Experimental EFIS

Peter:

 

Move the AOA indicator to the side of the AI protion otherwise you have too much clutter. 

 

The HSI shows a deflection indicator line that is vertical.  This should be in line with the other two green lines to idicate deviation from that "radial".

The engine monitor portion shows a continuous and thick colored bar showing the "value" to compare with the range.  I'd rather have the range  thicker and a pointer (could be a triangle) pointing to the area on the range.  Have the pointer change to red or yellow as appropriate.

Peter, that's a very exciting project.  Does your $1,000 price include the boxes to generate attitude and heading data as well?  For comparison, you should take a look at the MGL Avionics Xtreme EFIS/EIS/GPS...not yet in production but expected to be available this summer.  See http://www.mglavionics.com/html/xtreme.html.  Cheers, Matthew

Since this is a 'DIY' project the estimate is based on the materials cost.  Gyro's and accellerometers are not that expensive however here is the basic budget:

 

Inertial components (3 axis incl magnetometers) - design is for an Anaglog devices ADSI unit ~ US$400.00

Cpu board, including A/d converters and pressure transducers ~US$150

Display with 2 knobs for QNH, HDG ~US$200

So about 750 for the big ticket items, leaving about 250 for other stuff to mount the system into the aircraft.

 

Only rough estimates at present.

I am hoping to get the cpu board I am designing built without the AHRS unit and offer it as an assembled board at cost.

Someone out there may want to offer a more complex 'kit' and I would probably support them if they were reputable. but for now its a community project.

 

regards

 

Thanks, it is a bug

Must have crept in somewhere, will fix it.

I have been working on the gauge code to allow for different styles.  There is bar (what is show), pointer and vertical.  WIll add another display type of SmallIndicator for a pointer.  Would probably look quite good.  The G-Meter is currently a pointer type.

One thing that is not shown here is that my concept is to provide a bluetooth interface on the hardware to allow you to change the display ranges/settings.

I don't like the idea of people having popup menu's or buttons on the primary display.  Enough distractions occur in the cockpit anyhow and fiddling with a EFIS while in-flight I think is just bad airmanship, so the AIO is intended to just have 2 spin buttons, one for QNH and one for HDG.

So the idea will be that you doa flightplan with your laptop, customise the display with a program and then you upload it using bluetooth to the AIO and go flying.

Another idea is that by using a bluetooth GPS (about 60.00) you can then have the GPS/Flightplan work.  I am working on the gps display now.

 

The position of the AOA will be able to be set with the configuration app (left/right/center).

I like the idea of having the roll/pitch/aoa all in the center as these are the 3 critical things on this instrument, personal preference only.

 

regards

Peter,

Can you give me the instructions for running your AIO program on a Linux system?

 

Thanks!

 

Glen

There are 2 binaries on the sourceforge page, 1 for linux x64 (I have not built it for i386) and a windows binary that will run on windows XP.

These are just demo systems at present and will be updated tonight with the latest build.

Otherwise you can build from the source tree if you know how.  I am setting up a wikiwiki to give more info on this, but for now moving the code along is the most important to me.

The url for sourceforge is:

https://sourceforge.net/projects/diy-efis

regards

I have a PC104 CPU board with a 900 Mhz Tulatin Celeron processor and a 1024 x 768 VGA video output. The OS is Red Hat Linux 4. Any chance that I could get a copy of your source code? I'd like to compile your EFIS software and see if it will run on my equipment.

 

By the way have you seen the OpenGC page at: http://opengc.sourceforge.net/screenshots.html

 

Hi there,

 The source code is on the sourceforge site.  At present I have not created a tarball to build from as the project is under heavy development.  I am hoping to bring a demo one to Oshkosh this year to show anyone who is interested.

 If you would like to build from the snapshot you will need to use subversion to get the code.

 1. On your redhat box install the gnu toolchain and include subversion

 2. There is a basic linux hal that uses X11.  I am working on the ecos hal at the moment and expect to have a framebuffer version running at Oshkosh.

 3. get the code: svn co https://diy-efis.svn.sourceforge.net/svnroot/diy-efis diy-efis

 4. do the following:

cd diy-efis
automake -a -c
 make
./aio

The AIO screen should now come up.

Note however this is un-stable source.  You may have compile issues which are best handled by sending me a message using the sourceforge lists.  I am currently building under windows and it sometimes breaks the linux build.

