A Device which Measures RF Frequency

[cuauhtle]

Hi,

I want to build a RF circuit, which can measure the frequency of the radio wave. The range is quite wide 100-1000MHz. The accuracy is not that important for the time being.
Also I probably will use multiple antennas whose BW is wide for different regions (like 315MHz, 433MHz).

Having few experience in RF world, I have been reading some documentation for some weeks, I quite understood the antenna phase (I mean the structure of the antenna and building the resonant/matching circuit), but following that, it seems it is a must to use a receiver IC to amplify/filter the waveform (Unless I design the LNA, PA, mixer or even PLL? etc.).

From the data-sheets of the receivers that I have read so far -like rfRXD0420, ATA5743 etc- the conversion to intermediate frequencies (on the order of tens of MHz), amplification, filtering and demodulation is done in-circuit and mostly only an IF signal which is modulated by an external crystal and a low frequency data is given as an output to user-which will not allow me to observe the UHF signal.

Here best option seemed me to design the blocks, which are contained in the IC, myself but in beforehand I wanted to ask you if I can manage to get the signal by using any receiver IC's output pin. Once I can manage that, I plan to use a frequency divider circuit and use its output to count the frequency which I had already managed to build using a basic micro controller.

I hope I could be clear about the problem,

-Of course I do not ask for a solution of my project but tips, advices and resources.

Thank you.


cuauhtle

Microchip's rfRXD0420 www.datasheets.net/pdf/microchip.com/70090a.pdf
Atmel's ATA5743:www.atmel.com/dyn/resources/prod_documents/doc4839.pdf

 

[biff44]

No you are not clear at all. Those two chips are very narrowband, and can not possibly tune over the 100 mhz to 1 Ghz range.

WHAT are you trying to measure, and to what accuracy? Be very specific.

 

[cuauhtle]

No you are not clear at all. Those two chips are very narrowband, and can not possibly tune over the 100 mhz to 1 Ghz range.

WHAT are you trying to measure, and to what accuracy? Be very specific.

I want to measure frequencies of remote car keys between 300-500 MHz (I think 315 and 433 MHz are standard)

I thought I could use a chip for each antenna.
The accuracy is not strictly specified but I intend to obtain a resolution around 100KHz.

Added after 22 minutes:

Also, for wider bandwidths, what kind of ICs should I focus on?

Thanks.

 

[betwixt]

These kinds of devices send out short bursts of data which makes measuring the frequency quite complicated. The classic method is to open a gate for a specific period and count how many cycles pass though it. The trouble with data transmissions is that a lot of cycles may be missing and you get a wrong count.

It may be better, especially as you do not need absolute accuracy, to use a spectrum analyzer. The can be VERY expensive but you can make a simple one from a TV tuner module. They are usually PLL tuned so with the addition of a small microprocessor circuit you can make them sweep across all or part of the frequency range while you monitor the signal coming out of them. As they tune through the transmission you get a rise in output level at that frequency.

Brian.

 

[cuauhtle]

Yes, the data length is around 200 bits (strictly depends on the remote).
If we consider that the rate is 2-4kbps it lasts nearly for 0.1s, is it a short duration for capturing the preferred RF signal?

TV tuner modules have wide range receiver right? Using a wide-range tuner, switching a gate sounds nice, but still I need to amplify the signal?

Sorry for my poor knowledge, thanks.

cuauhtle

 

[Kripton2035]

this cn be what you are searching for ...

**broken link removed**

 

[biff44]

Assuming you have a big signal level (ie you have the key fob in your hand and are trying to measure it), you need to be able to quickly scan the who frequency range and figure out the frequency transmitted while the data is coming out.

There are probably a number of ways to do this. TA couple of ways that might work are:

1) down convert the frequency to a slightly lower range, say 50 to 350 MHz, adc sample it with a high clock rate, and do an FFT. This is going to be expensive due to the the high clock rate ADC, which will need to be at least 750 MHz.

2) build a delay line frequency discriminator, and use the analog voltage out of it to figure out the frequency. The problem is to get the 100 KHz accuracy. Might need to use two different delay line lengths to get the accuracy.

3) have a narrowband scanning system (maybe a 20 Mhz bandwidth) and push the key fob many times while the system scans the entire frequency range.

4) use some sort of scanning LO oscillator, downconverter, and a narrowband IF filter and a envelope detector or log amp. When you detect a hit, you look back at what frequency the LO was scanning thru at that moment to figure out the frequency.

I encourage you to look at ALL of the block diagrams possible for doing such a system, and figure out which one is the best for your (still mostly undefined) system requirements.

 

[FvM]

To add another option, I would attach a wire loop to my high speed oscilloscope, place the transmitter on it and determine
the frequency from the acquired waveform.

 

[biff44]

He wants a "circuit" to do it. If you have access to lab equipment, you would just hook up a spectrum analyzer and use its peak hold to figure out the frequency.

 

[FvM]

He wants a "circuit" to do it.

I see.

In the transmitter proximity, where you have sufficient field strength and don't necessarily require selectivity, a kind of
triggered frequency counter is one of the possible low complexity solutions, I think.