Super sensitive RF detection circuit

[Nora]

I built a **broken link removed**that is called an extra sensitive RF detection circuit. I couldn't get it to work although other people say they did.
I also don't fully understand how it works, and therefore can't troubleshoot it. I see 4 op-Amps coupled with capacitors. The author of the circuit says it is 4 (one quad package) high gain inverting op-amps LM324 with a 3-6" wire antenna as input and LED as output.
The configuration of the inverting circuit is also confusing to me, is it correct to put R1 to ground?

And how to analyze when Vout = -Vin(Rf/R1)?

Here's the link to the circuit again. Also- the author of this circuit has written a number of great circuits and electrical articles.
**broken link removed**

Also- I am trying to build a cell phone (900MHz) simple signal detector, please help with theory and circuitry.
Thanks in advance-
Nora

 

[biff44]

The only thing "super" about that circuit is its ability to fool inocent experimenters, like you. I will not work because the op amp has 0 dB of gain at 1 MHz, and drops like a rock after that.

 

[Nora]

Theoretically, is this the right way to approach a simple RF detection circuit?
Cut wire antenna to length of detection frequency, then use op-Amp gain circuit to amplify signal?

And why is the op-Amp in the previous circuit configured that way?
Thanks!
Nora

---------- Post added at 23:36 ---------- Previous post was at 23:33 ----------

Here is a cell phone signal detector circuit from the same author ... worth building or not?
**broken link removed**

Thanks!
Nora

 

[biff44]

In certain limited cases, yes. A diode detector (at the very end of the chain) is a fair way to detect RF energy. Having some amplifier gain in front of the detector will improve its sensitivity, but it will NOT increase the detection dynamic range!

So assume your 900 MHz signal is at a low but detectable level, like perhaps -70 dBm. And your diode detector is a sensitive RF schottky diode type, it would put out a fair amount of voltage at an RF input of -20 dBm. Then you could detect signals at that low input level IF you used a 50 db gain in front of the detector and a decent monopole antenna.

Now, 50 dB all buy itself on a pcb will simply oscillate. So you have to lay out the circuit so that that gain is split up. Perhaps you put 25 dB of gain and the diode detector inside of a connectorized metal box to shield it. Then externally you have the antenna, a ground plane for the antenna, and another 25 dB of gain. You can get such gain stages from Avago or RFMD.

Now you can determine the power level from -70 to -40 dBm at the atnenna output. If the signal is above -40 dBm, it will no longer tell you the level, but will correctly say it is above -40 dBm.

If you need to detect the actual power over a bigger range, such as -70 to -10 dBm...you will need something like an Log Amp (analog devices makes them)

One problem with all these circuits is that they will detect signals at all frequencies--only limited by where the atnenna works in frequency. So if you specifically only want to detect 900 MHz signals, but ignore the local FM radio station down the road....you will want to add a bandpass filter, probably at the output of the first gain stage. The make cheap saw filters that cover 902-928 MHz, for instance.

Rich

 

[Nora]

Thanks Rich.
Some more questions. Since US cell freqs are between 800 and 2700, I'd be detecting in that range. So OK, a really large band bandpass filter.
Too large of a band to be effective?

I looked up diode detector, what is this used for? - from Wikipedia: A diode detector is simply a diode between the input and output of a circuit, connected to a resistor and capacitor in parallel from the output of the circuit to the ground. If the resistor and capacitor are correctly chosen, the output of this circuit should approximate a voltage-shifted version of the original (baseband) signal. A simple filter can then be applied to filter out the DC component.

I wasted a nice protoboard soldering the first circuit. Is soldering necessary with RF circuits? It seems like the more bridges and gaps, the more the circuit might be effected. Or can I use a solderless breadboard?

The antenna...was planning to use a wire sticking up. Maybe once I understand the circuit better I would work on having a PCB antenna layout.

When do you use a monopole? When do you use a dipole? What is a wire sticking up called? How do you know when to use half wave or quarter wave?

Thanks!
Nora

 

[jiripolivka]

To detect and evaluate RF signals from ~10 to > 3 GHz, the best approach now is to find "logarithmic" detector ICs, available from Linear Devices, MAXIM and Analog Devices (and others). I used AD 8313 to cover the above range; with a 20 dB preamplifier, the output DC voltage with a microammeter can detect signals from -80 dBm.

A DC-biased diode detector with LM324 DC amplifier can work quite well, too; the lower level limit is ~ -30--40 dBm. Adding preamplifier ICs is easy.