Build your own VOT for $5 - no, no joke

[photofly]

Because we value something only in proportion to the cost and difficulty of obtaining it. If a test is done with a $5 part and something you already own and carry in your pocket every day, it's impossible for it to give results as valuable and good as a device that costs $1000 from Aircraft Spruce.

[NeverBlue]

makes something very simple sound very complicated

Actually a VOR signal is kind of complex...the most of any aircraft navigation signal that I've seen.

[Rookie50]

You know, speaking as one who is horrible incompetent with electronics -- must have failed shop class -- I'm impressed with PF's initiatives, even though I doubt I could replicate them. I'm a bit scary with tools......

Two thoughts that came to me: First, we can all choose to live our lives as creators or critics,

And -- if one chooses to work and create (in aviation or elsewhere) solely for money as a motivation -- that will be the only reward given.

Follow your passion, and money will follow as a byproduct, along with many more rewards.

Kudos to those who create and innovate!

[CpnCrunch]

According to the webpage, the range is 100m with a simple wire antenna. It's also not a perfect square wave:

http://raspi.tv/2013/rpi-gpio-0-5-2a-no ... -to-use-it

so it won't just generate nice harmonics at multiples of 108Mhz. The only way to know for sure is to check with an RF spectrum analyzer.

[photofly]

makes something very simple sound very complicated

Actually a VOR signal is kind of complex...the most of any aircraft navigation signal that I've seen.

Respectfully, I disagree. The engineering required to broadcast the signal is complex, but the signal itself is very straightforward: sufficiently simple to be decoded in a 1950s era radio set without any digital logic.

According to the webpage, the range is 100m with a simple wire antenna. It's also not a perfect square wave:

http://raspi.tv/2013/rpi-gpio-0-5-2a-no ... -to-use-it

so it won't just generate nice harmonics at multiples of 108Mhz. The only way to know for sure is to check with an RF spectrum analyzer.

It will just generate harmonics at multiples of 108MHz. It may not be a square wave exactly, but it's still periodic with a very stable frequency, so it can only contain frequencies that are integer multiples of the fundamental. If it's a symmetric waveform then it can only contain odd harmonics: 1st (fundamental), 3rd, 5th etc.

As far as energy content in the unwanted harmonics goes, square waves are pretty bad. Something trapezoidal will be better.

[Posthumane]

For Posthumane, here's the code that churns out the baseband signal.

It's based on the code from this webpage: http://soledadpenades.com/2009/10/29/fa ... ve-module/

The wikipedia page on VOR was mostly unhelpful, because it makes something very simple sound very complicated. This reference was much more use: http://www.nist.gov/calibrations/upload/tn1069.pdf

Let me know what you try and how it goes.

produce_VOTwav.py 2.zip

Thanks for the links. I probably won't be able to get to it for a while, but I'll give it a try sometime. SDR is a fun thing to play around with (at least if you're a nerd like me). I've been thinking recently of trying a few cheap designs for an automatic direction finder in the VHF bands. I agree with you about the wikipedia page on VOR not being very useful.

It will just generate harmonics at multiples of 108MHz. It may not be a square wave exactly, but it's still periodic with a very stable frequency, so it can only contain frequencies that are integer multiples of the fundamental. If it's a symmetric waveform then it can only contain odd harmonics: 1st (fundamental), 3rd, 5th etc.

As far as energy content in the unwanted harmonics goes, square waves are pretty bad. Something trapezoidal will be better.

CpnCrunch is not wrong here, you may in fact get some small harmonics in other frequencies, due to various non-linearities in the components that make up the Pi. A non-linear junction will often radiate a 2nd harmonic stronger than 3rd, and you can have various intermodulation products, especially when you have various active components with their own clocks on the same line as your antenna. That being said, these various harmonics will generally be orders of magnitude smaller than the base signal, and not something you have to worry much about unless you have a very sensitive receiver.

