Connecting safely to aircraft avionics circuits

While waiting for my Mega to arrive, I was hoping for some help on how to safely connect to two avionics circuits in an aircraft.

My project is to two fold:

  1. repeat a radar altimeter (R-Collins RADALT 55A) analog altitude readout on a 4 element 7 segment LED display driven by the Arduino.

  2. Repeat a 'decision height' light on an LED, again from the Arduino

For 1, I want to piggyback the arduino onto the radar altimeter output signal, which is a 0 - 16.4 VDC signal in two linear segments dependant on altitude. The voltage is normally fed to a analog height gauge. I want to tap this line using an analog input on the arduino.

For 2, the aircraft light is a 28VDC incandescent bulb. It is either off or on; I assume I will tap this to a digital in, but need to reduce voltage to 5 VDC and limit the current? Do I need to measure the current draw of the light first ?

My concerns are not damaging any aircraft equipment and making sure the oem fitted gauge and light continues to function along with the piggy backs. Eg I don't draw too much current off or cause any spikes ?

What resitstors/diodes etc should I connect between the signal(s) and the arduino ? I'm an Electronics noob so be kind :slight_smile:

The Mega is my prototyping tool (along with some fun and learning), but I eventually plan to make up to ten repeaters for the fleet I work with, so a smaller arduino may be better for the 'mass production' run.

I'll need to voltage reduce aircraft 24-28VDC to 9-12V power the arduino and the LEDS, but thats the next phase ... welcome any help regarding this bit too.

I could probably answer a few of your questions but first I'm a little concerned about any legal aspects of this. Are there any FAA type laws or regulations covering these kinds of modifications or attachments? Not preaching because I really have no knowledge of aircraft laws.

PS: I note now that you are posting from down under, but the source of the question still applies I guess.


Yes very good point. After I posted, I thought that the safety/legal thing might hold back some folks from answering, and I should have added the following, but I didn't want the initial post to be too daunting to read.

I can reassure folks there will be absolutely no comeback about legal issues or safety aspects, for any advice given here whatsoever.

I am working under the supervision of a liscensed aircraft mechanic (radio/avionics) ) who is rated on aircraft type. Any mods will be submitted to a CASA (FAA equivalent) certified avionics firm for issue of a legal engineering order before fitting. This is the normal legal way of doing aircraft mods.

So at the moment its really a 'feasability study', we have wanted to do this for a while and when I spotted the arduio thought this might be the cost effective solution.

Unfortunately the electrical guy at work is just too busy to help me with the nitty gritty, and also because he knows nothing of arduino's etc. An old fasioned 'non-digital/non-software' kind of background.

But he supports the idea, so he pointed me at the schematics & suggests I prototype and then submit 'up the chain'. We intend to test on boxes in the lab only anyway, so no flight safety implications.

So please, do not be worried about giving advice, I assure no risk to anyone.

I'm not sure I fully understand the signal for #1; "two linear segments"? Is that one signal, or two?

In any case, assuming there is enough current, a simple voltage divider may take your 0-16.4VDC down to 0-5VDC for the arduino's analog input.

That analog input is only 10 bits, so the entire range of the device will be divided into 1024 steps. If, for example, the instrument indicates 50k' at full scale, the arduino will only provide 49' increments. combine the resolution with the error in the ADC, and you may have accuracy to +/- 100' or so.

For #2, there was a recent thread on sensing 12V. As a digital input, it would be pretty easy (resistor + diode, or a voltage divider if the signal is 28V or 0V). If there's enough current to light an incandescent bulb that's daylight visible, there's plenty of current to drive a digital input.


For signal 1, there are 2 altitude regions, 0-500 foot gives 0.02mV per foot, then the signal switches to 0.003 mV per foot from 500-2500 feet. It doesn't read out above 2500 feet.

I hadn't thought about the resolution issue. I'd be looking for 10 foot resolution, but since we only need 2500 feet at 10 feet steps that would be enough with 1024 available I guess.

Thanks for the info on the divider. How much current would the Arduino take ? I'm wondering if it would compromise the voltage of the signal to the regular analogue gauge, ie lower it or something ?

I'm not sure yet if the voltage signal is regulated, thus not affected by a second load. I hope I am making sense.

I'll also see if I can find the thread on the 12V takeoff. Thanks for that.

0-500 foot gives 0.02mV per foot, then the signal switches to 0.003 mV per foot from 500-2500 feet

500 * 0.02mV is only 10mV.
Is that correct?
Did you mean 20mV (0.02 V) per foot?

Did you mean 20mV (0.02 V) per foot?

Ah Yes. Warning Will Robinson. My brain fade LED is on.

The Mega and breadboard turned up today ;D Off to order some resistor/diode/capacitor bits etc :sunglasses: