https://github.com/RamboRogers/rfhunter
This project is an RF Signal Scanner built using an ESP32, AD8317 RF detector, and various other components. It's designed to detect and measure RF signals in the environment and display the signal strength on an OLED display. It's useful to find hidden cameras, wiretapping devices, and other RF-enabled devices.
(Score: 5, Interesting) by VLM on Tuesday October 29, @12:05PM
The chip is spec'd down to 1 MHz but works lower (its an old chip from just after Y2K and I played around with it in the past) so it kinda-sorta works for EMI/EMC screening but not so well. You can pick up a lot of switching power supply noise below 1 MHz but its not spec'd for that although it will respond somewhat.
The chip goes to some effort to generate a pretty good dB ratio measurement which might or might not be overkill for some applications. It outputs an analog voltage not I2C so its not overly precise. I vaguely remember "dozens of mV per dB" slope in output voltage. So if you stuck a -10 dB standard on the input you'd get some random voltage around "whatever" and then if you stick a 3dB attenuator inline the voltage will drop (I think?) by precisely 25*3.000 mV, the delta will be pretty precise, but its anyone's guess what the starting and ending voltage will be. If you slapped a relay on the input with good isolation (good luck finding a cheap one) your software could autocorrect itself, including if you had amps/filters/attenuators on the input. So that's cool.
It's "only" linear over a range of maybe 40 to 50 dB from memory.
General aliexpress cost for the RF board with SMA connector ready to wire to a microcontroller is under "six dollars" and its a 5V tolerant chip so it'll work with pretty much anything (not ESP32 specific)
Its an old chip from the 00s, you could buy a Ramsey Electronics kit that used it in the 00s, they work pretty well. Fundamentally this chip/dev board is a magic black box that turns around 0 to -50 dB from 1 MHz to 10 GHz into an analog voltage that has a constant-ish linear change with dB change in input, and the rest of the project / kit is merely a voltmeter with a fancy display and some power supply, etc, more or less. If you just want to F around with the chip on the lab bench literally all you need is 5V and a voltmeter. IIRC the output impedance was rather large so forget driving a LED directly its meant to feed opamps or comparators or a microcontroller A/D input.
If you're doing EMI/EMC and just looking for signal, or doing generic "peak for max output" tuning, then just use a diode with a voltmeter clipped across it. Or if you're doing "serious RF work" like getting a multiplier peaked up on the correct peak, use a spectrum analyzer. But for some simple uses, this can be a good chip.
I believe it was originally developed / marketed for IF "RSSI" meter purposes (although not as low as 455 KHz), so if you're doing something very vaguely like that, the chip will work very well indeed.
(Score: 5, Funny) by RamiK on Tuesday October 29, @12:30PM (2 children)
Back-when I glued together an RF meter module (might have used an ad8317? no idea), a 12v battery pack for the VCC and a cheap 2v dial voltmeter... I believe they were something like these: https://www.aliexpress.com/item/4000144175681.html [aliexpress.com] https://www.aliexpress.com/item/1005005898356202.html [aliexpress.com]
Went down to the celler with an old walkie talkie and another "real" $400 RF meter I borrowed from the lab to "calibrate" by finding the right multiplier for the curve... Used a pocket calc to convert... And the fixed voltmeter dial was an upgrade from measuring the OUT with a multimeter...
They gave me my own "real" RF meter a week later. Well, after I forgotten my DIY at the bus stop... And the police detonated it... And the manhunt... lol... not joking...
compiling...
(Score: 4, Interesting) by VLM on Tuesday October 29, @12:57PM (1 child)
Narrower band, higher price, unnamed chip, no characteristics available. Probably better off with an AD8317.
My guess on the higher cost / narrower input bandwidth is it has an input amp so it works to lower signal levels. IIRC the AD8317 bottoms out around -40 to -50 dBmW input. Without any identification or specs, all I can do is guess. Was I correct? If it detected your walkie talkie from a modest distance it probably has some kind of input amplification.
Gotta be careful on AliExpress because they'll sell what the market will bear and the market doesn't think very carefully. The new hotness would be something like a ADL5513 for about $10 from the 2010s and IIRC that has a usable dynamic range near 70 dB at best (although marketing markets it as 80 dB range LOL because it technically "works" at extreme ranges, regardless its near twice the dynamic range of ye olde AD8317). AliExpress is the same way for microcontrollers, they'll gleefully sell you some obsolete small memory capacity model of a ESP8266 or ESP32 model from years and years ago at about the same price, or even more, than newer better chips.
Ah the societal price of fearmongering-as-a-business-model it makes a lot of money for the right people, and the authoritarians absolutely love it, but sometimes hobby electronics get blown up.
(Score: 2) by RamiK on Tuesday October 29, @01:59PM
For the life of me I can't remember. It was quite a few years ago and I got the module at a local RadioShack-type place when I was looking up bnc probes or something and noticed there was a cheap board and voltmeter so I could just DIY it instead...
It's not universal but old micros with lower memory capacity usually have better voltage and current tolerances. If memory serves this is actually a considering with Espressif parts for some popular designs... But honestly I didn't really get into them until the riscv ones came along so I wouldn't know.
compiling...
(Score: 2) by SomeGuy on Tuesday October 29, @11:28PM (2 children)
I'd really like to read more on the practical uses of this and how it responds to other RF-emitting devices.
On youtube, there is a video of a low frequency RF device (requires a dumb smart phone, of course) that is supposed to find pipes in walls, but can be used, in a limited way, to sort of see other things through walls. Enough to see vague shapes and movement. Would this detect that sort of thing?
On the main page, it talks about finding hidden cameras, but as far as I know that only works if they are dumb enough to put a wireless transmitter right on the device. What about a USB/ethernet camera or wired microphone? Won't do squat will it? Nobody ever talks about that.
Does this thing detect RF put off by LED lightbulbs and other noisy modern electronics? Although, would a TRS-80 through the entire thing off?
(Score: 2) by namefags_are_jerks on Wednesday October 30, @02:26AM
RF Signal detectors are over a century old. That ESP32 would have like 400 million transistors (as a guess); you only need one.
The old-tech Detector diode way would also pick up other RFI, which would be answer your question -- the ~150 kHz RFI from DC power convertors, the 10(*N) MHz harmonics from wired ethernet, and the other birdies from electronic devices. Like you said, a smart creep would cloak the recorder's RFI with another nearby RFI source.
The old 'FM Radio Bugs' kits that're entirely analogue and whose emissions are within the FM band would be quieter.. (although there are of course methods for detecting those)
(Score: 2) by owl on Wednesday October 30, @10:11AM
Wired USB/ethernet will give off some RF -- whether this device (or whatever other device one may have) is both sensitive enough at the correct frequencies, and can pickup the USB/ethernet RF out of the noise floor, is a different story.
A wired mic will also give off RF, but RF at audio frequencies will be very tricky to "detect".
And, this of course assumes that the "hidden camera installer" is smart enough to not just do the easy stupid method of "wifi hidden camera" as you say. For most "peeping tom" types, they likely have no idea how easy it is to detect their "wifi hidden camera". It works simply because only 1 in 1000 know to bother to look.