A little tidbit you might find interesting - I've worked a bit with non-linear junction detection and when you illuminate a complex electronics board with two frequencies, f1 and f2, you can get back some funny intermod products such as 2f1 +/- f2. 2f1 +/- 3f2, 3f1 +/- 2f2... The interesting bit is that different electronics will emit different intermod products which you can sometimes use to identify a particular item based on its harmonic signature.

[photofly]

A little tidbit you might find interesting - I've worked a bit with non-linear junction detection and when you illuminate a complex electronics board with two frequencies, f1 and f2, you can get back some funny intermod products such as 2f1 +/- f2. 2f1 +/- 3f2, 3f1 +/- 2f2... The interesting bit is that different electronics will emit different intermod products which you can sometimes use to identify a particular item based on its harmonic signature.

CpnCrunch's assertion - that because the output isn't a square wave, it must contain frequencies other than the integer harmonics - is plainly wrong on the mathematics, and we don't need to address it further.

Your assertion - that intermodulation distortion generates new sum-and-difference frequencies is trivially obvious: that's how we're achieving the AM modulation of the audio signal. To use your nomenclature, f1 is 108MHz, and f2 is a 10kHz bandwidth audio signal.

However if you wish to assert that IMD is throwing energy into VHF frequencies other than the integer harmonics of 108MHz, you're going to have to show there's another f2 (in the VHF band) present - not already a harmonic of 108MHz - to mix with. You can take as many sum and differences of the harmonics already present in the 108MHz "square" wave, and and they can only generate other already-present frequencies. Nothing new.

In fact there is no evidence of any other f2 present on the output. If there were it would show up as ripple or instability in the PWM output. It would actually cripple the ability of the PWM output to do the job it was designed for.

[CID]

One of the issues with signals like this is that the "noise" is not limited to multiples of the fundamental frequency. As mentioned, the "square wave" won't be perfect and the modulation method is crude. The truth of the matter is there is no way to be sure what this gadget is spewing until you subject it to a test with a spectrum analyzer. There are fairly inexpensive alternatives to this that would involve a little solder and a simple oscillator so I don't know why so much effort is being spent on making this square peg fit in the round hole.....

[Posthumane]

.


However if you wish to assert that IMD is throwing energy into VHF frequencies other than the integer harmonics of 108MHz, you're going to have to show there's another f2 (in the VHF band) present - not already a harmonic of 108MHz - to mix with. You can take as many sum and differences of the harmonics already present in the 108MHz "square" wave, and and they can only generate other already-present frequencies. Nothing new.

In fact there is no evidence of any other f2 present on the output. If there were it would show up as ripple or instability in the PWM output. It would actually cripple the ability of the PWM output to do the job it was designed for.

No, I wasn't saying that there would be intermod interference in the VHF band, just that almost every active board such as a Pi would have some additional oscillations present, and those do give rise to other intermod products. There's a 1 GHz clock somewhere on the Pi zero (I haven't looked if it's internal to the processor or if it has an external xtal) and with a sensitive receiver you'd be able to see that, plus sums and differences of the harmonics of your transmission. That is all pretty academic, of course, since those would all be so small as to be trivial and they wouldn't be in a band you care about. So not at all a criticism of the design, just an interesting factoid, and only relevant to those working in fields that make use of those spurious emissions.

When I get around to trying this out, I'll see if I can throw a wideband spec-A next to it and either feed it directly or with a wideband antenna to see what's coming out of it. I suspect that things outside of the intended band will be hardly even visible above the noise floor of the DC-8GHz spec-a.

[photofly]

There are fairly inexpensive alternatives to this that would involve a little solder and a simple oscillator so I don't know why so much effort is being spent on making this square peg fit in the round hole.....

Design and build one, then. Share your expertise. If you can do better, then... do better. (Actually I believe you'll find building a stable 108.00MHz oscillator with just a few components isn't straightforward. But you're the analogue RF design engineering expert in this thread, apparently.)

Meanwhile I'm still waiting.

[NeverBlue]

Respectfully, I disagree. The engineering required to broadcast the signal is complex, but the signal itself is very straightforward: sufficiently simple to be decoded in a 1950s era radio set without any digital logic.

Respectfully???....well

You're confusing date of technology with complexity of...

the VOR signal is complex to transmit because IT IS complex.

