Using a 34465A for power metering ... ?

[emax]

Hi all,

I wonder if I can use a 34465A for power metering.

The thing is that I have only a consumer-power-meter here which is not capable of measuring a low-power consumption around 1 Watt reasonably. Can I use a 34465A for this, considering the fact that this is the only instrument I have currently available?

And if yes: Any hints how to do this?

(I simply didn't find a way to configure such a setting maybe because I didn't yet master this meter).

[fiddleback]

The 34465A doesn't provide any kind of direct power measurement. However. you should certainly be able to use it to measure power at a one watt level. You'll just have to do it the old-fashioned way. Take a measurement of the voltage drop across the Unit Under Test. Then take a current measurement (in series with the UUT). Multiply the two, and you've got your power consumption. This will work for either AC or DC.

I believe that you can wire the 34465A current probes in series with the UUT, and at the same time, connect the voltage probe across the UUT. Then just switch back and forth between voltage and current measurements, and you can get the two readings in very rapid succession - without changing probes. Take the product of the two, and you've got your power measurement.

[niner_007]

Is this for AC or DC? I don't like the solution proposed by fiddleback, because it will be limited in speed, and will require constant relay switching, which depending on speed of reading would just not work and age the relays in the DMM.

34465A is great but it isn't that fast, depending on how fast you need it to be, it might not work. What is your speed, time, accuracy and resolution requirements? How long do you need to it for, and what is the interval?

Any digitizing DMM does already what you need, by digitizing voltage. The way I would do it, create a little power probe, to convert current into voltage externaly, amplify it, and switch input of the multimeter with a relay, externally between the voltage and the current to voltage input, control all of it via serial or USB via a computer.

[razvan784]

A less widely known feature of the 3446x series is that you can read the individual terms (HI-LO and SHI-SLO) in DCV ratio mode. So you can avoid the relay switching if you can live with an external current shunt, maybe calibrating in periodically with the meter Amps function. Another pitfall is that the sense channel is not differential, so SHI and SLO are sampled at different times. If your current is constant you should be good.
But I have a feeling the OP was talking about AC - in that case you can certainly measure Amps AC RMS, but you can't just multiply that with voltage without having some phase information, which I have no idea how to measure with just a DMM. Maybe you could use digitizing mode to sample the current, while triggering on the voltage zero crossing with an optocoupler on the external trigger input?

[2N3055]

Short answer: no.

Long answer: You need to explain in detail what are you measuring (DC 12V, 240V AC ?) and what kind of power you need: instantaneous, peak, average, for AC power factor, etc... ?
What is the device, is that device running in full power or you are measuring device in standby, but when it wakes up it consumes more....

[emax]

Hello all,

thank you for the replies, they gave me some valuable hints.

For the power-question: It is a 230V AC radio-controlled wall-socket-switch. A relay which I can plug into a wall-socket and turn on and off with a remote control. I want to use this to disconnect my standby-devices from a wall-socket which is hard to reach (behind a sofa).

But, I of course want to know what power consumption this "energy-saver" has. And with simply measuring the current I won't likely get any reasonable answer: My fluke 289  states about 21 mA, which I can't believe: 5 Watts is just nonsense, I guess.

The popular "power-meters" which are used in consumer surroundings are not a solution, so I am thinking about a more precise method.

This is not a critical thing, of course. But I do not want to replace a standby-consumption of maybe 3 Watts (TV, PVR, Amp etc.) with an "energy-saver" that consumes 5 Watts. 


BTW: You all do of course know, that 3 Watts permanent current sums up to 26 kWh/year ...   

[2N3055]

Hello all,

thank you for the replies, they gave me some valuable hints.

For the power-question: It is a 230V AC radio-controlled wall-socket-switch. A relay which I can plug into a wall-socket and turn on and off with a remote control. I want to use this to disconnect my standby-devices from a wall-socket which is hard to reach (behind a sofa).

But, I of course want to know what power consumption this "energy-saver" has. And with simply measuring the current I won't likely get any reasonable answer: My fluke 289  states about 21 mA, which I can't believe: 5 Watts is just nonsense, I guess.

The popular "power-meters" which are used in consumer surroundings are not a solution, so I am thinking about a more precise method.

This is not a critical thing, of course. But I do not want to replace a standby-consumption of maybe 3 Watts (TV, PVR, Amp etc.) with an "energy-saver" that consumes 5 Watts. 


BTW: You all do of course know, that 3 Watts permanent current sums up to 26 kWh/year ...   

You really need a power analyser for this. Your Fluke measured current correctly. Bu current doesn't have to be in phase with voltage, and real power (one that you pay for) might be close to zero, with most being reactive power(that is not metered by electric company).

[Kleinstein]

With some of the cheap power meters could be tweaked to work for lower currents.  Some of the units use a current transformer to measure the current and there is can be possible to change this to something like 10 primary turns instead of just 1. So it would read 10 times the current and power. Ideally one may have to ad an extra fuse, if the wire is thin (e.g. 0.75 mm²).

With an electronic device the lower real power compared to apparent power can be the phase angle, but also a highly distorted waveform.

[Ysjoelfir]

You should definitely think about using a real power meter for that purpose - thats why they are made And designing a proper, precision power analyzer is especially challenging - a whole company is needed to design a good working device Especially measuring AC Power isn't as easy as I*U=P if you need to measure things like SMPS or frequency converters. Even simple measurements of unswitched passive loads on the AC line aren't that easy if you take into account the sometimes pretty nasty deformed mains voltage...
And if you go into three phase measurements,... well, that is even nastyer.

However, to tackle your problem I would advice you to search a cheap, used power analyzer for lab use on ebay. If you are happy with a more or less rough estimate you could also take a look at - for example - the gossen wattavi. I recently sold some of those for 20€ each, so they may be worth looking for.

[emax]

> Your Fluke measured current correctly.

I have no doubt about this. :-) But I thought eactly what you told as well:

> Bu current doesn't have to be in phase with voltage

For that reason, a scope measurement might be the better choice than a 34465A.

[emax]

You should definitely think about using a real power meter for that purpose - thats why they are made

Yes, I agree. But I guess, for the price I'll never burn the power we are talking about. I only know very expensive solutions. But if you know some cheapo solution which does the job, please let me know. 

I have seen some arduino projects which take care of the phase, maybe I'll build one.

And designing a proper, precision power analyzer is especially challenging

Well, it's finally not about certified values, only just better than this "home-energy-saver".


PS: My proposal for an eevblog-video: "Arduino Power Metering, worth thinking about it?"

[MiDi]

There are cheap and good energy check meters for max 20Eur.
I have a Conrad Energy Monitor 3000 and a Profitec KD 302, both measure precise enough down to 1W and below.
The KD 302 was 10-15Eur, but both are not awailable anymore.

Look for tests of current energy meters tested at low power and pick a reasonable one.
The good ones have a special chip for this purpose - e.g. KD 302 has one and was cheap.

To give some rough numbers: the cheap rf wall-switches take ~0.7W (0.5-1W) - this is ~2Eur/year.
Most modern devices these days take less power in standby than these switches.

[emax]

Hello Midi,

I know the article from heise. As you write, the KD302 has disappeared, and so I decided to buy this unit from ELV. It seems to be quite precise in low-power measurements, though it is far from being really "precise". But maybe i't good enough. I didn't yet receive the order and so I still can't give a verdict.

I have seen another device here. Lots of eye-candy but only 0.5% accuracy (from whatsoever ...). I am still considering ...

[emax]

I have just found an eevblog thread for the PR10 here. Seems good.