Meters disagreeing

[bxcfilm]

I have had solar PV panels installed, and the installation was complete on October 21st. The panels are fixed on the roof, and I export the surplus. There is a generation meter, but no export meter.

The generation meter is a Landis & Gyr 5235A, and I also fitted one of these http://www.energymonitorworld.com/prod2.cfm?product=71698

On 7th November, the readings were:

Generation meter: 50 units
OWL meter: 51 units

Today the readings are:

Generation meter: 57 units
OWL meter: 63.6 units

So why is there an increasing discrepancy between the two? The weather has been a lot gloomier in the last 2 weeks, so the units per day figure is well down, but that ought to affect both meters the same (?)

I downloaded the L&G manual and found this. I'm no electrician, so I could do with a translation, but it seems to me to be relevant.
----------------------------------------------------------------------------------
3.2 Anti-creep
Below starting current, the meter enters into an anti-creep mode. In this
state the metrology LED is permanently lit and the registers do not
increment. The LED remains lit until the meter current is increased in a
forward direction beyond the starting current. The starting current is shown in the technical summary.
----------------------------------------------------------------------------------

The starting current is 0.4% of lb (whatever that means).

Is the L&G meter not recording all the current?
Is it "Working As Intended"?

I would appreciate any input.

 

[DetlefSchmitz]

The OWL (assuming it hasn't changed recently) only uses a current clamp, and doesn't monitor the voltage at the same time. So it has no awareness of power factor and therefore cannot give you an accurate reading. They are probably not very closely calibrated either. They are great for a general idea of household power consumption but not precision instruments.

 

[bxcfilm]

The OWL (assuming it hasn't changed recently) only uses a current clamp, and doesn't monitor the voltage at the same time. So it has no awareness of power factor and therefore cannot give you an accurate reading. They are probably not very closely calibrated either. They are great for a general idea of household power consumption but not precision instruments.

I'm aware the OWL is not necessarily 100% accurate, but its reading is already 10% ahead of the L&G meter. Should that degree of difference be acceptable?

Also, I have noticed often how the OWL meter is showing current, which the L&G meter says there is none. Is the anti-creep mode the reason for this?

 

[JohnW2]

The OWL (assuming it hasn't changed recently) only uses a current clamp, and doesn't monitor the voltage at the same time. So it has no awareness of power factor and therefore cannot give you an accurate reading.

True, but I think you're getting unnecessarily sophisticated - since it doesn't monitor voltage, it will not be able to measure kWh in the face of varying supply voltage, even if the PF were always 1.000. Furthermore, one wonders if it's even been configured with the 'most usual' supply voltage for the premises, or whether it is using some default (probably 230V or 240V) for it's kWh calculation.

They are probably not very closely calibrated either. They are great for a general idea of household power consumption but not precision instruments.

Indeed, and that's probably the most likely explanation. Indeed, even expensive and sophisticated meters using 'clamp' sensing are never going to be 100% accurate or reliable.

Kind Regards, John.

 

[Space cat]

-- it has no awareness of power factor and therefore cannot give you an accurate reading.

This power factor thing is a bit technical for non-electricians so I'll try to keep it simple. If you plug in a pure resistive load, like a filament light bulb, the current is in phase with the voltage. That is to say, when the voltage is positive, so is the current and likewise when the voltage goes negative.

But not all loads are resistive. Motors and transformers are partly inductive and their current lags behind the voltage. When the voltage switches from positive to negative, the current remains positive for up to half a cycle before it turns around. Similarly, when the voltage goes back to positive, the current doesn't reverse straight away.

Whenever the current and voltage are opposite, the load is actually pushing power back into the supply. A true power meter takes account of this. A simple current meter doesn't; it measures the current and multiplies by the voltage (which it thinks it knows) but is oblivious to the how much of that power actually goes into the load. Since many loads around a typical house are partially inductive, you should expect your current meter to have the higher reading.

