Cheapo AI

[Jan Olieslagers]

Given the (more or less) recent appearance of low-cost R/C helicopters and quadcopters, I assume there must be cheap pseudo-gyroscope electronic sensors available. Wouldn't it be feasible to apply these in a low-cost AI? Such a device could obviously only be used in non-certified planes - but such is mine.

NB perhaps this is not the right place to discuss this kind of idea - I'll be glad to learn of better. I used to be on nttp://rec.aviation.homebuilt but have been shut off from usenet by my internet provider, and never managed to find an equivalent forum.

[Whiskey Kilo Wanderer]

Hi Jan,

I think TruTrack use something similar in their turn co-ordinator and ADI. I've used the turn co-ordinator for a few years now and it works well, being more reliable and taking less power than the previous electric gyro version.

The ADI is a full attitude indicator and heading (actually track) sensor. Not exactly cheap, but might solve a collection of problems if panel space is limited.

TruTrack ADI (http://www.trutrakflightsystems.com/products/ADI.html)


Safe Flying,
Richard W.

[cumulusrider]

There is me thinking that AI meant Artificial Intelligence.
I dont know what made me think intelligence has anything to do with some posters on this forum:)
Cumulusrider

[IO540]

This has been done but there are technical issues concerning the way in which a slow background self erection process is implemented.

In a vac horizon the self erection is done using pendulous vanes (IIRC); in an electric horizon it is done (IIRC) with little switches with weights attached.

In a purely solid state gyro horizon you don't get any gravity sensing so you have to make a stab at when the plane is straight and level, by using one or a combination of airdata or GPS track/altitude data. IIRC Aspen use the former whereas Garmin use the latter. None of this is hard (especially the GPS version) but it means it isn't completely cheap and trivial. This (http://i101.photobucket.com/albums/m74/peterh337/Ipad%20stuff/ipad-ai.jpg) for example is virtually useless.

[Mark 1]

About the cheapest that I've seen that I would trust in IMC is the MGL (http://www.mglavionics.com/html/infinity_singles.html).

That's about $800 for the sensor and $200 for the cockpit display. Add another $260 and get a remote compass too. For not too much more, the Dynon D6 has a full 6-pack of instruments and fits in a 80mm panel cut-out. I have actually flown that in IMC and it was rock solid. It can also have an internal battery that gives you 2 hours use if you have electrical failure.

The low cost solid state gyros are actually rate gyros, so need to be integrated to provide attitude data and this introduces drift errors that need fancy algorithms and additional input to correct. All this adds up to the cost of the budget instruments that I've mentioned.

But this technology is still developing thanks to the US experimental market that is allowed to use it in real IFR. So things will probably get better and cheaper.

[A and C]

While I see good quality AI's getting cheaper with the advance in technology I dont ever see something reliable enough to trust my life to as becoming cheap.

[Rod1]

MGL do a low cost Dynon D6 competitor as well parts for aircraft, aircraft spares uk, pfa aircraft, microlights, homebuilt aircraft (http://www.partsforaircraft.co.uk/). I have has an MGL setup for some time and consider it 1st class. I suspect much more reliable than a vac pump.

Rod1

[Jan Olieslagers]

@ a&c: flying a microlight, I am supposed to never come anywhere near a cloud anyway. If I consider an AI, it is just against the odd chance of finding myself where I never ought to be.

@cumulusrider: though it is against my nature and upbringing, I am making a fair effort of using as many codes, abbreviations and acronyms as I can manage, because that is obviously the correct way of posting here.

@all: thanks for the pointers, but I have failed to make clear that (having some skills at soldering and PCB design) I was thinking of a homebrew concoction around the R/C heli sensors. Apologies for not being sufficiently clear.

[Jan Olieslagers]

It is that too, I am considering redoing the dashboard and real estate is precious indeed. Also, I think the AI a "nice to have just in case" but as said I am not really supposed to ever need it, so I am not going to spend a 1000 euro's on it - even at today's prices that amount will buy me gas for many hours of flying.
But above all, it seemed a nice occasion to heat up my soldering iron and do some jolly tinkering.

[abgd]

It's often seemed to me that there is something of a gap for an 'open avionics' suite to be developed for the homebuilt market. Though I'll admit I haven't looked to see what's available already. I do a fair amount of home electronics - enough to realise I probably wouldn't trust my life to anything I designed - but when I look at the prices of relatively simple things like tachometers and engine temperature gauges I can't help thinking that a little PIC and graphical display could do the same job for a fraction of the cost.

[abgd]

Here's a website regardin a homebrew flybarless system:

About the project - 3Digi (http://3digi.wikidot.com/einfuehrung-ueber-das-projekt-en)

[abgd]

Actually, couldn't you combine a magnetometer with a gyro to partially solve the erection problem? How about a little lead tray with cosmic-ray detectors top and bottom?

[Mark 1]

In the real lightweight and experimental arena; I met the guys from Belite (http://beliteaircraftstore.com/avionics.html) at Sun'n'fun earlier this year. They even got their exhibition stuff under hard cover before the tornado arrived!

They seemed very impressive and knowledgeable guys and were showing off some very lightweight and capable kit.

Maybe worth a look.