RF power measurment in a band

[resistor]

Hi all,

I'd appreciate any feedback on a project idea I've been noodling about.  I have no experience working with RF signals, so I'm worried that I'm way off the deep end.

I happen to own a sail boat, and like a lot of sail boat owners I'm always worried about conserving battery when anchoring out overnight.  A big trend in recent years has been replacing old 12v halogen lights with 12v LED modules, which saves a ton of power.  However, a concern has been raised over shoddy LED modules potentially radiating RF in the marine VHF bands (156-174MHz), which could be a safety concern as VHF is your primary SOS facility on a boat.

What I want to build a simple detector to determine whether an LED module I have plugged in radiating within the marine VHF band.  I need to build in the ability to set a baseline, since the amount of ambient energy in the band will depend on time of day, location, whether my neighbor is using his radio, etc.  The idea would be to set the baseline, flip the light switch, and indicate if power in the band is now above the baseline.

So my idea currently is to attach an appropriate antenna to a band-pass filter for the 156-174MHz range.  I think I then need to feed that through a logarithmetic detector (AD3818 modules seem easy to come by?).  What's I'm unsure of is what to do from there.  I've seen some projects labeled "Arduion RF power meters" that just sample the output of the AD3818 directly, but I don't see how that's actually measuring power.

I see that AD also makes true RMS RF power ICs - are those what I need?

[MasterT]

Get RTL2832 module from China, plug into laptop and scan a band. The chances to make good RF detector w/o experience close to null.

[resistor]

If I just wanted to solve it with money, I'm sure one of those $50 signal analyzers could probably tell me the answer.  I'm noodling about how to do it as a learning exercise...

[srb1954]

Why not just try using your receiver to see if there is excessive interference?

Tune your receiver to a quiet channel, flick the LED lights on and off and see if there is a noticeable increase in receiver noise level when the lights are on. Repeat the procedure for as many channels as possible.

[resistor]

Tune your receiver to a quiet channel, flick the LED lights on and off and see if there is a noticeable increase in receiver noise level when the lights are on. Repeat the procedure for as many channels as possible.

This definitely works, but is somewhat time consuming.  I'd like to be able to quick test different modules.  I'd also always be wondering if it might be fine when receiving a strong signal but still cause issues when trying to understand a far away signal, or when transmitting to a far away station.

[bob91343]

Noise measurement is a subject of itself.  There are expensive noise figure meters and so on.

You could modify a marine or amateur VHF transceiver and use its receiver.  Eliminate the narrow IF filter (about 15 kHz) and you have a much broader band.

You can also find a way to make the receiver sweep the band, rather than measure the entire band at once.  A spectrum analyzer can do all that if you are willing to set one up on board for the test.

[srb1954]

Tune your receiver to a quiet channel, flick the LED lights on and off and see if there is a noticeable increase in receiver noise level when the lights are on. Repeat the procedure for as many channels as possible.

This definitely works, but is somewhat time consuming.  I'd like to be able to quick test different modules.  I'd also always be wondering if it might be fine when receiving a strong signal but still cause issues when trying to understand a far away signal, or when transmitting to a far away station.

Unfortunately, everything to do with EMC testing is very time consuming and results can be quite variable depending on test set-ups particularly relating to the orientation and location of cabling. The advantage of testing the LED lighting in situ is that you get a more accurate indication of the real world results and eliminate variations relating to cable positioning etc. Testing on the bench will help weed out really bad LED lights but might not help all that much in determining the optimum choice for the final installation.

I don't think that you are going to have much problem with transmitting: the worst that is likely to happen is that your transmit signal quality will be degraded and maybe you will cause annoyance to other users by splattering spurious noise onto adjacent channels. There shouldn't be any decrease in transmitted power. Any noise would probably enter via the transmitter power supply and this can be fixed relatively easily with some filtering and shielding on the supply input lines.

Your main problem is going to be noise from the LED lights swamping out reception of very weak signals and this is not so easy to fix with filtering and shielding.

If your receiver has a signal strength meter you could get some indication of the magnitude of the  problem by comparing the meter readings between lights on and off on a quiet channel. You can then compare those meter readings against the meter readings of weakest signals you can clearly receive to get an indication of how much the reception will be degraded.

[David Hess]

Common untuned receiver designs will not have enough sensitivity because it takes a very low level to interfere with a receiver.

Any VHF SSB or AM receiver could be used to measure the noise levels.  Using an RTL2832 as a spectrum analyzer as suggested by MasterT could be a good option.  A higher performance but still low cost alternative is the SDRplay spectrum analyzer with an RSP1A receiver:


https://www.sdrplay.com/spectrum-analyser/
https://www.sdrplay.com/rsp1a/