|
Post by daveatl on Nov 8, 2020 18:24:19 GMT -8
Hi,
I got my Skyweather up and running, bit I'm getting some strange measurements. First, the outside humidity reads 71.8% while the inside humidity reads 54.5% even though both sensors are right next to each other on my bench. Second, the wind vane, even though it seemed to be giving correct readings the first few times I ran skyweather.py now says WNW when the vane is pointed north and N when pointed any other direction.
Does anybody have any idea what may be going on with this thing?
Thanks, Dave
|
|
|
Post by SDL on Nov 9, 2020 10:50:04 GMT -8
Dave,
I would guess you have the wind vane plugs and rain bucket reversed. Carefully check that.
The humidity thing is more problematic, You can leave them on a few days (reading them if possible) and see if you bake out the difference. It might be that kind of a problem.
BP
|
|
|
Post by daveatl on Nov 27, 2020 9:19:22 GMT -8
Hi, sorry for the delay in replying. I was getting good wind speed readings, but i double checked the plugs anyway and they were in the right jacks. I ran the weather rack test, pointed the vane in different directions and got these readings: When pointed North I got 4.118V and 292.5 degrees NE 2.430V 0.00 degrees E 0.488V
|
|
|
Post by daveatl on Nov 27, 2020 9:28:57 GMT -8
Dang it, don't know what i hit, but i wasn't ready for that to post!
Here's all my measurements:
N 4.118V 292.5 degrees NE 2.430V 0.00 degrees E 0.488V 0.00 degrees SE 0.970V 0.00 degrees S 1.515V 0.00 degrees SW 3.316V 337.5 degrees W 4.975V 0.00 degrees NW 4.657V 270 degrees
The voltages look reasonable based on the datasheet for the weather rack, but i have no idea what is going on with the conversion to degrees. Also, I tried to leave the system up and running to bake the humidity sensors, but it keeps losing the WLAN connection and rebooting and I haven't set it up to restart skyweather.py on reboot yet. No other devices in my house seem to think the WLAN dropped out.
|
|
|
Post by SDL on Nov 28, 2020 9:30:56 GMT -8
This sounds like a Raspberry Pi issue. Losing the WiFi shouldn't necessarily cause a reboot.
Can you open a terminal window and see what it says before it reboots?
Post the results.
Regarding the degrees, what are the results from running testWeatherRack.py ?
Post them please.
BP
|
|
|
Post by daveatl on Dec 6, 2020 15:30:28 GMT -8
Ok, here is the output from testWeatherRack.py while vane was pointed North, East, West and South:
----------------- SDL_Pi_WeatherRack Library WeatherRack Weather Sensors ----------------- Rain Total= 0.00 in Wind Speed= 0.00 MPH max Ever Wind Speed= 0.00 MPH MPH wind_gust= 0.00 MPH max Ever Gust wind_gust= 0.00 MPH (' value = ', 4095.375) (' voltageValue = ', 4.095375) Wind Direction= 292.50 Degrees Wind Direction Voltage= 4.095 V ----------------- ----------------- ---------------------------------------- ----------------- SDL_Pi_WeatherRack Library WeatherRack Weather Sensors ----------------- Rain Total= 0.00 in Wind Speed= 0.00 MPH max Ever Wind Speed= 0.00 MPH MPH wind_gust= 0.00 MPH max Ever Gust wind_gust= 0.00 MPH (' value = ', 486.5625) (' voltageValue = ', 0.4865625) Wind Direction= 0.00 Degrees Wind Direction Voltage= 0.486 V ----------------- ----------------- ---------------------------------------- ----------------- SDL_Pi_WeatherRack Library WeatherRack Weather Sensors ----------------- Rain Total= 0.00 in Wind Speed= 0.00 MPH max Ever Wind Speed= 0.00 MPH MPH wind_gust= 0.00 MPH max Ever Gust wind_gust= 0.00 MPH (' value = ', 1508.4375) (' voltageValue = ', 1.5084375) Wind Direction= 0.00 Degrees Wind Direction Voltage= 1.505 V ----------------- ----------------- ---------------------------------------- ----------------- SDL_Pi_WeatherRack Library WeatherRack Weather Sensors ----------------- Rain Total= 0.00 in Wind Speed= 0.00 MPH max Ever Wind Speed= 0.00 MPH MPH wind_gust= 0.00 MPH max Ever Gust wind_gust= 0.00 MPH (' value = ', 4949.625) (' voltageValue = ', 4.949625) Wind Direction= 0.00 Degrees Wind Direction Voltage= 4.938 V ----------------- -----------------
Catching what happens when the pi reboots may be more difficult, it tends to happen randomly after it's been running for an hour or 2 and the first indication i usually see that it rebooted is when the Blynk app shows something like "WLAN down, pi rebooting" at the bottom.
|
|
|
Post by SDL on Dec 8, 2020 18:50:33 GMT -8
WLAN down means it is losing the WiFi connection. Try disabling the WLAN detection in conflocal.py and see if that fixes things. It does sound like that option is enabled in the software otherwise it wouldn't get to Blynk.
BP
|
|
|
Post by daveatl on Dec 11, 2020 18:54:46 GMT -8
Hi, Thank you! That fixed the rebooting issue. It has stayed up for 24 hours now, instead just 2 or 3 like it was doing.
Any ideas about the wind vane readings? It still thinks it is pointed north no matter which direction I point it. Except when it is actually pointed north, then it thinks its pointed west north west.
Thanks again! Dave
|
|
|
Post by SDL on Dec 12, 2020 8:48:17 GMT -8
|
|
svejk
New Member
Posts: 14
|
Post by svejk on Dec 21, 2020 13:56:55 GMT -8
I have the same problem and I tracked it down to under-voltage on the power supply to the SkyWeather box. I have a long USB run from the transformer to the box. The test setup worked fine, but once installed with a long power supply run it started to behave as you described. I tested my theory this morning by taking a long extension cord outside, plugged the power supply into it, and and plugged the power supply directly Into the pi. All directional readings were suddenly correct! When I put it back on the long USB run, the problem resumed. :-(
If voltages from the WeatherRack are not within 5% of the expected value for the current wind direction then one of two things happens: 1.) It will report an incorrect direction corresponding to a voltage that it finds within the + or - 5% range, or 2.) if there is not a valid voltage within the + or - 5% range to match your measured value then it will report back a default of 0 degrees (North).
I haven't figured out an alternate setup for power to my SkyWeather yet, but at least I now know why wind direction is incorrect.
|
|
|
Post by SDL on Dec 23, 2020 9:09:44 GMT -8
Svejk,
You nailed it. That makes perfect sense to me. A low voltage to the Pi screws with the ADC on the Weather Vane. If you look at the spec, you see that some voltages are close together and require small bands around the directions.
I'm glad our new WeatherRack2 doesn't have this problem.
BP
|
|
svejk
New Member
Posts: 14
|
Post by svejk on Dec 23, 2020 10:21:16 GMT -8
I measured the voltage at the GPIO pin headers. Pins 2 and 4 (5v) measured at 4.75v while pins 1 and 17 (3.3v) measured correctly at 3.3 volts. I'm using a 2.5 amp power supply, however, I'm wondering if the voltage measurements are under load or if they drop when when a wind vane reading is taken. Not sure how to determine that.
I did notice that some voltages are very close together and + or - 5% is a narrow window, so it doesn't take much variance for the readings to be off.
Don't rub in WeatherRack2. I just got my SkyWeather up and running. ;-)
|
|
svejk
New Member
Posts: 14
|
Post by svejk on Dec 24, 2020 9:20:30 GMT -8
I don't understand why ADJUST30R5 is set to "1" in SDL_Pi_WeatherRack.py rather than ".66" as documented for the Raspberry (i.e. it's set to 5v rather than 3.3v.) When I run testWeatherRack.py it sometimes reports wind direction voltages over 3.3v (4.328v, for instance). If the wind vane input voltage is actually 3.3v then no returned value should never exceed 3.3v. This suggests that 5v is actually the input for the weather vane on a raspberry, vs. 3.3v as documented.
To test that question I set ADJUST30R5 to ".95" to account for my input voltage drop (4.75v measured vs. 5.0v expected) and guess what? Wind direction readings are now all correct.
Am I correct that 5v is used for the wind direction on the Pi? If so, then the documentation is incorrect.
|
|
|
Post by SDL on Dec 24, 2020 15:19:19 GMT -8
How are you connecting the Pi and the WeatherRack? If you are connecting it to a 5V interface, (like the Pi2Grover), you use 5V as specified. If you are connecting straight to the Pi, you would use 3.3V.
Seeting it to 0.95 would adjust for your lower voltage to the WeatherVane, I would assume must be lower than 5V for some reason. Can you measure it?
BP
|
|
|
Post by daveatl on Dec 24, 2020 17:22:11 GMT -8
Thank you, svejk! I checked the voltage and measured 5.27V on the GPIO header. I guess that's just too high. I plugged in a different power supply that measured 5.02V on the GPIO header and now I'm getting the wind direction readings i would expect. Thanks again!
Dave
|
|