenta their own set of HEX offsets for the software... and nothing is documented... kinda...

To me, if someone was handing me a bunch of offsets and saying here you go, I would have grouped them - 737, A320 and CRJ and then noted it in the system variable file. Nope... not today and apparently, not yesterday either.

So poor Rick gets to have fun unfurling the mess. Figuring out what panel goes where, what is does and making sure that what system variables that are triggered work is time consuming. Remember how many phidgets there are... well each digital has 16 channels, each analog has 8 channels and the LED's have 64 channels.... wow... and when we got done with the majority of it, we only had two digital phidget boards (32 switches) that had no system variable attached to them.

Now, we have to make a decision - do we get them all working or simply get the trainer back to it's original level which was 4. I'm all for getting every single switch to work... so that is what we are doing.

So I left Rick to get the System Logic working again... he's having fun.... We found the landing gear! (now the plane won't crash in the sim) Throttles are working! EICAS is working! APU is working! FMS is working! Glass is showing what it should show. We still have a few switches to do.

Step 9 - Autopilot and Glareshield Electronics box

So at this point, we have found where the auto-pilot is hooked up, but it isn't working. The master warning is always on. So I check the auto-pilot for a phidget file and sure enough, there is one! The serial numbers are wrong and the file is formatted the same, but never mind at this point, updated the serial numbers and added the encoders. Rick gets in the cockpit and we get in the air... turns on the auto-pilot and TA-DA! Auto-pilot works!

Figured out that the electronics box (that has the phidgets) for the auto-pilot and the glareshield have to be connected to the computer that runs the glass. Have no idea why, but it works. Wasn't originally hooked up that way - beginning to wonder if the people that used it before didn't know what they were doing and trying to fix it...

So that offloads some of the phidgets from the Prepar3d computer. Prepar3d talks to the Glass computer via WideClient... I bet that it is handling that. Oh look! WideClient has an INI file that allows you to load software like an old style batch file. Cool... let me modify this and get the right software in the INI file and put WideClient in the startup folder for Windows and TA-DA! The software for the glass loads when Prepar3d loads up.

Step 10 - The Radios...(groan)

The phidgets for the radios are spread between two different phidget boxes. Hummm... When the software gets moved to the new computers, I kinda want the Glass computer to run the radios. So how to get this done. They are identical and use the exact same system variables. I wonder if I can attach a system variable to two different phidget serial number and make it work? Splitting the radio display with a splitter to go to two different places is easy and keeps the amount of software running to a minimum... wonder if that will work? And the Glass computer is dying... sigh.... time to move the software to the new Glass computer.

Step 11 - Check out

So this will happen in the next installment.... I have to get a pilot over here and see what he thinks, but Rick has to figure out the logic first....

Till next time.... may the wind be every beneath your wings!

CRJ Taking Off

feedback control loader, control loading, actuator, FAA Level D, flight controls, simulation, simulator

The Overhead Panel - That basically does everything!