EMC pre compliance low cost lab

Hello,

We are thinking to buy material for a EMC precompliance lab. We have problems mainly with radiated emissions.

We are thinking to buy close field probe set. Are they reliable? Can i measure improvements in design before send the product to the lab?

Another option is buy a cheap bilog antenna (UPA 6192 Teseq), but havent′t got any anechoic chamber. Can i use it in open field (3m-10m)?

Another option is to buy a GTEM cell, but it is very expensive.

Thanks!

Field probes measure local E and H field strengths, not radiated power. In so far, it's not a question of reliability rather than designer's skill to use them effectively. In my opinion, they are powerful tools to locate EMI sources. Also open field measurements can be helpful.

Hi,

> We are thinking to buy close field probe set. Are they reliable?
> Can i measure improvements in design before send the product to the lab?

They give you an estimate of the Electric field at 3 m.
You will also need an EMI receiver.

> Can i use it in open field (3m-10m)?
Yes, absolutely. In this case, you will need an Open Site Area of 10 meters.
The steps are:


1) The EMI receiver measures the Electric Field in the environment with your device off,
at any polarization and height of the receiving antenna (read the norm).

2) Turn on the device and repeat the measures.

3) Program the EMI receiver to substract the environmental Electric Field from the Field radiated by your device.

In this way you can get a pretty good estimation of the Electric Field radiated by your device.


Enrico Migliore

Hello Enrico,

I have a FSH6 R&S Spectrum analyzer. I think that it hasn′t got the option to substract but at least i going to have an stimation. Thanks!

you need to tell us the frequency range you are interested in.

I think, exact subtraction won't work anyway. You should rather record the existing background "noise" and discrete lines to know which signals are generated by your device. Consider that the background spectrum won't be constant.

Hi,

> I think, exact subtraction won't work anyway.
It does. This option is present in recent EMI receivers and should be used to figure out the direction/directions of maxium radiation and of course the frequencies where the Electric Field exceeds the limit.

Enrico Migliore

Hello all!

I have been reading about EMC testing and OATs. I would like to make the tests in a parking (asphalt) bui i don′t know if i need a ground plane. The main goal of the test are to compare some circuits, not to create a OAT for certification.

Hello All,

Finally i bought the precompliance antenna (UPA 6192 Teseq) but i can′t see the exact emission patern we obtain in the certificate lab. I can see some peaks but not the exact profile. I think that the noise of the FSH6 R&S spectrum analyzer hide the radiations. I try it in a OAT but can see noise around 50Mhz. Can i make something?

Hi,

keep in mind that you need a conductor plane connected to earth, below the wood table.

Your LISN must be connected to earth.

The metal plane is needed because they want to measure:

how well the lines of force of the electric field of your DUT induce DISPLACEMENT currents on the plane.

If you don't have a metal plane your DUT won't induce any DISPLACEMENT currents.

Best Regards,
Enrico Migliore

Right for the respective conducted emission measurements, but not related to the previous discussion of radiated emissions.

To see the low level signal, you should use a calibrated low frequency amplifier to boost it by maybe 20 dB. Also, you need to measure BOTH polarizations to get the lab results (i.e. you need to rotate the antenna 90 degrees on its axis during half of the measurements).

Also, in general, you are not going to repeat the exact measurements you see in one lab in another. There are just too many changes, and they all effect the measurement.

I agree about the pre-amplifier suggestion as best option. SA noise level strongly depends and demodulator and video bandwith settings. Did you use a standard EMC 9 kHz BW setting? Also, did you switch-off the input attenuator?

At these lower frequencies, where the inefficiencies of a compact wide-band 'test' antenna falls off one *can* construct a 1/2 half wave dipole and make measurements, just substitute the appropriate correction or 'antenna' factor to offset gains/losses properly ... I have used a couple of dipoles cur for 40 MHz and also 50 MHz, to cover that end of the range where the sensitivity of the wideband (but lossy) compact antenna doesn't drive the signal analyzer (could be a spectrum analyzer or a tuned RFI receiver) with much signal.

BTW, It's the way they used to do it, and, it is the way they verify operation of those wide-band compact antennas.

A paper on how to construct a suitable "Robert's BALAN" and use with the associated antennas:

Title: 20 to 1000 MHz ANTENNAS DEVELOPED BY ROBERTS
. . http://dashfoundation.com/Archives/R...20Antennas.pdf

Roberts was an engineer with the FCC involved with compliance testing.

Jim

- - - Updated - - -

This could affect radiated emissions ... where the power leads/power cord can 'radiate' if not routed to a LISN (which acts as a LP filter to RF). I just briefly glanced at a radiated test protocol and it showed a ground plane in use below the UUT ...

Prudence would lead one to use a LISN tied to the ground plane although the radiated test protocols may not specifically call for the use of a LISN ... one would have to double-check their test requirements to be sure though. The LISN does act as a termination to a power lead or wire (on account of the 50 Loads imposed due to test equipment or the pads in front of the test equipment.)

Jim