RFXCOM - use cases

[nduku]

Hi,

at the moment I've a WMR928 connected to a Alix box with MeteoHub installed, and it's working pretty well. What I don't like is the limitation of the sensors (e.g. no UV sensor on WMR928), therefore I'm looking for alternatives.

So, I came across the RFXCOM and they're advertising a huge number of supported sensors on their web page which sounds interesting, but when I scan through them I see - apart form Oregon scientific sensors - mainly home security related stuff which I'm not really interested in. More interesting would be energy monitoring, heating system monitoring (e.g. monitor the temperature curve of the buffer storage), ...

What I'm interested in are sample use cases that others already have set up. I'm really curious what interesting setup's out there that might ease my decision to purchase the RFXCOM

...and of course experience with the RFXCOM. Is it reliable? Working fine? Known issues?

Any input is appreciated!

Cheers,
Axel

[wfpost]

I have added an RFXCOM to my existing VP2/meteohub installation two years ago and my experience is positive.
Because additional Thermo sensors are very expensive with Davis I opted for RFXCOM and are quite satisfied.

The Oregon UV-sensor is a different story: I had one connected to my earlier Oregan WMR-100 weatherstation before I got a VP2.
It has to be said that the accuracy and precision of that sensor is very modest to say the least.
The fact that Oregon is not releasing specifications about that sensor is telling a lot IMHO.

So, Thermosensors from Oregon in connection with RFXCOM are doing the job very well.
But from my experience for Solar and UV measurements the only, though not cheapest solution of gadgets, is coming from Davis and Davis only.
At least in the semi-professional area of weather equipment.

BTW:
If someday I had too much money (not the case right now ) I´d go for this weather gadgets of a company located in the area I´m living:
Three times more expensive than Davis, but some very nice stuff like a Clouds Sensor.

http://www.reinhardt-testsystem.de/engl ... ensors.php
http://www.reinhardt-testsystem.de/engl ... WS_9_5.php

[nduku]

Hi,

so you added additional Oregon Thermo sensors to RFXCOM. The Oregon UV sensor experience is bad news, because this would have been one of the first sensors to add for me

Does RFXCOM also support Davis sensors? The web page does not mention them. So There would be no way to use a davis uv sensor without the Vantage, right?

If someday I had too much money (not the case right now

) I´d go for this weather gadgets of a company located in the area I´m living:
Three times more expensive than Davis, but some very nice stuff like a Clouds Sensor.

http://www.reinhardt-testsystem.de/engl ... ensors.php
http://www.reinhardt-testsystem.de/engl ... WS_9_5.php

In fact, looks very interesting, but I think the DWD would then ask you for the weatherforecast

2500 € for a sensor is far away from my weather-enthusiast-budget. BTW: The company is also near my hometown, just 30 km away

[admin]

davis is using a frequency hopping technique the rfxcom cant handle. so davis sensors are out of the equation when it comes to rfxcom.

[wfpost]

as I still have my UV800 from Oregon, I´ve connected it to my current rfxcom.
The problem with the UV800 is that it shows far to high UVI values in the summer.
I recall it had high values of 14 when there was only UVI 10.

What I did then, was using the calibration within meteohub to decrease the UVI values over 9 from the UV800 and that worked more or less.
I´m trying to set up the UV800 as well and maybe I can show you a comparison graph between the Davis and Oregon UV sensor.

[skyewright]

I've had an RFXCOM since I first started using Meteohub (March 2008). I have found it to be very reliable. Signal reception is far better than for OS consoles (I've had both WMR928 & WMR200). I have a mix of sensors from my WMR928 & WMR200, plus extra 3-channel sensors (the Temp only ones with a 3m lead), and a extra 10 channel temp/hygro sensor. Having multiple sensors using the same channel (e.g. I have 6 '3-channel' temp/hum sensors!)works but you have to accept that at intervals pairs of sensors sharing the same channel will conflict (if they have similar signal strengths then both will be lost for a while, or if one is closer then the further one will be lost for a while. One way I deal with this is to give the 3-channel sensor I don't want to drop out at all [a black-body based night-time cloud sensor] a channel to itself, and crowd several less critical sensors onto a different channel).

I've had a UV800 since New Year. As it only has integer resolution it didn't register any non-zero values till some time in February, but accepting that limitation I am happy with it so far.

I also have a Davis so theroetically I could add a UV to that but the cost, even imported direct from the US, would be massively greater than for the OS, and frankly UV isn't that high on my priority list so I decided to give the OS unit a try out (I did import a Davis solar sensor, and I'm very happy with that, but that's only half(?) the price of the UV unit, and solar energy measurement is of greater interest to me).

In addition to the OS sensors I also use the RfxCom with RFXMeter/RFXPulse to monitor electricity import and microgeneration.

[nduku]

Hi,

thanks for the replies so far! So, to recap:

-RFXCOM seems to be relieable
-Davis sensors are not supported
- wide range of OS sensors supported
- OS UV800 has mixed results

use cases I see so far:

Temp sensors
UV sensor
RFXmeter for power consumption monitoring (would be very interesting for me, but 140 € for that sensor is far away from cheap, maybe an alternative could be that OWC stuff from UK or ELV sensors)

One idea I have in my mind, maybe someone has tried the same:
I have a heating system with a buffer storage containing 1000 ltrs of water and I'd like to monitor the temperature of the water in relation to the solar energy monitored with an UV sensor. Result would be an overview to see of how much solar energy goes into the buffer through a solar thermal unit.