As far as the glass cockpit project goes.  It was written to be a simulator for flight sims.  It uses openGL which is not suitable for use on an embedded system.  The diy-efis is intended to run on low power basic systems.

 

regards

 

 

 

Hi Peter!

 
I was able to use Subversion to get your source files from sourceforge.net. The "automake" ran ok but the
"make" generated a long list of errors. I'll try to troubleshoot the errors over the weekend.
I already have a Mircrostrain 3DM-GX1 AHRS that I plan to interface with your program.

http://www.microstrain.com/pdf/3DM-GX1%20Datasheet%20Rev%201.pdf

Thanks for sharing your source files!

Glen

Looks like a good unit.  The source code may be broken at present, sorry about that.  Try on saturday and should build ok.  I am changing the xml parser so that we can load flightplans into the gps unit.

 

Sun 25th-April

I have checked in all of the changes.  If you get a copy from sourceforge it should now build ok.

You may need to get libx11-dev if you don't have it, on ubuntu:

sudo apt-get install libx11-dev

also in the root of the diy-efis you will need to make configure executable.

Here is a sample build setup for ubuntu:

----------------------------

sudo apt-get install build-essential libx11-dev automake autoconf subversion

svn co https://diy-efis.svn.sourceforge.net/svnroot/diy-efis diy-efis

cd diy-efis

chmod +x configure

automake -a -c

./configure

make

src/aio

---------------------------------

I am also uploading linux 32 bit binaries to the project

 

regards

 

Thanks for making this!  It is way cool to see my two hobbies come together - technical geekery and now airplanes.

 

I'm tempted to make something similar in Java, perhaps Java ME.  How did you do your hardware selection process?

The AIO is intended to be hardware neutral.  I am trying to use off the shelf components.  I have ordered an 6.5" panel from http://www.winsystems.com for the base panel and am working on a pc104plus daughter card for the gyro's and airdata.  The engine controller I use for my aircraft http://www.linkecu.com has all of the engine data in a serial data form.  The gps interface that I am doing uses NMEA format  which is standard.

Equally the ahrs/airdata could be replaced with an off-the-shelf unit as well, such as a crosbow unit but the price is diabolical.  The only thing that is needed is to write a driver that converts each piece of data into a 32bit message that the aio can use.

Thr DIY part is really that you can use standard hardware and interfaces and then write a fairly sinmple driver to make it work.  The AIO code base will be targeted at either windows or linux/ecos.  This means you can run build it for:

• WIndows -XP
• Windows-CE
• Linux-X11
• Linux-Poky/X11
• Ecos-Framebuffer

The code base currently compiles for most of these environments.

I am hoping that if enough people get interested they will contribute any drivers they write back to the project so making it easier for anyone else to use that hardware.

I intend to keep the code base maintained for the  core O/S and of course the hardware in my aircraft.

regards

 

Have you looked at the products available from DIY Drones (http://diydrones.com/)?  They make sensors for RC airplanes, but I don't know why their stuff couldn't be used in a full scale aircraft.

Check out the ArduIMU+ V2 board they sell for $100.  It has 3 axis accelerometers and gyroscopes and includes open source firmware.  For airspeed, they sell the ArduPilot Shield V2 for $36 that has a differential pressure sensor.

Sparkfun Electronics, http://www.sparkfun.com, sells two different controller boards, ArduPilot ($25) and ArduPilot Mega ($60), that are intended for RC UAV applications, but might be suitable for this application.  They have open source software available, so maybe it could be modified to just send the data to a display board.

Best Regards,

Brian

Great project and as a open source project hopefully we will see continued improvement.  Personally I prefer bargraphs for temps on all cylinders, CHT and EGT.   The new Dynon Skyview has the right idea for form factor.

Mark

Thanks for the feedback mark.  I will add a bar-graph display to the gauge base class.

 

At present I am getting ready for oshkosh, and hope to bring a computer with the software running on it with me.  Don't know if it will have any avionics hooked up, aircraft not finished yet

 

One think I am hoping to bring is the client application that will allow the uploading of flight-plans and customisations to the display.

 

Hopefully I can catch up with anyone who is interested on the project.

 

regards

 

Homebuilding at its best. Love what you're doing!

Oh, and I'd recommend a solid state hard drive. They've gotten much cheaper in the past couple of years.