Off topic Aircraft Instrument

I know this is off topic, but I would love to hook my Sanguino upto some electronics in my airplane.

The signal that I am interested in reading and or driving varies between +/- 150 millivolts.

Anyone know of a interface circuits to use?

Thanks

I think the Sanguino might be a bit big for that ?

What's sort of device is outputting/inputing those voltage levels ?

If you are inputting to tha sanguino, you could put three diodes in series, pointing towards the signal source, in series with a 10 k resistor to 5 volts on your sanguino. Tie the negatives together of the sanguino and your signal ground, and the signal input will be offset by the three diodes, to approximately 180 mV higher than it actually is, each diode will drop about .6 of a volt. That will give you a range of about 30mV to 330mv between the resistor and the first diode. The circuit relies on being able to sink some current into the signal source though, to develop the correct bias levels.
Outputing a voltage within your desired range is more complex though. If it's not high current, then perhaps a capacitor charge pump, PWM, diodes for level shifting again, and an op-amp running off your pheripheral to boost the output stability.

Freds, et al,

First off, a couple questions:

Is this a model airplane, or an airplane with people aboard?

Is this signal you're reading/writing at all critical to flight safety?

(I'm a great fan of experimentation, but not when it might cause an accident. So please use all due caution!)

IMHO, an instrumentation amplifier is a good way to change a small differential signal into a larger, single ended (non-differential) signal -- one more suitable to reading via an Arduino/Sanguino analog input. I've recently used the TI/Burr Brown INA2126 (that's a dual, the single is INA126.) These are handy because you can set the gain (for both inputs in a matched fashion) with a single resistor, and they can be made to work on 0/+5 V (and can be had in DIP packages, ~$4 for the dual from digikey, I think.) There are many similar chips from Analog devices, Linear Tech, and others. If your application has low requirements for precision/common mode rejection, you can wire up either the 2 op-amp or 3 op amp instrumentation configurations with whatever op-amps you like and 1% (or better) resistors.

To go the other way, you could first produce an analog voltage by low-pass filtering a PWM output, and then run that into differential by running it into a pair of amplifiers, with gains equal in magnitude and opposite signs. There are also single-chip "differential drivers" that can do the job. But I haven't used any, so I won't recommend specific chips.

HTH,

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

Thanks guys!

Yes this is for a human occupied airplane(mine), not going to take a nap while it is in the loop. It already has an autopilot, just not alitiude hold capablity.

I found a I2C presssure sensor for $50.00, so going to have it drive the forward and back trim for the elevator to hold +/- 100 ft with automatic disengage if it exceeds 200ft per minute rate change or the 100ft tolerence band.

Airplane has a manual switch to kill trim power so only switching 12 volts between the forward and reverse motions.

I have been flying for 25+ years and the airplane gets the best of care, not quite ready to spend 5K for altitude hold if I can do it for $100.00. Though it probally will be stopgap until I can afford the real fancy autopilot that I have been lusting after.

There was a guy that used to make a pic controller that listened to a hand held GPS and would feed the correction signal to the autopilot; since the hand held GPS is much more acurate then normal naviagation radios.

Actually in my reading the AVR Analog inputs can go up to 1/2 volt negative before they get clamped so the 150mv should be ok. I have an extra instrument head that I can experiment with on the bench, before anything in the airplane.

I assume this is an experimental where you have a little more flexibility in making changes
without having to be an A&P.
Will this have correction for pressure changes. Remember hi to low, look out below.
I would be concerned about the trim. I would hate think that the controller would jam the
trim to full up or full down before you got a chance to disable it.

I will prove it out in a friend’s experiential catagory airplane first, before the A&P route for mine.

Though I am thinking I can reprap a 3D solenoid mount to drive the pilots trim switch for no electrical connection to the airplane.

Airplane trims out really well and trim change is fairly much on the slow side as it is(quicker to manually trim), so hard to image a trim run away that wouldn’t cause an altitude excursion and not be cancel by the electric trim power disconnect switch which it right at hand.

Yep high to low, current IFR rated; went out and drilled some clouds today.

NASA precision mapped the Greenland ice caps with a GPS controlled tone generator that drove a low power transmitter on an ILS frequency, so been doodling on that also.

I'm no pilot, but if I were to put some homemade contraptions in charge on an airplane, I think I'd make damn sure that there are at least two independent (as in, separate and made of entirely different parts) systems for each task, and any significant difference in their decision would make a christmas tree's worth of warning indicators and some hefty horn go off in my face to let me know that there's something very wrong, and in need of my immediate attention, going on...

Edited 1 time(s). Last edit at 04/12/2009 10:09PM by Enleth.

You do that when you are flying anyway, your always checking and monitoring and looking for something to go wrong.

Actually in learning to fly it's the first and last 500 feet that are critical and reacting correctly if something goes wrong durring other phases of flight.

The rest of the time its almost like learning to drive a car in an empty parking lot.

I have no problem taking someone who has never flown a airplane up and having them manipulate the controls while I talk them though it. I did that yesterday with a lady friend as our first date.

Flying isn't that difficult though it has a lot of mystic around it, but if you can handle the complexities of RepRap then you could learn to fly an airplane.

Though it is a skill that you can always advance to new levels, I would love to fly an airplane where I was worried about speed limits to break; but that takes burning jet fuel.

And there are two independent systems, manual and power(electic) assist, just going to give the power assist a bit of intellegence that can be turned off in the same manner as if the trim actuation switch had stuck in position.

IF YOU KNOW WHAT YOU'RE DOING, and think it through, and are prepared to accept the consequences, then it should be OK.
You know a lot more about aircraft safety than I do. A reasonable risk assessment should be carried out - write down (makes you think about the answers):

what are the likely failure modes?

what is the worst failure mode?
(stuck trim? continuously changing trim? feedback voltages through the sensors blowing other instruments in the panel?)

what is the worst possible time for a failure?
(during landing / takeoff 10ft up?)

What are the consequences of the worst failure mode at the worst possible time?

Would I put my parents/children/Significant Other/siblings (pick one that is closest) in the aircraft and let someone fly it? (this one gave me a different mental perspective whan I was designing software for the Merlin Mk1 ASW chopper)

How can I minimise the risk?
e.g.
big 'emergency off' switch?
design it with a plug/socket so you can disconnect the entire apparatus?
procedures (like it will only be used above 5000ft? it will be completely disconnected or umplugged for takeoff and landing?)
(write these down now - when you're cold and sober, not excited with the 'just built this, now I want to test) - and stick to them religiously!
It looks like you've thought this through, but writing it down does concentrate the mind wonderfully.

I went through something like this before I built a RepRap.
Maybe we ought to have a safety page and risk assessment on the wiki?

Off topic, but have you looked at the accelerometers? There are a number of balancing robots (toy cars, etc) that balance on two wheels, I'm guessing that the logic must be somewhat similar?
Just thinking, please shoot me down... :-)

Hi David

Yes everything in your failure assessments has come to mind at some point. And I plan on very much taking it in stages. Altitude readout, Altitude deviation nag, fast acting vertical speed indicator, etc.

Accelerometers alone would not be much use for altitude hold unless you built up a full inertia measurement unit and then you would want to tie it back to something like the altitude sensor I am using to reduce its error budget.

There are projects out there where they are building IMU’s for remote controlled airplanes and helicopters. You can find most of the links about it on DIY DRONES.

Here’s a link for an IMU from Sparkfun that is Arduino based.

[diydrones.com]

So the 12th step past darwin is a bot that rolls around delivering parts that it builds on the way LOL!