how to measure wattage?

Measure the resistance of the filament, then use Ohms law to get a rough idea of the wattage. Most cheap multimeters would not be particularly accurate at the level you are measuring, hence you will only get an approximate reading, unless you have a 'better' multimeter.

that will only measure the cold resistance. The lamp wattage is based on the hot resistance.

If they are the same as normal incandescent bulbs the resistance will show as almost nil unless the bulb is running.

The only way I can think of is to have an ammeter in series when they are running. Wattage will be volts multiplied by amps so 20 watts at 12 volts will be about 1.67 amps. Be careful. The start up current when the bulb is first turned on will be a great deal higher.

It'll be even more marked with halogens as they run hotter.

Afraid that won't work. The resistance varies with temperature and is much higher when the lamp is working.

You need to measure current and voltage and multiply the two for watts.

If the old ones are still working, you need a multimeter with a very low resistance range (anything more than 10 ohms full range is too high), to compare the resistances of an old one and a new one. If they're about the same, then they're the same power, if the new ones are roughly half the resistance of the old, then the new ones are higher power. 12V G4 bulbs normally only come as either 10 or 20 watts, except for some "Eco" bulbs which are 14W or so.

If you're still unsure, then dig out the manual....

cooker hood was already in house so never had a manual, however have managed to locate on line and the max wattage is 20W so thats prob whats in already.

It is true that the start up current is higher. I do not think that is a problem. If it was, you could just measure over a longer period.

I have this energy-meter:

Here you see energy measuring by an expert:

The problem is that the start up current might well blow the protection fuse in the meter

.... or blow up the meter - which is why I mentioned it.

Yes, good point, but measurement still stands as an easy way to find out, measure the 2 different lamps, and it can clearly be seen if they are the same wattage.

Would it not be possible to make a cable with a parallel bypass switch which is then opened once the bulb is operating, thus avoiding the higher switch-on current through the ammeter?

Part Timer was thinking very hard :

Possible, but it would need to be a make before break - because the lamp would cool down during the switch over.

Pretty useless doing it on a cold filament

Simples - you short the meter out on startup

switches in both cables, one with meter in series.

open switch on cable with meter and close other switch before connecting power.

Connect power and when bulb lights first close switch in series with meter and then open other one.

Take meter reading in amps and turn it all off.

Multiply amps by 12volts.

Seems a lot of trouble!

On the contrary, my cheap multimeters are (surprisingly) every bit as accurate as my relative expensive Flukes. However, as others have pointed out, measuring the cold resistance of a filament gives little indication of its hot resistance. You are much better off measuring the working AC current, and use 'watts = voltage x current'.

But surely the cold resistance is not going to be the same as when they are on? Brian

Exactly. A switch across the meter. Make sure the switch is "on" before you power up your test rig, and when the lamp glows you turn the switch "off" so that the lamp current flows through the meter. Hopefully your meter has something like a 10A AC range.