Low voltage downlights - individual transformers?

Hi, I'm about to fit 8 low voltage downlights in my kitchen ceiling (yes, I'll replace them with LEDs in a few years when that becomes viable). Question is, is there any advantage to individual transformers (which my bathroom lights have) rather than use a pair of

250W transformers to power 4 bulbs each. The cost saving is not really significant, I'm wondering if there's any advantage to individual transformers other than losing only a single light when it fails.
Reply to
pcb1962
Loading thread data ...

Its lightly easier to arrange low loss LV cabling with a transformer per unit.

Otherwise I have found almost zero difference in either cost or performance.

Reply to
The Natural Philosopher

4x as many transformers means 4x as many failures per time, but each failure affects 1 light rather than 4. So fitting that many is probably counterproductive.

NT

Reply to
NT

Individual transformers are great for retrofitting LV in place of mains fittings. Less advantage if going that way from scratch. Personally I would go for 2 transformers doing 4 each.

Reply to
John Rumm

The Natural Philosopher wibbled on Monday 28 June 2010 15:46

I have also found no issues with 5 x 20W lamps off one PSU, even though 3 of the lamps are are the far end of the physical circuit and 2 lamps are about

1m closer to the PSU in terms of the cable.

I used 2 core 1.5mm2 fine stranded sheathed flex for mine (silicone to the lamp drops - not *really* necessary but utterly heatproof, and regular flex interlinking the 2 JBs back to the PSU. Up to 3m of cable from PSU to furthest lamp. No problems - all lamps appear equally bright.

Combined them with a dimmable PSU and dimmer pull switch and the result is quite nice for a bathroom. Soft start on the dimmer might make the bloody bulbs last longer than 5 minutes too! ;->

I think the OP's 2 PSUs is a perfectly good arrangement WRT to being less inconvenienced if a PSU blows up. Putting the PSUs in an accessible position would be even better (I did) :)

Reply to
Tim Watts

Particularly relevant if you are putting them in a suspended ceiling below the real one...

And, if Tim's 5x20W are not on a 250W transformer, there might conceivably be problems if a later user were to opt for a super bright option. Not likely in this case but as I noted in the similar recent thread, our kitchen ones were originally designed for 20W lamps, with no spare capacity, and when one 30W lamp got into the system, they kept turning themselves on and off as the transformer overheated. Hence, I also know about the suspended ceiling bit...

S
Reply to
Spamlet

Oh dear. Might want to look back over many threads here on effective kitchen lighting schemes.

LED lighting may well become viable eventually, but I don't think it will ever be in the form any anything which you can retrofit a 50W MR16 and get equivalent output - it will be completely different designs.

When I've used LV lighting with multiple fittings, I use more than one transformer, but not one per lamp. You need to be able to make very good LV (high current) connections though.

Reply to
Andrew Gabriel

hear hear

formatting link

Reply to
NT

To be fair the current crop of real LED MR16, not ebay 3 for a fiver fodder, will give a narrow spot 20W halogen a run

formatting link
of them getting close to 35W equivalence, not cheap so payback in non commercial settings slow.

Cheers Adam

Reply to
Adam Aglionby

I already have very effective halogen under-cupboard lights to provide task lighting, the downlights are just to light the floor area and to provide illumination over the sink.

But it will still fit in a round hole in the ceiling - I wasn't thinking of a direct MR16 replacement, when LED downlights become a satisfactory option I expect to replace the mix of halogen GU10s, CFL GU10s and MR16s that I currently have with new fittings throughout.

Yes it seems that 4 bulbs each on a pair of 250W transformers is the sensible choice.

Reply to
pcb1962

250/4 =3D 62.5W an odd wattage indeed.

Underrunning a 250W wound transformer will let the voltage rise andd will give you short lamp life.

Cheers Adam

Reply to
Adam Aglionby

,

Not it its an electronic transformer, or a decently wound toroidal.

Anyway 3 ft of 30A cable will drop the volts another volt if it bothers you.

Reply to
The Natural Philosopher

Don't ever buy cable from TNP, it must be cr@p he has.

Reply to
dennis

? Explain. It's a long time since school, but I thought the output voltage was fixed by the turns ratio and supply voltage. 'Underrunning' should simply just let the tf run cooler I would have thought. Voltage should only vary from bulb to bulb if they were connected in series: which, hopefully, is not being done here. For a safety margin, a tf should surely be big enough to take the maximum wattage bulbs that might be substituted for the originals by an unsuspecting later user.

On the other hand, there is, indeed, a need to make good connections, as I have found surprisingly hefty-looking connectors have melted when the grub screws weren't up to the job, and also that the connecting pins on the lamps themselves can melt - just as the ones in headlight bulbs often do.

S
Reply to
Spamlet

article,

Not sure what that calculation is designed to show... you could have done 250/5 = 50

The electronic ones are switched mode PSUs in reality, and as long as you have the minimum load on them, they are happy, and maintain a fairly well regulated output. A wound one is unlikely to fit through the hole in the ceiling!

Reply to
John Rumm

There is am implicit resistance in the primary and secondary windings too: How relevant that is depends on how cheap and nasty the unit is.

Also.'transformer';' means often 'SMPSU' and these are regulated units, so don't have much in te w3ay of a virtual output impedance anyway.

I see no brightness changes on bulbs if a mate dies. On my SMPSU type transformers.

I DO see a significantly slower time to come up and reach working temps in the triple that is fed through 2 meters of 30A cooker cable though. It may be only taking 13A or so, but it sure loses a volt or so down its length.

Reply to
The Natural Philosopher

Are you sure that's not just a soft start transformer ramping up?

I am not sure what constitutes "30A cooker cable", but let's say its

4.0mm^2 T&E. That exhibits a voltage drop of around 11mV per amp per meter. So even at full load you get 0.011 x 30 x 2 = 0.66V, and at 13A that would be under 0.3V. If you are losing a volt then it suggests either an under specced cable, or a dodgy connection somewhere.
Reply to
John Rumm

Now consider what I wrote, and consider what current the lamps take on startup.

The resistance makes a LOT of difference to cold filaments. Less at full brightness.

0.3 volts is 2.5% in any case, of 12v.

It may well be a tad longer than 2m as well.

Reply to
The Natural Philosopher

I'll be using these,

formatting link
four 50W lamps on each transformer.

Reply to
pcb1962

Missing factor is called regulation, E core trafos have better regulation than torrodials which can easily go 15% over at low load.

Dosent affect many electronic trafos really, some do have minimum load though.

Don`t wire them in parallel either , then you need the whole run rated to the full load....

Arguments either way really, trafo per lamp is really my preferred method now.

Its suprising waht a mere 4A at 12V can do ;-)

Cheers Adam

Reply to
Adam Aglionby

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.