 

[JohnW2]

This power factor thing is a bit technical for non-electricians so I'll try to keep it simple.

Indeed, but as I've just written, I don't think it's even particularly relevant. The facts that the OWL has no awareness of voltage variations and that it may not even have been set up for the 'usual' supply voltage of the premises are simple concepts for a non-electrician and, in a domestic environment, probably far more important than considerations of power factor!

Kind Regards, John.

 

[JohnD]

if the owl is clamped round the supply cable from the PV inverter, it will also be monitoring the current from the mains supply up to the inverter, such as it is. I expect if you look at the owl during the hours of darkness you will see a number.

 

[stillp]

The manufacturer claims the OWL to have an accuracy better than 5% above 3A, better than 10% between 1 and 3A, and "not specified" below 1A. It seems quite reasonable that a calibrated meter could read 10% different from the OWL, particularly since they don't state the conditions under which that accuracy is achieved.

 

[JohnW2]

The manufacturer claims the OWL to have an accuracy better than 5% above 3A, better than 10% between 1 and 3A, and "not specified" below 1A.

Maybe - but that must (or should!!) come with some small print that says that it assumes no variation in supply voltage AND that the device has been configured to calculate on the basis of an appropriate voltage. Let's face it, the supply voltage can theoretically be anywhere within a window something like 17% wide (wrt the bottom) and can also theoretically vary throughout that window in real time. Whilst extreme variations and deviations from nominal are probably rare, it is very reasonable to think that such variations could easily be big enough to totally invalidate the claimed accuracy figures.

It seems quite reasonable that a calibrated meter could read 10% different from the OWL, particularly since they don't state the conditions under which that accuracy is achieved.

Very much so.

Kind Regards, John.

 

[bhm1712]

hence why it is called a monitor not a meter.......

 

[DetlefSchmitz]

You're quite right, JohnW2. The reason I started with power factor is that most complaints I've heard concerning the OWL is in relation to computer equipment, where switched mode supplies can easily produce misleading results. However in terms of not monitoring the voltage; if the OWL assumed 230V then it would already be 8% off where I live. So it does make accuracy claims rather doubtful.

 

[JohnW2]

hence why it is called a monitor not a meter.......

Well, to be fair, 'monitor' usually implies some sort of measurement device which measures continuously, or undertakes a continuous series of measurements, rather than necessarily implying anything about accuracy (some 'monitoring devices', particularly in safety-critical areas, can be extremely accurate) ....

... but it does indicate why a device (of whatever name) which only measures current (with a measurement error) can never provide accurate kWh figures - unless the exact voltage is known, and known not to to vary!

Kind Regards, John.

 

[JohnW2]

You're quite right, JohnW2. The reason I started with power factor is that most complaints I've heard concerning the OWL is in relation to computer equipment, where switched mode supplies can easily produce misleading results.

... and that's one which is probably even more difficult to explain to a non-electrician, because I suspect the issue is at least (probably more) to do with waveform modification than to power factor in the normal sense! (I suspect most of the devices essentially measure peak current and then adjust it to get RMS, using a correction factor which will be wrong if its not a pure sine wave).

However in terms of not monitoring the voltage; if the OWL assumed 230V then it would already be 8% off where I live. So it does make accuracy claims rather doubtful.

Indeed. Even if the device has a facility to allow one's 'usual' voltage to be configured (I don't know if the OWL ones do), most premises will experience at least a few percent variation (sometimes a lot more) over time. If the voltage is not configurable, I would hope that ones sold for the UK market assume something like 240V - which is closer to what most people usually get than is 230V.

Kind Regards, John.

 

[ban-all-sheds]

This power factor thing is a bit technical for non-electricians so I'll try to keep it simple.

Was there any need to descend to idiot mode with all those smilies though?

 

[securespark]

I expect if you look at the owl during the hours of darkness you will see a number.

Two-wit - two-woo?