I’ve started working on one of the fundamental instruments on the plane: Attitude indicator.

The ones I’ll be using are from a 737-200 cockpit, and one can be seen on this close up from the cockpit:

Once opened, we can see the instrument is not difficult to interface with, it is basically a bunch of motors with some resolvers for position feedback. Also there are some coils for driving the localizer, glidescope, several flags and so on.

Once I’ve examined it in greater detail, though, a difficulty arised: The motors used in the instruments are not the usual DC motors, but some AC motor-tachometer arrangement. That accounts for the expected reliability and precision on the instrument. Essentially, the tachometer can be used to dampen motor movement.

So all I know if I have some nice small motors I need to move, and they have 10 wires coming out of them:

Fortunately I found some datasheets from a different manufacturer, and I was lucky enough the cable colors matched the motors I have, which I confirmed just measuring DC resistance on them.

On the motor I can see: Control winding: 26v/400Hz. Reference: 26v/400Hz, Tacho: 26v/400Hz.

I have all the details now! I only need to figure out easiest cheapiest way to generate a variable frequency/amplitude 26v/400Hz signal.

This is the easiest way I could find:

Generate a 400Hz sin using a PWM from the microcontroller. Varying PWM parameters is very easy modulate amplitude/frequency.

Run that signal thru a ULN2003, that’s a very nice mid-power darlington array. That will amplify the PWM on voltage and current. Then, filter the output signal with a simple LC network, and run it to the motor winding. This is the idea:

So after building it (this is a 3-channel version):

I connect the motor, I get a nice signal on the scope…

But the motor doesn’t turn 🙁

So, what’s going on here… After some head scratching I see what’s going on. I’m not sending real AC to the motor, it’s more a modulated pulse, but actually I never invert polarity on the motor. I need something more complex, something I try to avoid because I’m going to need a good number of these circuits to control all motors on the instrument.

I happen to have some TDA2030A IC’s on my lab. That’s a very nice power amplifier, I can use it as a proof-of-concept. I build a small demo board:

and this is the signal I get now (PWM filtering is much worse here, but that’s not a problem)

That’s very good, and the motor turns now!!! I really don’t want to use this as the final circuit, but I’m in the right track.

I’ve just ordered a bunch of N and P Mosfet, so I’ll be building a simple circuit to generate this.. More once the parts arrive!

Share on Facebook


Boullu Guy · December 2, 2010 at 11:21 am

Hi Mr Sedano ,

just discovered your site and i think we have the
same goal….. make real non modified instruments
work on our simulator ; i am not an expet like you as i can see , but i am able to simulate synchros or resolvers , starting with USBServos cards from Opencockpit ; PWM signal modified to
0-5VDc ( 5VDC until 10VDC ), with a very simple board ; 1 USBServos Card = 2 simulated 3 phases synchros , then signal hatched to 400Hz ….and
with a small Sioc code make all my 26v/400hz
instruments work perfectly and accurately .Also able to make DC Motors driven with a synchro…
I have the sames ADI and HSI’s , every things
working fine .
i you are interested give me your email and i will send you pics of my sim … not a 737-200 for the time , but certainly i a short future.

i am located in France , near Aix enProvence.

regards ,

Boullu guy .

Boullu Guy · December 28, 2010 at 11:11 am

Hi ,

did you receive my doc ?

regards ,

Boullu guy .

    fransedano · January 4, 2011 at 2:27 am

    Yes many thanks!!! Private email on the way

Edoardo · February 1, 2011 at 4:33 am

Salut Boullu, je voudrait savoir si tu peut me donner plus info sur le travaille que tu as fait.
Tu peut voir mon website http://www.blogspot.com et ecrire moi là.

Merci beaucoup


Chris Taylor · August 11, 2011 at 4:39 pm

Edoardo and Boullu guy, I am working on the same issues. I have a real Mig-21 cockpit with lots of synchro/servo gauges I would like to get running. Could you let me know how to contact you both to discuss where I have progressed to and see how you have solved the problems. Inerested to see the details of Boullu’s solution and schematics of the circuits. Particularly looking at the sefvo amplifier of the synchro signal. Regards, Chris.

steve · November 26, 2016 at 8:06 pm

Did you ever succeed at controlling an attitude indicator?
I bought one on eBay and I want to control it with an arduino.
I have some ideas but have not tried them yet.
I found a manual that I think is very similar to the indicator I have.
I would be interested in sharing ideas.


Leave a Reply to Edoardo Cancel reply

Your email address will not be published. Required fields are marked *