Add measurements for the Shelly Wave 1PM Mini Z-wave switch

[schwabe]

Device information

This is the Shelly 1PM Z-wave switch that can be put into a switch box behind a light switch.

Home Assistant Device information

Device info
QMSW-0A1P8
by Shelly Europe Ltd.
Connected via Controller
Firmware: 11.5.0

Checklist

• I have created a single PR per device. When you have multiple submissions please create separate PR's.
• For lights - I have only included the gzipped files (*.gz), not the raw CSV files.
• For lights - I have provided a CSV file per supported color mode. Look that up in Developer Tools -> States

Additional info

This has been measured by a speciality DMM that has no problems with low power measurements and should be be probably better than most of the plugs to measure this even though it is out of calibration.

[bramstroker]

@schwabe Thanks.
I see you used device_type power meter, which is not correct.
Could you please have a look here: https://docs.powercalc.nl/library/device-types/smart-switch/
And pick the example JSON for Smart switch with built-in powermeter.
https://docs.powercalc.nl/library/device-types/smart-switch/#smart-switch-with-built-in-powermeter

[bramstroker]

I'm not sure if these values can be correct.
Going by all the other data available in power profile library.
And also information available on the internet, and manufacturer specs never seen lower values than 0.3W for shelly devices.
I suspect there might be a problem with your multimeter not be calibrated correctly, but it's hard to prove.
So not sure how to proceed yet.

It's Zwave device, so power consumption will be lower than the Wifi counterparts shelly devices.
But I highly doubt it will be as low as 0.043 as we have never seen this for any Zwave device.

[schwabe]

I double checked and the device seems to be behave a bit strange. If you power it on, it stays at 45mW, if you turn it on via the switch, it goes to 168 mW but if you turn it off it again it goes to 128mW and stays there and never goes back to the 45mW. The apparent power (V*A) of the device is also much higher. 0.35 VA instead of the 128mW when off and 0,619 VA instead of 168 mW when turned on.

[bramstroker]

Very strange indeed.
Curious, so I just ordered this device to measure it myself as well.
Will keep you updated to see which measurements I will come up with.
I'll use the Zhurui PR10 and another power meter which should pick up <0.3W which I have laying around.
Let's see which figures I come up with.

[bramstroker]

@schwabe I had this shelly device delivered, and connected in parallel with 2 good energy meters I have laying around.

ELV Energy Master Basic 2
https://www.amazon.nl/ELV-Energy-Master-Basic-energiekostenmeter/dp/B00C7LIVVK/ref=asc_df_B00C7LIVVK/?tag=nlshogostdde-21&linkCode=df0&hvadid=710029310359&hvpos=&hvnetw=g&hvrand=7942757592619482585&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9064801&hvtargid=pla-421355751494&psc=1&mcid=b3d7b7d46e8d3b4981a342e101163315&gad_source=1

Zhurui PR10-c
https://www.amazon.nl/ZHUURUI-PR10-C-metering-recorder-flowmeter/dp/B06VWMS7QS

ELV is only having 0.1W precision, but is able to measure as low as 0.1W.
Zhurui PR10 is able to measure 2 digits precision, and this device is known to be very acurate for a consumer grade device, also on low power levels.

Here are my measurements. Toggled relay on and off 10 times, and measurements were reproducable exact constantly.

ELV Energy Master Basic 2

• off: 0.1W
• on: 0.4 ~ 0.5W (jumped between these two figures)

Zhurui PR10 Pro

• off: 0.14W
• on: 0.46W

Turning off without load meter went from 0.46 to 0.14 within a second, and turning on again to 0.46.

As you can see both my energy meters returned the exact same values and were reproducable.
So I'm pretty sure these figures are correct within a small error margin.

Also added a video: https://youtube.com/shorts/d7sUEDU5pYk?feature=share

[schwabe]

So your off values are about that same I have. I have 0.13 W and you have 0.14W, so that is the same, more or less. The intesting thing is that your on values are quite different from mine.

I will try to see if I can figure out what might cause this.

[bramstroker]

@schwabe were you able to figure it out or do you have any other measure device which you can try?
Otherwise I suggest to go with my figures from the Zhurui PR10 Pro which were reproducable each time and seem very reasonable.

[schwabe]

One thing about your measurement is that it would be outside the spec of the zwave device, which is < 0,3 W. I found one guy comparing your measurment device with the successor of my DMM (https://lygte-info.dk/review/Power%20Zhurui%20Power%20Recorder%20PR10%20UK.html) where they seem to be roughly the same which doesn't help making a decision here.

I have an osciloscope that can also measure power but the current clamp is not really precise for such low currents, so you get a lot of noise instead of a proper measurement:

So that does not really help either. But measuring these low power settings is always challenging. But I personally would trust my multimeter more than the 40 USD China power meter. What I suspect is happening is that the spike is short enough that typical approach for sampling often enough per second to get close to the real measurement gets us some aliasing effects here which results in the different measurements.

The PR10 specifies its accuracy a bit convoluted as:
[3].If power <100w, the resolution is 0.01w, the measuring accuracy is±(0.5FS+0.01w),power>100w, the resolution is 0.1w, the measuring accuracy is± (0.5FS+0.1w),the power> 1000w, the resolution is 1w, the measuring accuracy ±(0.5FS+1w);

But even assuming that FS (full scale) is 100W for the < 100W range, we still have +/- 0.5W + 0.01W for this range. So +/- 0.5 is not big as you approach 100W but for small power measurements, this is a significant uncertainity.

Bonus here, osciloscope shots of the IKEA Inspelling where my values are also a bit lower than the ones in the library (0,204 vs 0,23 Idle and 0,64 W vs 0,8W on) but stil quite close since I had the setup anyway:

Idle:

Power on:

[bramstroker]

You know a lot about the topic. Hard thing to measure these low consumptions.
That's a really cool and expensive osciloscope you have here.
Is it somehow possible to measure the shelly Zwave module with that one?

[schwabe]

Yeah the first picture is with the Shelly. The problem with the osciloscope is that if you look at the current (C3), it is too noisy to give any good measurement in that low current situation, so that osciloscope is nice too look at the graphs and what happens but the absolute results are unfortunately not very good for the purpose of getting hard numbers. The osciloscope adds a lot of power consumption from the noise at these low power levels.

[bramstroker]

I see. You get quite similar results as me however 430mW, when that's taken with the relay on of course.
Not sure how to proceed with this PR, and which values to pick?
Let me know what your suggestion is.

[bramstroker]

Suggested change

	"measure_description": "Manually measured with Gossen Metrawatt 29S (energy speciality DMM)"
	"measure_description": "Manually measured with Gossen Metrawatt 29S (energy speciality DMM)",

Forgot comma here that causes the JSON to be invalid