Anyone experience with something like that?

[YJB]

Hi,

For monitoring solar energy you will really need a solar sensor, an UV sensor is not suitable for that (as far as I can determine).

The UV Sensor measures the sunburning portion of the UV spectrum. The solar sensor measures solar radiation , which is the portion of the solar spectrum.

BTW,
I'm also a happy RFXcom user (wind, rain, 6 * temperature, electricity and natural gas).

Ysbrand

[skyewright]

RFXmeter for power consumption monitoring (would be very interesting for me, but 140 € for that sensor is far away from cheap, maybe an alternative could be that OWC stuff from UK or ELV sensors)

I have a "CurrentCost" CC128. Meteohub can read direct from that (i.e. RFXCOM is not needed).
The main problem with the CT (i.e. clamp) based electricity monitors is that (except for the Brultech, from the US) they work on fixed assumptions about power factor and voltage and thus are simply not accurate; once you are used to how they work in your situation they give a useful guide but no better than that (I well recall spending ages trying to track down mystery overnight usage until I discovered that the problem was just a matter of power factor and I was trying to find something that didn't really exist!).

The RFXPulse is expensive but accurate (it's counting the LED 'flashes' on your actual utility meter.

Nowadays I only use the CC128 for monitoring electric heating (resistive heating has a consistent power factor of 1, so the power factor problem is less of an issue).

solar energy monitored with an UV sensor.

As YJB says, a UV sensor isn't suitable for that (plus the fact that the OS sensor only reports to integer resolution means it would not be very good for the purpose even if UV was a useful measure in this case).
You'd probably find a "temp in a jar" style solar sensor more useful (the OS temp only 'on a lead' sensors are good for that purpose).

[YJB]

The main problem with the CT (i.e. clamp) based electricity monitors is that (except for the Brultech, from the US) they work on fixed assumptions about power factor and voltage and thus are simply not accurate;

RFXPower is using a clamp as well, but it takes the power factor into consideration. In real life it turns out to be very accurate, I barely see any difference between my utility meter and the RFXpower counter.

[skyewright]

RFXPower is using a clamp as well, but it takes the power factor into consideration. In real life it turns out to be very accurate, I barely see any difference between my utility meter and the RFXpower counter.

Thank you. I didn't realise that, however, given the high standard of all the other RFXCOM bit's, I should not be surprised.

[skyewright]

As YJB says, a UV sensor isn't suitable for that (plus the fact that the OS sensor only reports to integer resolution means it would not be very good for the purpose even if UV was a useful measure in this case).
You'd probably find a "temp in a jar" style solar sensor more useful (the OS temp only 'on a lead' sensors are good for that purpose).

To illustrate this I've attached an image showing solar (the red line), UV (the blue line) and "temp in a jar" (pale blue shaded area) for yesterday.

The "temp in a jar" is actually plotting the difference in temperature between the normal outdoor temp sensor and the temperature inside the "jar".

I've not included the scales, it's the relative shapes of the plots that are of interest rather than the figures measured.

[nduku]

You'd probably find a "temp in a jar" style solar sensor more useful

Just to get that straight:

What you're telling is, that the difference of the regular outdoor temp and a sensor put in a glass will be a very cheap and accurate alternative to expensive and not always accurate solar sensors? And in addition it also seems to be a way for an UV sensor to be connected to the WMR928!

I've not included the scales, it's the relative shapes of the plots that are of interest rather than the figures measured.

Do you have any idea what that means in terms of UV index or solar power? I've to confess that I'm far away from havin' a clue on solar energy measurement.

[nduku]

After reading this

http://www.weather-watch.com/smf/index. ... 506.0.html

I think I'll first go for the homemade solar sensor as I have a spare sensor already that'd be good for this. Thanks for the hint!

To get back to the topic I think I'll purchase the RFXCOM in the near future, as it has some promising features and general feedback is positive. Additionally it seems to be a nice toy

[skyewright]

You'd probably find a "temp in a jar" style solar sensor more useful

What you're telling is, that the difference of the regular outdoor temp and a sensor put in a glass will be a very cheap and accurate alternative to expensive and not always accurate solar sensors?

Cheap: yes
Useful: yes
Accurate: No. Just a useful indication, provided you keep an eye on and regularly adjust the temp difference that equates to "noon clear sky".
WeatherDisplay users can get an estimate of W/m2 from a temp in a jar sensor but the figure is very much an estimate rather than a measure.

I'm not sure where you get the impression that the expensive solar sensors are not accurate.

and in addition it also seems to be a way for an UV sensor to be connected to the WMR928!

I'm not sure where you are seeing that.
I know that someone did a modification that linked a UV sensor through to a temp sensor and use the result for solar, but that isn't "temp in a jar" and it still has the problem that UV isn't a good measure of solar.