Multi Meters

[The Medway Handyman]

Can I ask the electrically literate here two questions?

I have a basic digital multi meter

to test a lighting transformer the other day, 230v to 12v, 105w. I could accurately test the input voltage, but the output voltage gave results that made little sense & seemed to vary.

Is this because digital multi meters aren't good at this, or are transformers difficult to test?

Next question, is it possible to test the starter on a fluorescent light with a digital multi meter to see if its working?

Would I be better off with a different type of meter?

[Grunff]

The output voltage of a transformer will vary depending on the load. The open-circuit voltage will be a lot higher than the full load voltage. I'd guess the open circuit voltage will be somewhere around 18-20V. If you then measure it again with a 105W load, it'll be around 12V.

Not really - they're made up of a capacitor and a gas-filled bimetallic switch. There are several different failure modes. Easier to just carry a few spares.

No, I think you just need to get to know the limitations of the instrument. For testing transformers, if you have an output voltage, the transformer is ok.

[Andy Wade]

I'd expect the regulation of a 100 VA transformer to be a lot better than that - 15 V off-load, maybe. However it was probably an 'electronic transformer' with either a high-frequency or a DC output that was confusing TMH's meter.

Or stock-up with electronic starters like

throw away the glow starters at every opportunity.

I'd recommend some safer (GS38 compliant) test leads before you go prodding around inside consumer units in the course of work.

[Andrew Gabriel]

If it's an electronic transformer, the output is something like 40kHz AC and probably not a sine wave, so the meter isn't likely to give a reliable reading. Also, it will propably shut down if the load is too low.

No. Well, if it registers a short, then it's dead, but open circuit could still be dead or alive. If it's an electronic starter, I don't know what you'd see with a meter.

Not in either of these cases. They don't lend themselves to testing with a meter. Carry a few spares and test by substitution.

[T i m]

Assuming it was just a transformer (ie, not rectified and / or regulated DC output) then the meter on AC / 20V range should have shown a pretty constant 12V (or whatever).

Neither from my experience?

Pass. I have only ever tested them by putting them in a fitting. ;-)

I don't think so. That meter is fine for most everyday domestic testing though.

All the best ..

T i m

[The Medway Handyman]

Thanks for that link. I'll use those.

You won't find me anywhere near a consumer unit apart from turning off the power. Prodding? No way pedro.

[Dave Plowman (News)]

Was this a true transformer (heavy and probably doughnut shaped) or an electronic one (switch mode power supply - light weight and probably tubular or rectangular)?

The former will probably give quite a high voltage with no load applied - about 16v DC. The latter might show no output at all without load. But both should give near enough 12v AC with the load in place. Of course if they did you'd not need to test them. ;-)

I don't think so. But it's one of those things best tested by substitution.

I would find it rather lacking as I'm used to one which can read min and max as well as average and hold any such reading. Also measures electronic components, frequency and temperature. But cost about 200 quid.

[The Medway Handyman]

It did have 'electronic transformer' writ upon it. My meter was on DC, reading seemed to go up & down.

[Bob Eager]

If it's just a transformer, its output will be AC...

[<me9]

I would expect from a decent transformer that the output regulation from no to full load to be in the region of 5%.

An unregulated DC power supply would be in the region of the figures quoted.

Some "transformers" may be the latter, but the output would be DC.

[The Medway Handyman]

Aha! I assumed, perhaps wrongly, that it would be DC. That may well have caused the confusion.

[Dave Plowman (News)]

The ones I've played with were AC - but high frequency rather than 50Hz, which could give a false reading on some meters.

[The Medway Handyman]

The latter.

In this particular situation, the customer was saying that bulbs kept blowing & lights kept flickering. She couldn't get any of the leckies she called to come out for such a small job, so I agreed to have a look.

I told her I could check the wiring & suggested changing the transformer to eliminate that as a possible cause. House around 10 years old. She was happy with that. It now seems that some of the lights have loose connections with the G4 bulbs, so we are looking at replacing them.

Testing under load with the meter set at 20v AC would seem right then?

[Andrew Gabriel]

Like I said, I would not expect an accurate reading if it's an electronic transformer. For an iron cored transformer, that should be OK.

[meow2222]

was it correctly loaded during test? If not, the test results dont mean anything.

yes, so its not really a transformer. Therefore none of the usual behaviour of transformers applies here.

sounds like you've done exactly what's practical, check wiring, bulb holders, and replace psu.

No. The thing outputs high frequency and likely non-sine ac, so your meter is entirely unsuitable. There are ways to test it with what you've got, but they're not so simple and not worth doing.

Bear in mind that if you ask about a complex technical subject on uk.d-i-y youre going to get a lot of wrong answers.

If you want more info:

[John Rumm]

Electronic "transformers" are usually switched mode power supplies. The output waveform may be not what you expect. The frequency could deviate dramatically from 50Hz. It might be rectified DC (with ripple), or it might be smoothed. The output may turn off altogether without a load.

[Lobster]

What's the benefit of these - efficiency/longevity/long-term cost/all of the above...?

David

[The Wanderer]

You will. You definitely will. One day you'll be tempted to take the lid off and check out something inside. We all do it.

[Owain]

And when the lighting MCB suffers an attack of sparkenflashenpoppen and its little lever goes all floppy, the only way to get the lights back on that weekend is to move the circuit to the another MCB of about the same rating.

Besides, the CU is the easiest place to wire into for a handy battery-backed emergency light over the CU...

Owain

[Harry Bloomfield]

Then it would be pulsing, looking for the load. The only way to test a switch mode power supply (electronic transformer) is to apply a load to it. Even a 10w load would probably be enough to get it to fire up and hold a steady output.