Using a spectrum analyser with commercial probes to sure up my designs

[swmcl]

Hi all,

First post I think ...

I'm thinking of using a spectrum analyser to check out anything I design that might get to the day where I have to pay for compliance testing.  Is this possible ?  For real ?

I'm not sure how to calibrate things nor what to buy but whatever the case, I do not want another plaything on my bench.  I would buy if I could have a reasonable surety regarding the intended purpose.  My pockets are not that deep but I would be interested in commercial probes and pre-amps.

I'm thinking Rigol RSA-3015E-TG at present.  I have an Ethernet to Serial board that will be designed commercially and my other projects are more PIC based.

Honest opinions please !

Cheers,
Steve

[tautech]

Consider the SSA3015X Plus or 3021X Plus with the EMI promo at a much reduced price.
https://int.siglent.com/info/detail-40.html

SSA3021X Plus are proving more popular here as they can be modified to the SVA1032X whereas the 1.5 GHz model can not.

[Kean]

Hi Steve,

Welcome to the forum!

Both the Rigol and Siglent spectrum analysers mentioned are pretty neat for tracking down the source and verifying mitigaion of compliance issues ... once you know what you are looking for.
They can be tricky for doing "blind" pre-compliance measurements without a lot of experience, largely due to the nature of calibrating & correcting any RF measurements against the limits.
Measuring conducted emissions is a lot easier than trying to measure radiated emissions, but both will require a bunch more equipment than just the SA.

I'd suggest reading the following posts to start, and then maybe come back with any questions.

https://www.eevblog.com/forum/testgear/pre-compliance-setup/
https://www.eevblog.com/forum/testgear/emc-pre-compliance-antennas-in-australia/
https://www.eevblog.com/forum/testgear/rigol-dsa815-tg-emc-pre-compliance-questions/

Also check out the recently started "EMC for Everyone" video series by Kyle Hunter (author of the first topic linked above) at https://www.youtube.com/c/MicroTypeEngineering/

Kean

[EE-digger]

Fortunately, some of the most useful probes are ones you can replicate at your bench.

Near field probes for magnetic (H) field are easy to make from coax or better, rigid coax.  For electric field, E field probes are also easy to make, generally a short exposed center conductor coax.

Be sure that what you make is well insulated so you never need to worry about your analyzer's front end or that of a preamp.

At first I thought you wanted scope probing on a spectrum analyzer.  Apparently not but if the need arises, check out the old Tektronix 1103 power supply and Tek's active probes.  They are fairly priced, unlike Agilent/Keysight probes.  The 1103 can be used forever, having a BNC output to anything you want.

[nctnico]

Fortunately, some of the most useful probes are ones you can replicate at your bench.

True. A LISN and a current clamp (https://www.edn.com/using-current-probes-to-estimate-e-fields/) are easy to construct yourself. For near field probing you can use a piece of coax in a loop but in my experience these are easy to break. I bought a set of near field probes from RF explorer (https://www.seeedstudio.com/RF-Explorer-Near-Field-Antenna-Kit-p-2784.html ) which are good value for money.

However, these tools can only provide some insight into how a device will behave during an EMC test. The usual way is to do a pre-compliance tests which shows you the problem areas. Then you use your own tools (LISN, current probe and H field probes) to track down the source and verify the fixes you made result in a significant reduction of the emissions. Then you go back for another pre-compliance test and if that works out OK, go for the rear compliance test.

Having an EMI option on a spectrum analyser is not really necessary IMHO; if you can do max-hold and averaging the functionality is good enough. Any EMC compliance measurement you make has a very low accuracy anyway due to the rest of the setup.