Products > Test Equipment
EMC pre compliance test equipment recommendations
<< < (3/3)
basilic:
Hey, I'm replying to your message because I'm currently at the next stage.

Three years ago, I explained to our managers that we can't claim something is good without any measurement, and we failed our emission tests (conducted but with additional precautions, and then radiated).

We were authorized to spend a little to purchase a spectrum analyzer (R&S FPC1000 3GHz, not enough budget for an FPC1500 or higher), a small biconical antenna (due to limited storage space), and a second-hand 50µH LISN. With this equipment, we successfully patched the design (both hardware and software patches).

Last month, we failed the conducted immunity tests (for industrial level).

After extensive discussions about the impossibility of reproducing the phenomena and solving it without many tests in the lab, we are planning to purchase more equipment for immunity testing: amplifiers, a TEM CELL, attenuators, and a generator.

Regarding the generator, I'm seeking advice on the best choice, whether to invest in a new spectrum analyzer with a tracking generator (SSA3075X) or just a generator (SSA5060X).

We need to test immunity up to 6GHz (previously limited to 2.7GHz) as per the standard.

In either case (generator or spectrum analyzer), we will need amplifiers for conducted immunity (level 3 requires at least 40dBm with a CDN).

If anyone has an opinion, I would appreciate hearing it before investing a significant amount of money without a clear direction.
nctnico:
I'd invest in a general purpose RF generator. For radiated immunity you'll also want modulation (at least AM). Another item you'll need is a power meter so you can measure the output of the amplifier versus frequency and create a correction table. With some simple Python scripting + pyvisa it is pretty simple to create an automated setup that sweeps through the frequencies.
basilic:
If the budget were unlimited, I would prefer an A-brand. However, after many hours of consideration, I only have a budget of 12-14k€ to purchase a setup for conducted (10Vrms 150kHz - 80MHz AM 1kHz) and radiated immunity up to 6GHz (10V/m and 3V/m).

With this budget, my preference is towards Tekbox amplifiers and brands like Siglent for the generator, supplemented by some near-field probes and a DIY CDN.

In the 61000-4-6 standard, the level is set as continuous wave (CW) without modulation, and I can use our oscilloscope to perform this calibration.

I can't quite grasp the distinction between an RF generator (aside from being more powerful but not sufficient in terms of output level, as far as I see it, primarily used to generate continuous wave frequencies) and a spectrum analyzer with a tracking generator. Admittedly, I've never used it. :-//
nctnico:
Do you really need to go to 6GHz for radiated immunity? Recently I got a product tested and in the 1GHz to 6GHz band, only a few spot frequencies (the LTE bands) where tested at the EMC lab.

Unless your product fails those tests, I'd concentrate on the 80MHz to 1GHz band because in that range your product is mostly likely to be affected. For the pre-testing I did, I used this (surplus) amplifier: https://nl.aliexpress.com/item/1005004001551855.html which works pretty well but I've used it up to 5W as I couldn't dissipate more (see below). I've also seen a 16W version. Be sure to mount it on a heatsink though.

If you are going to use an open-sided TEM cell for radiated immunity testing, make sure to also get a shielding tent to contain the RF field. And you'll need a dummy load of some sort to terminate the TEM cell. I bought a Mini-circuits 5W, 20dB attenuator for this purpose (from Mouser as I wanted to get the real deal). The advantage of using an attenuator is that you can also use it to test the amplifier gain.

Another neat test to do is see at which level a product fails when subjected to radiated fields. For practical purposes 10V/m is a bit low; it is possible your product ends up receiving much more due to faulty equipment which is still your problem because your product misbehaves. I'd make sure that it stays OK up to 30V/m.

RF generator versus tracking generator: The generator typically gives a cleaner tone (less harmonics) and has a more precise output level. With a tracking generator you can typically choose a few levels and thats it. If you are going to drive an amplifier, you will want the ability to adjust the output level to follow the frequency characteristic of the amplifier while doing the stepped sweep.
basilic:
For the level, we will conduct our tests at a higher level than required by the standard. The goal is to succeed in the certified lab and exclude uncertainty from our system. We will not declare conformity solely based on the limit line.

We use the 61000-6-2 standard for industrial radiated immunity.

In the 2006 version, the test levels for radiated immunity with 80% AM (1kHz) are as follows:

    80MHz to 1GHz: 10V/m
    1.4GHz to 2GHz: 3V/m
    2GHz to 2.7GHz: 1V/m

This standard changed in 2019 (we were not aware of this change), and now the levels are:

    80 to 1GHz: 10V/m
    1.4GHz to 6GHz: 3V/m

without specifications... but in the 61000-4-3, we can only cover the phone and RF emitter.

I know there isn't a problem with the TEM cell (up to 1GHz) or probe (up to 6GHz) to achieve this level with a small amplifier (0.1W or 0.5W), and it's not an investment for our company (can be placed in current expenses).

I checked the output level resolution in the band 150kHz to 80MHz; we need 38.5dBm (7W) as the output for a 10Vrms test, according to the standard (61000-4-6) on a CDN.

For the spectrum analyzer: SSA3075X-Plus is designed for -40 to 0 dBm by 1dB step (similar on other spectrum analyzers).
For the generator: SSG5060X is designed for -110 to 15dBm by 0.01dB step.

In any case, we need an "expensive" amplifier for the conducted immunity.

To achieve this level, we would buy an amplifier TBMDA-CDN25 that provides up to 44dBm with a 0dBm input (3 dBm maximum input).

If I understand correctly, with the spectrum analyzer and amplifier, we can conduct this test from a minimum level of 4dBm to 39dBm with a maximum of 34/35 steps and much finer steps, and start at a much lower level with the real generator, right?

I don't know if the cleaner tone is an argument for an immunity test.

I will make calculations to see the effect of a 1dBm change on the signal level (I will try to do this on Monday using LTspice to simulate steps of the spectrum analyzer).
Navigation
Message Index
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod