LV transformer specs

I think you should check out your wiring

What you're saying sounds a bit like "there was no fault until the fault occurred" if you see what I mean

Reply to
geoff
Loading thread data ...

I do, and I agree. The problem is I'm reasonably competent and safe with basic household wiring, but this has got me stumped.

Here's a picture of how the lights are installed (it was a kit from B&Q):

formatting link
transformer is located in the ceiling cavity above where I'm standing to take the photo. I've moved the lights closer to that end of the room to shorten the circuit, and the new transformer appears to be working fine (it barely even got warm last night).

Apart from the failed transformer, nothing else has changed as far as the household electrics go.

Here's the current problem. I have a portable radio in the kitchen, and whenever I turn the lights on now, they generate quite a lot of RF interference (enough that I can't hear the radio).

I have *no clue* how to diagnose the cause of this, any ideas?

Cheers,

Reply to
Jeremy Collins

No great mystery here: the new "transformer" you've now put in isn't a "transformer" of the classic form (primary coil of larger number of small-guage turns, ferrite core, secondary coil on other bit of core with smaller number of larger-guage turns, transfer of electrickery into magnetism and back again, ratio of voltage-in:voltage-out equal (ignoring losses) to ratio of primary turns to secondary turns), but a switched-mode power-supply unit, whose design revolves around a relatively-high-frequency oscillator running at hundreds of kilohertz, and b'ing about with that waveform's shape to keep the output voltage constant under varying loads. The design of SMPSUs is inherently prone to causing radio-frequency interference (RFI), which is wot you're hearing on your radio.

The cheaper the "electronic transformer" (SMPSU), the less attention is likely to have been paid to reducing RFI; and those 12V output leads make a fine aerial for broadcasting the RFI onwards - this, as pointed out in an earlier posting in this thread, is one of the reasons for the manufacturer telling you to keep the output leads short (this helps them claim conformance with RFI regs, you see ;-).

You can try passing the LV output through a clip-on ferrite lump, if you can come into possession of such (if you have an unused and unloved PC video cable with a Gert Big Lump on it, you can try using said Lump); you can try shortening the LV leads to the minimum; you can try a different SPMSU (hint: old computer PSUs have 12V outputs, but not rated for 100W, and some need a dummy load on the 5V output to regulate properly); or you can stump up for a simple, honest-to-goodness, "real" transformer, a toroid by choice, which runs at the 50Hz which God and Maxwell and Tesla intended AC to flow at ;-)

HTH - Stefek

Reply to
stefek.zaba

Change to a toroidal transformer.

Reply to
The Natural Philosopher

Reply to
Jeremy Collins
[snip]

Aahh...

Ah, yes (again).

I have one, is that enough? And just splice the section of video lead with the GBL onto the 12V leads with a connector block? Worth a go I reckon!

Can't really be done, they have to stretch the length of the kitchen (>3M) to suspend the lights (photo in previous post).

I've been building PCs for about 8 years, so I have couple spare PSUs, and while your idea does have a certain appeal, I'm not sure I want to put a PC PSU in the ceiling cavity. Wierd noises from above, plus they tend to suck in a lot of dust.

Although the 5V dummy load *could* be a PC case fan for extra cooling, I doubt it'd be enough to disconnect the internal PSU fan.

Nice idea though! (Imagine the reaction of future owners when they discovered it...)

I think I'll try the DIY RF shielding first, I've already bought two transformers this week :-)

Thanks for the informative reply!

Cheers,

Reply to
Jeremy Collins

Definitely the simplest answer, but even at tlc a lighting transformer of that size isn't far off 40quid.

Alternative option is filtering, but the problem is that your fancy lighting setup has created a near perfect aerial (long and lots of lovely loops) for even the smallest amount of RFI escaping the 'transformer'.

May be worth spending a up to a tenner on filtering, but without proper knowledge of the problem (and sometimes with) there is no guarantee that the filter will work, so the tenner would be wasted. Options are:

  1. Ferrite ring
    formatting link
    stock code AM35Q (2quid) - take the LV cable from the transformer close to the transformer and loop through the ring as many times as it will fit (try to get 6 or more turns (loops)) before going off the lights; probably use a length of flexible twin for the winding. If this improves the situation, more can be added in a chain.
  2. Common Mode Choke
    formatting link
    stock code 188-9040 (9quid) or equiv, but must be 10A rated. Just wire in with a chocolate block red to red, black to black. Can also be used in conjunction with the ferrite ring.
  3. Purpose made RFI filter - forget it, too expensive, buy the toroidal instead.

Good luck.

btw: What are you listening to, AM or FM, what band?

ps: Just saw your reply to Stefek before sending. Forget shielding, it won't do a thing for you as it will be your leads that are radiating - or did you mean filtering? The ferrite from the monitor lead will be 'tuned' for 25MHz up so isn't likely to be of help, but try it anyway. Ferrites come in many flavours so one designed for just HF RFI will likely have better effect - actually if you have a scrap computer PSU then hack it open & pinch a toroid from that - there's bound to be one filtering the mains input. Use lots of turns - efficacy goes up for the square of the turns.

Once again, good luck.

HTH

Reply to
fred

I'm going to try that.

Thanks for the link. So the suppression choke sits in series on the LV output from the transoformer?

FM. Sod's Law applies, in fact. The radio is tuned almost exclusively to Radio 4, which gets the worst interference! Radio 1, 2 etc aren't affected at all!

I think I did mean filtering, this is a bit new to me. I've got a couple if PDUs that stull work, so rather than hack one apart I think I'll try your previous suggestions, in order of cheapness.

Thanks for taking the time to reply; I'm learning a lot about LV lighting from the contributions to this thread!

Cheers,

Reply to
Jeremy Collins

Yes, that's right. The choke will just be a ferrite, but of material that suits your kind of interference ie harmonics from 100-200k ish switching, so say

100k to 100M.

It's just that FM is much less sensitive to this kind of interference, so you must be pumping out quite a lot. Good news is that once your filtering drops the RFI below a certain threshold the interference to the radio will stop completely, almost in the same way as with digital, but different . . .

I'm afraid a lot is down to your fancy lighting rig, having the feed & return separated by about a foot across the ceiling like that is creating a fabulous loop aerial cf. portable TV aerial. I have some really dirty light dimming switchers, but have never had a squeak from the radio as all feed by cables side by side ie twin & earth.

Some caps between the feed & return wires may help too, must be non polar and also small value to avoid upsetting the stability of the transformer, consider maplin stock code RA49D (100nF ceramic, 10p) or RA52G (470nF ceramic, 24p). Best placed away from the transformer I think (stability), maybe just before/after the choke, maybe just before the cable feeds you christmas tree ;-). Watch the body of the cap when bending the leads, they're really fragile, use pliers just before the body, keep leads short.

Welcome

Reply to
fred

Hi,

Try unplugging the bulbs and plugging them in one at a time, might tell something.

Or add a coil of wire into the LV circuit, might reduce interference on your favourite station.

The above has about a 5% chance of working though :))) Chokes or a normal transformer are a far better bet, 2 separate chokes would be better than a common mode one.

cheers, Pete.

Reply to
Pete C

That will be the switched mode power supply generating RF, your power rails being fairly good aerials The only way round it is to go back to a wound transformer

Reply to
geoff

no no. Remove lump from lead and thread lump over existing wires. But to be honest youre probably wasting your time. The solution is a toroidal transformer. You might get a refund on the tronic one on the basis that its not fit for the purpose.

Regards, NT

PS a PC PSU wont give enough power to run the lights.

Reply to
N. Thornton

Yep, (bugger). I tried a couple of ferrite rings since I needed some stuff from Maplin anyway, but they didn't make much difference.

Bugger. Oh well, I've learned a lot from the experience!

Reply to
Jeremy Collins

Sorry to hear that.

If you are in a mood to play some more, try splitting LV feed & return near the transformer & winding 10 turns of each single conductor on a separate toroid and, if you have any, add a cap as previously described between the two. This will tackle a different mode (differential) and, as someone else suggested, may be a more likely problem in your setup.

This sort of thing:

________ | | | |----TOROID1---|------------- FEED | TRANSF | === CAP 100NF | |----TOROID2---|------------- RTN |________|

Keep the toroids separated by 25mm+

May still be crap, but costs nothing . . .

Reply to
fred

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.