Can we get away without using shielded cable?

[Faringdon]

We have  a plastic enclosed product comprising multiple PCBs...with a 20W offline SMPS supplying power as 12V rail. System Contains 3 other Buck converters. The 4 SMPS's are all on the same PCB. This PCB supplies power by cables to other PCBs, which dont have any SMPS's  on them. For the non-SMPS PCBs that are connected by wires between them, can we get away without shielded cable for these?

We wish to pass EN55032 (radiated emissions)

[wraper]

If designed properly, it can pass radiated emissions without any shields and shielded cables. Look at phone chargers, they have similar power rating and I don't know any which are shielded. Also It would be great if you don't create tons of threads about the same thing with some variations.

[Faringdon]

If designed properly, it can pass radiated emissions without any shields and shielded cables. Look at phone chargers, they have similar power rating and I don't know any which are shielded.

Thanks, as you know, the phone charger plastic enclosure is likely to be foil lined...but if not....its quite likely to not pass radiated EMC regulations...

yes, but its interesting, ....Please add your thoughts to this....just imagine a company's product didnt comply with radiated EMC  regs....then the standards bodys would take samples of the company's product, do radiated EMC tests on them...and ban their product, right?....As you know, this never happens. A Radiated EMC test on an apparatus containing an offline SMPS takes ages to do  and costs utterly  shed loads of money...and can only be done in super-expensive  chambers which are pretty rare to find......the EMC antenna has to be moved all over the place to re-scan the product from different angles and elevations...then any failure peaks have to be re-scanned to see if they really are fails....takes ages  and ages. So no standards bodies have organised  radiated EMC testing for company products.
So many company's products simply dont comply with radiated EMC.....but what a small startup has to bear in mind, is that a big competitor may well sneak their product into a radiated EMC test...so that they can proove it fails....then they would put the small  startup out of business by using their highly payed lawyers to do this.
..But even that rarely happens....because if you bomb your enemies...then your enemies may bomb you too. So in fact, in radiated emissions, as well as many other  electrical standards areas....companys in a particular sector tend to have "amnestys", whereby neither criticises the radiated EMC (and other) failures of each other....and they live and let live. This often also happens with PFC (EN610000-3-2)...companys simply avoid using PFC on SMPS >75W...because all their competitors dont use it either....and neither criticises the other.
I believe we all know that the standards bodies (who answer ultimately to the governments), dont actually test samples of company products in order to see if they comply with radiated EMC regulations...as discussed...its just too expensive, and takes too long....there is a reliance on the goodwill of companys  to oversee it themselves. I mean, Western Governments have suffered such a  "dereliction of duty" such that the vast majority of the Western general  electronics sector has been outsourced to companies in the Far East.....leaving gaping holes in Western capability.....i'm sure we all realise that Governments that behave like that, are far from having the "upright-ness" towards  implementing measures to check and ensure products have the necessary  radiated EMC compliance.

[Gyro]

You really are becoming a stuck record Treez. Hell, what do I mean becoming? 

https://www.eevblog.com/forum/renewable-energy/offline-isolated-smps-must-be-in-metal-enclosure-(to-pass-radiated-emissions)/msg3607222/#msg3607222

I count at least ten threads that you've started on this 'product'.

[wraper]

If designed properly, it can pass radiated emissions without any shields and shielded cables. Look at phone chargers, they have similar power rating and I don't know any which are shielded.

Thanks, as you know, the phone charger plastic enclosure is likely to be foil lined...but if not....its quite likely to not pass radiated EMC regulations...

You just pulled it out of nowhere. They are not. https://www.chargerlab.com/apple-18w-usb-c-power-adapter-a1695-teardown-review-beautiful-inside-and-out/

[Faringdon]

Thanks, but has it actually passed radiated Emissions testing to EN55032?
Also, is it using "special" plastic infused with conductive material so it is covertly shielded?

Heres another one with no foil lining...just plastic case...they show full Conducted EMC scan plots with pass....but no radiated emissions plots are shown , or even mentioned.....sound ominous to me....
https://www.power.com/sites/default/files/product_document/design_example/der-600_45watt_usb_pd_3_power_supply_using_innoswitch3-pro_powigan.pdf

Power integrations test their offline SMPS's to hell and back in their Application reports (just download  one and see). The fact that Power integrations do not show any kind of radiated emissions test on their non-metal-encased offline power supplies is very ominous. It suggests that they do not pass radiated emissions (unless of course, they are placed in a metal enclosure, which of course, they often will be.

