Circuit to test noise immunity of Electronics products..

[ocset]

Hello,
We  wish to see if our  offline Switch Mode LED lights  are immune to noise.   
Therefore, we will make a special “noise  producer” as in the attached. It is simply an LLC converter with one of its switching nodes dragged out of the enclosure for 30 metres. This 30 metres is loosely twisted (it’s  also insulated wire). We will drape this 30 metres of  “noisy wire” over our products as they are working –to see if it makes them go wrong.
Do you think any of the products will go wrong when the “noisy wire” is draped over them?    Might any  operating electronics products be disturbed by being  brought near to such a wire? 
The waveform of the square wave of  the wire is also shown. It is a square wave with 7ns rise time.
LTspice simulation of "noise producer" attached also.

By the way, the LLC  converter ('noise producer') shown  feeds off the output of a fully isolated full bridge converter, which itself feeds off the output of a 240VAC mains PFC stage
 

[dmills]

FFS Treez, there are perfectly good standards for this stuff, apply the immunity tests described in the appropriate standards, not some random lashup (which will NOT be giving you 7ns edges at 30m, pretty sure of that)!

For random ESD testing for getting some confidence before breaking out the real deal, there are far simpler things.
I have a set of lightly modified piezo gas BBQ lighters that by way of various bits of copper tape and screws can produce wicked electric and fast magnetic fields within an inch of so the device, simple, cheap and almost disposable.

Hell a kettering ignition system with a pair of telescopic aerials braised onto the spark plug is a fairly nasty test and is simple to throw together from bits from a scrapyard.

Here https://www.edn.com/electronics-blogs/the-emc-blog/4378152/An-EMC-Troubleshooting-Kit--Part-1a-Emissions- are some ideas.

Regards, Dan.

[CopperCone]

Dude i wanna see your vibration resistance tests i can see your boss going to accounting with a nuncha hammers and power tools he got from his garage and making a buncha office people attack your products like office space in the company parking.

Your company must be running on thin film margins.

You should get a broad band antenna, and a power amp to do testing from 100MHz to 18GHz. You can do this with like 2 horns for 5 to 10k$ and cost of amps probobly 20k for half decent stuff that might give you a clue

I dont know how to rconomically test the lower end of the spectrum. You might want to look into helmholtz coils.

If you connect a bbq starter to a loop it will make a fairly high frequency antenna. You can also jingle coins in a bag or hersies kisses in a plastic dish around it if you are interested in esoteric esd precompliance rituals.

Your wire idea will basically do random ass shit. You need to get yourself access to a vna and connect a wire to it then jiggle it around. Like connect two bnc to alligator clip attachments together and connect it to a vna. Then move it around and look from like 1 to 100 MHz you will see how god damn unstable and random it is.

Maybe if you stand over the device being tested and jiggle the wire around like a cat toy or snake having a seizure you might get sone kind of spectrum test.

And another one is to run a brushed motor near the device with its filter cap removed if it has any. You might want to run some brushed drills over it. Also maybe seperate the wires going to a brushed motor of fair power and put the device between them so the wires dont couple to each other.

To be fair these tests do have merits.

Also test near full power cell phone.

[dmills]

Walkie talkies are good too, as are Ham transmitters for the HF band, but really, if you want more then some vague idea do it properly.

Making up some semi proper kit is not that hard or expensive, and ferrite clamps are not rocket science.
Conducted susceptibility can be improvised to within maybe 6dB or so over the HF band fairly easily.

EFT? You can do something, but it is going to be very approximate.

Conducted emissions you can build a LISN and as the requirement only extends to 30MHz, a SDR can get you into the ballpark with some software to do the Qpk @ 9KHz BW.
Calibration is problematic, but if you have some box with known emissions measured in a proper facility you can derive a rough cal by measuring it with your kit.

Radiated? The FCC engineers back in the day published some designs for dipole/balun combinations that had known antenna factors, add a SA and an open field site... It is an expensive pain to do proper measurements but you can do comparative stuff fairly easily, and once you get above a few hundred MHz or so there are companies making PCB printed log-periodic antennas with more or less known factors for not stupid money.

An old 20 foot shipping container in the carpark can be your friend for some of this stuff!

Regards, Dan.

[basinstreetdesign]

For conducted noise tests you can try the "file" test.  One previous employer used this regularly on his products prior to sending to an actual test facility.

Go to the hardware store and buy a $5 wood file with coarse teeth.  Drill a hole through it.  Get an electric mains cable and interrupt the hot side of it.   Connect the file to the wall plug side of the wire break.  Using insulated gloves, plug the cable into the DUT and into the wall socket.  With the DUT-side wire connected to the file, get the DUT operating and stable.  Then run the DUT-side wire up and down the file to generate lots of sparks, shit, crap and corruption with noise conducted and radiated.  If you can do that for 60 sec without a hiccup from the DUT, it passes.

[oldway]

In the 80's we were using a rather rudimentary but very effective test.

A relay with the  suitable coil voltage was connected as an oscillator (a contact normally closed in series with the coil) to the low voltage supply of the DUT. If everything continued to function normally, the immunity was considered sufficient. 

[ocset]

Thanks, we are still interested in the possibility of a "dragged out" switching node exactly like shown causing other electronics to fail. -Or even making the power supply itself fail (the one thats had its switching node 'dragged out' as shown)
Its a long story why...if you wish i can tell you?
This isnt anything to do with official regulatory noise immunity testing.

[Gyro]

Dude i wanna see your vibration resistance tests i can see your boss going to accounting with a nuncha hammers and power tools he got from his garage and making a buncha office people attack your products like office space in the company parking.

Hey, don't knock it, I used to take Set-top-boxes out into the car park and drop them to see if anything safely critical came loose. A crude, cheap, and effective first line test. 

[CopperCone]

maybe you can line up the workers in 2 rows and drag the product through on a rope like a gang initiation 

and I am interested about why you would do independent weird tests I like extra effort, it makes me feel I am getting a better product when I get extra tests

[dazz1]

Hi
Google "chattering relay".  It is a recognised test method.
https://www.edn.com/electronics-blogs/the-emc-blog/4407544/Unusual-EMI-Sources

Dazz

[jmelson]

There are commercial immunity testers.  When IBM commissioned a couple 370s at work, they brought in this gadget with an antenna mounted on a box, and the device mounted on a tripod.  They set it a couple feet from the machine and plugged it in.
It  apparently had a HV power supply and a spark gap in it, and would sit there going snap-snap-snap.  They moved it around the machine, facing all sides of the cabinet, while the CPU ran diagnostics.  Then, the opened one of the covers on the machine, and turned the EMI source in, and the diagnostic immediately failed.

I think you could easily build something like this.  Getting the antenna set up to give a flat, wideband spectrum might be a little bit of a challenge.

Jon

[CopperCone]

I am gonna build a chattering relay with a old elevator relay I have its fascinating.