Digital signals today are as about as simple as they could be to encode/decode.

The VOR signal is both an AM and FM signal in one carrier.... 30 Hz Variable, 9960 Hz FM'ing +/- 480 Hz at a 30 Hz rate and 1020 Hz all AM'ing a VHF RF signal.

[photofly]

Respectfully, I disagree. The engineering required to broadcast the signal is complex, but the signal itself is very straightforward: sufficiently simple to be decoded in a 1950s era radio set without any digital logic.

Respectfully???....well

You're confusing date of technology with complexity of...

the VOR signal is complex to transmit because IT IS complex.

...

The VOR signal is both an AM and FM signal in one carrier.... 30 Hz Variable, 9960 Hz FM'ing +/- 480 Hz at a 30 Hz rate and 1020 Hz all AM'ing a VHF RF signal.

Whether you think it's complicated or not is a judgement call. One can describe it completely in a single line (you pretty much did that yourself) and produce it with three looped lines of computer code, using just some trig functions - two lines, if you don't want an ident. That's where this project started. The raspberry pi was just the quickest way I could find to modulate the signal onto a carrier. I have a synthesized signal generator that will generate the RF and modulate it, but it weighs 80lbs so it's not particularly easy to take it out on the ramp.

Most of the engineering complexity in the VOR transmitter is having to build the counterpoise, to alleviate terrain effects, and the complicated multi-antenna array with phasing arrangements to use the Doppler effect to avoid having an actual rotating antenna like the original stations did.

Digital signals today are as about as simple as they could be to encode/decode.

Now if you want a complex navigation signal, try synthesizing your own GPS data. Then encode it with BPSK using the correct CDMA codes. Bet you can't describe that in one line.

[SuperchargedRS]

Don't know why everyone is s***ing on the OP for this - the knowledge and that little bit of motivated ingenuity is pretty f***in' impressive.

My head is chock-a-block full of flippin' enough BS airplane stuff - the fact the OP has enough left-over cells to learn a bit of Python is pretty neat.

'bout time we had that modern taste of motivation in this profession anyways

Besides, what the heck do ya think your IFR4000 is doing anything different than this little unit anyways!

Very true, to the OP good stuff, the non knuckle dragging folks on this board love seeing someone posting cool tutorials like this

[NeverBlue]

Now if you want a complex navigation signal, try synthesizing your own GPS data. Then encode it with BPSK using the correct CDMA codes. Bet you can't describe that in one line

Uh...you just did...phase shift keying is not complex...it's FM...sort of...easy-peezy stuff to decode/encode...it's just a word...it's just software...that's why you can buy a GPS for your car for less than $100.
It comes in every smart phone now...complex????

VOR is not easy...your set-up is not common.

[photofly]

I'm sure the "aircraft are not science experiments" brigade will look up over their reading glasses at me in a disapproving manner because of the sheer pointlessness of it all, but I think this is cool:

http://www.rtl-sdr.com/decoding-aviatio ... h-rtl-sdr/

[CID]

I'm sure the "aircraft are not science experiments" brigade will look up over their reading glasses at me in a disapproving manner because of the sheer pointlessness of it all, but I think this is cool:

http://www.rtl-sdr.com/decoding-aviatio ... h-rtl-sdr/

Cute. Pointless. Harmless.

[CpnCrunch]

It comes in every smart phone now...complex????

If you think a smartphone isn't complex, perhaps you should try designing and building your own from scratch.

[NeverBlue]

I don't know what you're reading...

but I NEVER SAID THAT AT ALL!

what is the matter with you?

[CpnCrunch]

I don't know what you're reading...

but I NEVER SAID THAT AT ALL!

what is the matter with you?

You said a GPS isn't complex because it comes in every smartphone. So you're saying GPS and/or smartphones aren't complex. You can buy a GPS chip for under $1, but that doesn't mean it isn't complex. Tell you what, try building your own GPS receiver from scratch and see how complex it is.

[photofly]

It's a fair amount of work:
http://www.aholme.co.uk/GPS/Main.htm