I just got my electricity meter hooked up to the RFXmeter.
(http://www.domoticaforum.eu/topic.asp?TOPIC_ID=2809).
It seems I get 1200 pulses per 1KW. That seems a lot everyday. Above the disk I choose there is another one with 10x less pulses. Is it better to take that pulse to avoid frequently roll over moments of the pulse counter or is it better to keep the high accuracy?
RFXmeter best practice
Moderator: b_weijenberg
RFXmeter best practice
Hi Bastiaan,
The RFXPulse has a maximum counter value of 16777215. At 1200 pulses per kWh that means a roll over every 13981 kWh. In an average household that would be once every 4 years.
I'm using 2000 pulses/kWh power meters connected through the S0 output to RFXCom pulse interfaces and I find that accuracy is barely enough
. Note that at 2000 pulses/kWh and the minimum transmission interval of the RFXMeter (once every 30 seconds), the "resolution" you get is 60 Watt (i.e. at 60 Watt energy use, you'll see exactly one pulse per 30 seconds).
Below that, the signal will "jitter" between one or less than one pulse per transmission, making it hard to guess whether a device didn't use energy for a long time and suddenly started using +/- 60 Watts or whether a device is steadily using a very small amount of energy that suddenly added up to a pulse.
In your 1200 pulses/kWh case the resolution would be 100 Watt. And 120 pulses/kWh would mean a resolution of 1000 Watt...
In your case, using a single meter for total consumption, this might be less of an issue as idle load will usually be well above 100 Watt. Also: if you only want regular readings (once per hour), it all really doesn't matter that much. It's more the problem of instantaneous power consumption measurements.
See my graph in http://www.domoticaforum.eu/topic.asp?T ... ge=2#10136 where the artefact is visible in the green line (power consumption is steady but the readings jitter).
Lennart
The RFXPulse has a maximum counter value of 16777215. At 1200 pulses per kWh that means a roll over every 13981 kWh. In an average household that would be once every 4 years.
I'm using 2000 pulses/kWh power meters connected through the S0 output to RFXCom pulse interfaces and I find that accuracy is barely enough
. Note that at 2000 pulses/kWh and the minimum transmission interval of the RFXMeter (once every 30 seconds), the "resolution" you get is 60 Watt (i.e. at 60 Watt energy use, you'll see exactly one pulse per 30 seconds).Below that, the signal will "jitter" between one or less than one pulse per transmission, making it hard to guess whether a device didn't use energy for a long time and suddenly started using +/- 60 Watts or whether a device is steadily using a very small amount of energy that suddenly added up to a pulse.
In your 1200 pulses/kWh case the resolution would be 100 Watt. And 120 pulses/kWh would mean a resolution of 1000 Watt...
In your case, using a single meter for total consumption, this might be less of an issue as idle load will usually be well above 100 Watt. Also: if you only want regular readings (once per hour), it all really doesn't matter that much. It's more the problem of instantaneous power consumption measurements.
See my graph in http://www.domoticaforum.eu/topic.asp?T ... ge=2#10136 where the artefact is visible in the green line (power consumption is steady but the readings jitter).
Lennart
RFXmeter best practice
Thanks Lennart,
I use seriously more than average but it still means one roll over every 1-2 years. You are right that resolution also averages out errors.
'll keep the setup as is, it seems to read very consistant. Just need extra RFXmeter.
I use seriously more than average but it still means one roll over every 1-2 years. You are right that resolution also averages out errors.
'll keep the setup as is, it seems to read very consistant. Just need extra RFXmeter.