[mikeselectricstuff]

yes, but its interesting, ....Please add your thoughts to this....just imagine a company's product didnt comply with radiated EMC  regs....then the standards bodys would take samples of the company's product, do radiated EMC tests on them...and ban their product, right?....As you know, this never happens. A Radiated EMC test on an apparatus containing an offline SMPS takes ages to do  and costs utterly  shed loads of money...and can only be done in super-expensive  chambers which are pretty rare to find......the EMC antenna has to be moved all over the place to re-scan the product from different angles and elevations...then any failure peaks have to be re-scanned to see if they really are fails....takes ages  and ages. So no standards bodies have organised  radiated EMC testing for company products.
So many company's products simply dont comply with radiated EMC.....but what a small startup has to bear in mind, is that a big competitor may well sneak their product into a radiated EMC test...so that they can proove it fails....then they would put the small  startup out of business by using their highly payed lawyers to do this.
..But even that rarely happens....because if you bomb your enemies...then your enemies may bomb you too. So in fact, in radiated emissions, as well as many other  electrical standards areas....companys in a particular sector tend to have "amnestys", whereby neither criticises the radiated EMC (and other) failures of each other....and they live and let live. This often also happens with PFC (EN610000-3-2)...companys simply avoid using PFC on SMPS >75W...because all their competitors dont use it either....and neither criticises the other.
I believe we all know that the standards bodies (who answer ultimately to the governments), dont actually test samples of company products in order to see if they comply with radiated EMC regulations...as discussed...its just too expensive, and takes too long....there is a reliance on the goodwill of companys  to oversee it themselves. I mean, Western Governments have suffered such a  "dereliction of duty" such that the vast majority of the Western general  electronics sector has been outsourced to companies in the Far East.....leaving gaping holes in Western capability.....i'm sure we all realise that Governments that behave like that, are far from having the "upright-ness" towards  implementing measures to check and ensure products have the necessary  radiated EMC compliance.

Standards bodies are responsible for writing standards, not enforcement or even testing.
Enforcement varies by country, in the UK it's Trading Standards, who are typically more interested in dangerous kids' toys, fake cigarettes, lethal phone chargers etc. AFAIK there have only been a handful of prosecutions here.
In most cases it's a complaints based  process. I believe Germany is the only place where they actively go out and buy and test products. I don't know if this is still the case, but chances are it would be high-volume consumer products.
However there are other situations - if you are selling via a distributor, the distributor may want to see paperwork that the product complies with all relevant standards.  If you are selling to another company, they may want to see documentation. Your competitors may also report your products if they find them noncompliant.
In practice, unless your product causes serious problems, chances are nobody will ever check.

[wraper]

Thanks, but has it actually passed radiated Emissions testing to EN55032?

That particular charger is for Chinese market so it passed another similar emissions testing.

Also, is it using "special" plastic infused with conductive material so it is covertly shielded?

No
Here is European charger teardown and it's not shielded either https://budgetlightforum.com/node/60580

[Faringdon]

Thanks, yes, ive taken a great many apart, some are foil lined and some arent.
The burning question for us, would be...do the non foil lined ones pass radiated emissions to EN55032.

[wraper]

the non foil lined ones pass radiated emissions to EN55032.

If they did not, they would be illegal to sell. And the big guys like Apple or Samsung sell chargers which are not shielded.

[Faringdon]

OK thanks, and then we go back to the details of the reply #2 above...

[Faringdon]

Hi,
We have a system  in a plastic enclosure comprising multiple PCBs interconnected by cables and looms. One PCB comprises the 4 SMPS's. There is an offline 20W SMPS on this PCB also. Our circuit ground is directly connected to earth. We need cable screens for all the cables. Do you agree that we must connect our cable screens to circuit ground at each  end of the cable? The thing is, do you also agree that we need to put ferrite beads in the connection leading to the cable screen, in every case?
You see, the situation is that we do not have a separate chassis ground, because our enclosure is plastic...so therefore we cannot  connect our cable screens to any chassis ground.

[SteveyG]

Hi,
We have a system  in a plastic enclosure comprising multiple PCBs interconnected by cables and looms. One PCB comprises the 4 SMPS's. There is an offline 20W SMPS on this PCB also. Our circuit ground is directly connected to earth. We need cable screens for all the cables. Do you agree that we must connect our cable screens to circuit ground at each  end of the cable? The thing is, do you also agree that we need to put ferrite beads in the connection leading to the cable screen, in every case?
You see, the situation is that we do not have a separate chassis ground, because our enclosure is plastic...so therefore we cannot  connect our cable screens to any chassis ground.

It depends entirely on the nature of the emissions. If you can avoid your cable being an antenna for whatever frequencies and harmonics are being generated by your converter then there's no need to use a shielded cable.

The first step is to understand your converter and design it with parts and in such a way that it doesn't cause a problem. Next would be to suppress the noise. Final is to shield the noise.

If you've got multiple converters, you need to do some analysis and/or testing to work out the best shielding strategy, but I would still try to suppress the emissions first.

[dmills]

Generally a mixture of a little filtering at each end of the cable (Remember noise can come in from the load end as well!) gets it done providing the power supplies are reasonably clean to start with.

Loop area matters, so twisted pairs can be your friends.
For each and every signal in the design consider the current loop formed (ALL of the current loop formed) and seek to minimise its area.

Sometimes synchronising the power supplies is helpful (For reducing input ripple current if nothing else), and in extremis synchronising to a deliberately spread spectrum clock is a thing.

Remember that for an aerial to be effective it needs to be reasonably long compared to the wavelength, so there is little point from an EMC perspective in worrying about circulating currents at frequencies where the total loop length is less then about 1/10th wavelength.

Get and read "Electromagnetic compatibility" by H. Ott and "High speed signal propagation - advanced black magic" by H. Johnson, and be enlightened, it is money VERY well spent.