SMPS & RFI

[pwdrive]

Any one had practical experience in dealing with this menace?

Lawrence.

[tiv48]

i starte.d a thread here about the local pub that bought a laptop mainly to supply music for quize nights and other functions, it was unusable because of noise from the smps this was proved simply by unpluging the ps .
the noise could only be heard through the pa system
after months of auguments with pc world who denied there was a fault they finaly changed the psu witch was different from the origanal this cured the problem
hp had surgested fitting a feret ring in the mains lead so they belived it was rf.
quite a few lcd tvs seem to emit rf at quite high levels enough in one case i know of the wipe out the bottem end of the fm radio band.
then there is rf output from energy effisent light bulbs
it amazes me we get radio receptio at all
rob t

[pwdrive]

It's a pain...

Lawrence.

[ed731pdh]

So much for the CE label.

[Skywave]

What follows will be known by many reading this, but it does attempt to answer the question posed in the opening post. I will refer to 'RFI' (radio frequency interference) in the broadest sense, not solely limited to that caused by SMPSU devices.

Ideally, and generally speaking, the best approach is to identify the source and fit suppression techniques there. Yes, I know - easier said than done! Of course, if the offending device is found to be under your exclusive control, then you are in a much better position to do that suppression than if the offending item is otherwise. In that latter case, contacting the owner of the offending device (if indeed it can be located and identified) is sometimes profitable. As regards, what to do electrically, the usual approaches are to fit filter capacitors and / or chokes and most important of all, ensure that the item is 100% screened in a metal case and that case is solidly earthed.

The next approach - which is usually a bit easier to implement, but less effective - is to reduce the level of interference that is reaching your receiving device. All interference is electro-magnetic radiation and as such it needs a path into your receiving device. Three routes are commonly identified: (1) direct radiation; (2) mains wiring and connecting wiring external to the affected device picking up the interference; (3) interference being picked up by any aerial attached to the affected device.
(1) usually occurs rarely: commonly caused by something very close to the affected device and the offending device producing a very large radiated field: usually relatively easy to locate. Usual cure is fitment of filters and complete screening of the affected device; a total elimination of the interference may not be possible: often only a substantial reduction. A typical example is a badly-adjusted Amateur Radio transmitter affecting that household's T.V. (But here tackling the source is obviously the best route).
(2) pick-up on external wiring: fit chokes and capacitors. Capacitors across mains inputs must be approved types: X and Y rated.
(3) aerial pick-up: if the source is directional, changing to a directional aerial (to reduce the amount of pick-up but not so as to dramatically reduce the wanted signal level) can help. Also paying attention to the aerial's down-lead, be this co-ax or other: are all the joints clean and intact? If an earth is needed (e.g. a 'long-wire' type aerial), is an earth lead fitted and is it effective ('effectiveness' of an R.F. earth is not a trivial matter). If the interference is not of a broad-band nature, (i.e. specific to a particular frequency or frequencies), the fitment of an R.F. trap in the aerial down-lead is a favourite cure. With a 'long-wire' type of aerial, it is worth remembering that most interference from house wiring is vertically-polarized. So if you can arrange that the lower part of the down-lead is screened (and that screen is connected to earth), the upper part of the down-lead may be in an interference-free zone. Although the screening will reduce the signal level, the overall effect will be an improvement in the interference-to-signal ratio at the receiver.

And finally, these days there are some forms of interference that we are simply having to live with - such as the wide-band interference generated by PLT, the nuisance-value of which is largely ignored by the powers-that-be.

References. For those who access to the following publications, the entries therein are relevant.
1. Amateur Radio Techniques, 7th. ed., Pat Hawker (G3VA); RSGB publication: p.274.
2. Ditto, 4th. ed., p. 200.
3.The Radio Communication Handbook, 6th. ed, RSGB publication, chapter 17: 'EMC compatibility'.
4. Ditto, 4th. ed., chapters 4 and 18.
5. Ditto, 3rd. ed., pp. 447 and after.
6. The VHF / UHF manual, 3rd. ed., RSGB, chapter 4.
7. The Radio Amateur's Handbook, 1976, ARRL, pp. 499 and after.
8. The Radio Designer's Handbook, 4th. ed., Langford-Smith, pp. 1279 and after.
9. The Radio & Television Engineers' Reference Book, 4th. ed., Hawker & Pannet, section 33.

And many others - but that lot'll do for now!

Al.

[Mark R]

Sorry Al, being a bit dim here and not liking acronyms, what is PLT?

cheers Mark

[tiv48]

i have tried fitting x2 caps ine the mains socket fitting ferrit rings and enclosing psu in an earted tin ,all with no luck, on one dell that was very bad i ended up useing a toshiba psu that cured the fault
rob

[Skywave]

Sorry Al, being a bit dim here and not liking acronyms, what is PLT?
cheers Mark

No need to apologise: electrical technology is full of damn acronyms: it was ever thus!
PLT: Power Line Telecommunications.

From OfCom:
http://stakeholders.ofcom.org.uk/enforce...ement/plt/

And for something that has more 'bite':
http://www.ban-plt.co.uk/

Al.

[Skywave]

Rob: with those free-standing external SMPSUs that power lap-tops, I too have found that most of the interference comes from the mains lead. Now I've never tried this, but in theory a narrow-band filter, resonant at 50 Hz, fitted at the mains input, should eliminate most of the hash. Trouble is of course that that will require very low inductance coils and quite large capacity capacitors. Nevertheless, it might be worth investigating, especially where the mains current is relatively small - as in a lap-top SMPSU.

Al.

[pwdrive]

I received a filtered multioutlet socket strip today, it has made a big difference, I would say the RFI is now about 90% down so it's a good start, filtering the supply to the table (my work bench) will be the next step.

Thanks for everyones input.

Lawrence.