how to build a frequency monitor?

[chancecasey]

Hi everyone!

I have very little RF experience (but know a little theory) and wanted to get some opinions on how I should proceed. I want to monitor the output of an AMB transponder (used in auto racing) - which transmits at about 3.58Mhz. I know nothing about the signal other than the frequency and that a "7 digit" ID is somehow encoded in it, along with some other data that I'm not really interested in. I just want to detect when a particular ID (transponder) transmits.

To do this, I figured I'd need an antenna, amplifier, and bandpass filter, then of course some way of logging "hi" and "low". I'm assuming the transponder uses amplitude modulation in a hi/low (binary) scheme to transmit it's ID. If I can gather enough data, I might be able to write software to correctly extract the ID out. Oh yeah, I guess I'd need to convert the signal to something usable by an RS232 interface.

Anybody done anything like this before?

thanks!
-Chance

P.S. - I'm not trying to do anything sneaky or rip off AMB's IP, as I'm not trying to "time" the cars, but rather have a simple system for detecting them, e.g. - knowing each one has left the track, or entered the pits, for example. The alternative is to buy one of their $10,000 detection devices that gives extremely precise timing and interfaces with their fancy software - which I do not need.

 

[RadioRon]

Chance, my advice is to get some help with this. What you need is a complete receiver and decoder for a signalling protocol that is unknown to you so far. This is not a project for a beginner. There are many questions that need answering before you actually design anything including:

- how much signal is available?
- what exactly is the modulation method used?
- how long is the signal available? (ie. is the car passing by at some speed or is it just sitting in front of you?)
- how far are you from the car (related to the first item above)
- is it ok that you are seen detecting this code, or is there any need for secrecy in operation?

If it were up to me, I would first bring some instruments to the location where the signal is. I would make a simple loop antenna out of scrap wire, connect it to a spectrum analyzer and watch the signal to try and determine the modulation type. I would also bring my communications monitor and view the signal in the time domain through AM, FM and SSB detectors to help identify the modulation type. Only after the modulation is understood would I try to put any electronics together to make a receiver. This is assuming you want to read such codes many times. If you only want to do it once, then just do it using existing test equipment, don't build a new receiver.

You need an RF engineer, technologist or technician if you are going to build a receiver that is to be used for anything other than a training tool.

 

[Sceadwian]

If you can't find the specs on how the signal is modulated it's not something a hobbyist can do. Though I'm sure the specs for the protocol have to be out there somewhere. 3.58mhz is not a very high frequency, and receiving and demodulating it shouldn't be particularly difficult. If you could find some refrences to the protocol and data formats used by the transmitters it might be possible to make a receiver. I would think someone out there has done something like this before.

 

[chancecasey]

For the first phase I would just do this on the bench, not the track, using a few different transponders with known IDs

The duration of the transmission is likely on the order of .00001 to .0001 of a second. The amount of data is small - if encoded binarily the information could be easily communicated in 25 or 30 bits. There is a "7-digit" ID, a 1 or 2 digit excitation field strength value, and an "unused" value, probably another 2 or 3 digit value.

I would imagine that with a single transponder it would be pretty easy to identify and analyze the transmission with the right equipment. So what equipment do you think I'd need? Would a scope do the trick? I guess it would need to "record" stuff? This is where my knowledge is non-existent.

Is the tricky part is just getting an encoded piece of data to work with in the first place because of all the modulation schemes? If I can just demodulate and get the encoded data, I think I have a decent shot at decoding it. I'd at least like to give it a try - and if the equipment isn't "too" expensive well hey now I can start doing all kinds of neat, easier stuff - like reading my car's knock sensors

Oh yeah and I'm pretty sure AMB has kept that under wraps - they have a monopoly in this business. And although it is not my intention to take any business away from them, I suppose someone could if they did this and weren't afraid of patent lawsuits. I'm in the software business and have some experience with patent law - not something you want to mess with.

 

[Sceadwian]

At only 3MHz it's likley AM. Build a simple AM receiver and tune it to that frequency, it's just a matter of using a different filter capacitor and inductor for the frequency required, all you need to demodulate is a diode. Schematics for simple AM receivers abound on the net.

 

[chancecasey]

I would rather buy some test equipment than build a receiver, if the equipment cost wasn't too high. There have been times in the past, and I'm sure in the future, where this kind of equipment would come in really handy. I'd like to hold off on trying to build a receiver until I can see if I can even decode the signal. If I can, then I'll look into building several receivers.

Given the data I've provided so far, what would I need? I assume a scope, but what capabilities would a scope need to help solve the problem above, assuming the signal is AM? Do I simply make a simple loop antenna and hook it up to th probes? Do scopes read mV at 3Mhz, and demodulate it (assuming AM again) and record it for later analysis?

If it weren't AM, I assume a spectrum analyzer would be needed - but these are quite expensive, yes?

 

[chancecasey]

Nevermind - gotta love Google - looks like the lines between scope and analyzer can get hazy, especially with PC/USB-based equipment. Wow, these things get pricey as they get faster! Just like race cars...

 

[RadioRon]

I would rather buy some test equipment than build a receiver, if the equipment cost wasn't too high. There have been times in the past, and I'm sure in the future, where this kind of equipment would come in really handy. I'd like to hold off on trying to build a receiver until I can see if I can even decode the signal. If I can, then I'll look into building several receivers.

Given the data I've provided so far, what would I need? I assume a scope, but what capabilities would a scope need to help solve the problem above, assuming the signal is AM? Do I simply make a simple loop antenna and hook it up to th probes? Do scopes read mV at 3Mhz, and demodulate it (assuming AM again) and record it for later analysis?

If it weren't AM, I assume a spectrum analyzer would be needed - but these are quite expensive, yes?

A scope is a great starting point. It may be all you need, since you have direct access to one of these transponders. This is because you can get right up beside it and pick up a signal with just a length of wire stuck into the scope vertical input. If you do it this way, you will see some sort of carrier signal presumably with modulation of some sort. If it is AM modulation, what you will see is a 3 MHz signal that turns on and off at rapid rate, or perhaps goes up and down in voltage at a rapid rate. You will have to fiddle with the sweep speed of the scope and try to capture one of the transmission bursts. For this reason, you really should be using a digital storage scope.

Here's a link of some pictures of AM on a scope:
https://www.electro-tech-online.com/custompdfs/2007/01/chap16.pdf

You can also post pictures of what you see and get our help.

 

[chancecasey]

Awesome - thanks for the info! Funny how it's so hard to get 1MB or more of memory in these things when a GB of memory for a PC is so cheap. Maybe it's SRAM? I guess a lot of memory isn't needed because you set up the "trigger" so that you don't "start recording" "by hand" ? I'm sure this will all become more obvious as I learn this stuff. I see a ton of used ones on eBay - a lot of analog Tektronic (sp?) and various others, and a lot of ones from China that are USB-based that seem to be like the ones from PICO - which by the way does anyone here have experience with, good or bad?