weathercloud.net data poster

[Kenneth Baker]

Hi,

I started working on a weathercloud.net rest plugin to weewx based in large part on the great plugins Matthew Wall has written along with documentation that weathercloud provided me.

There are a few values that weathercloud would like posted that I am not entirely sure the best way to go about producing.

1) wspd and wdir are supposed to be current values.  I believe these should be the instantaneous values from the loop packets.  Does that make sense?

2) wspdavg and wdiravg are supposed to be 10-min average values and wspdhi is supposed to be a 10-min maximum gust of wind.  Since the archive interval is adjustable (and mine is set to 5-minutes).  Can you suggest how I should compute 10-minute averages?  Should I compute the average (and max) based on the loop packets that come in over the 10-min interval as well?

Thanks!

-- Ken

[Thomas Keffer]

​Hi, Ken

1. Are they expecting a "Rapidfire" like feed? That is, an update very few seconds? If so, then wspd and wdir are best taken from the LOOP data. Take a look at the Weather Underground uploader: it does this. If they don't expect such a rapid update, then I assume they want the "last" value, meaning the last archive record.

2. Best to calculate the 10 minute aggregates from the archive database. Don't use LOOP packets: that's what the rest of weewx is for! Average and max wind speed will be easy, but average wind direction will require that you resolve the wind into x and y components, then do the vector average. You can use function archive.Archive.getSqlVectorsExtended() to do this. Its doc string explains what it returns, but it's basically two vectors: one with the timestamps, the other a vector of complex (x,y) numbers. You would then calculate the average x and the average y, that is average east-west wind velocity and average north-south wind velocity. Then use atan2() to calculate the average direction. Then convert to compass direction. Piece of cake! :-) It's real easy to make a mistake doing this, so test the hell out of it! 

Thanks for giving this a try!

-tk

[Lucas Heijst]

On Tuesday, 30 September 2014 19:21:03 UTC-3, Kenneth Baker wrote:

I started working on a weathercloud.net rest plugin to weewx 

Hi Ken,

Good news! Weathercloud may be an alternative for Weather Underground (with it's faulty interface for missed data).

Weathercloud is fast growing and looks very promising. Please let us know when your plugin is ready for testing.

Luc

(http://app.weathercloud.net/d0959484454)

[Robin Sanders]

Kenneth,

I am also keen to give this a try as soon as you are ready for testing.

Robin

[mwall]

hi ken,

i hope i'm not stepping on your toes with this, but here is an implementation for a weather cloud uploader.  before the weewx 3 release i am doing an update to all of the uploaders on the weewx wiki, and this was some partially written code in the larder.

this will upload the basic stuff with correct units, and it seems to handle errors properly.  however, there are still a few things not complete:

- no wind averaging
- does not handle extra temperature, humidity, and wind sensors
- does not calculate thw
- does not upload barometric trend
- does not upload battery voltage or status
- does not calculate interior dewpoint or heat index
- does not handle soil moisture or leaf wetness
- does not handle forecast icons

some of these are still not supported by weathercloud, but they are in the weathercloud api.

you must login at weathercloud then create a device.  choose 'vantage pro' for the device, and 'module for weatherlink' as the link type.  then you'll get an id and key for the device and you can put those into the WeatherCloud section in weewx.conf (see comments in wcloud.py for syntax).

imho, the cloud process for setting up devices is a bit brittle.  they really do not need to worry so much about the station type as much as the data coming from it.  for example, even if i specify a vantage station, there is no telling what sensors i might have for it, and i could be uploading data from non-davis sensors as well (e.g., from a frankenmix of rf, davis, and one-wire sensors).

anyway, let me know how it works for you.

m

[Lucas Heijst]

Matthew,

Can you give us installation instructions?

Where to put wcloud.py / settings in weewx.conf?

I already have a Weathercloud ID and key

Luc

[mwall]

On Thursday, October 16, 2014 8:09:32 AM UTC-4, Lucas Heijst wrote:

Can you give us installation instructions?

Where to put wcloud.py / settings in weewx.conf?

0) obtain credentials from weathercloud.net

1) put wcloud.py in the weewx user directory

cp ~/Downloads/wcloud-0.1rc1.py /home/weewx/bin/user/wcloud.py

2) modify weewx.conf:

[StdRESTful]                                                                   
    [[WeatherCloud]]                                                           
        id = WEATHERCLOUD_ID                                                   
        key = WEATHERCLOUD_KEY                                                 

[Engines]
    [[WxEngine]]
        restful_services = ..., user.wcloud.WeatherCloud

3) restart weewx

sudo /etc/init.d/weewx stop
sudo /etc/init.d/weewx stop

[Lucas Heijst]

Matthew,

Most values in the database are zero or N/A (bar)

In the protocol I used earlier most values had to be integers of (real value*10)

Example:

temp 34.3 Celsius should have value 343 in the message

bar 1010.7 hPa should have value 10107 in the message

As a test  I used for all fields format '%.0f' for the data of 15:50. My temp of 34 Celsius was interpreted as 3.4.

10/16/2014 15:50	3.4	3.2	3.2	2.6	4.1	64	69	-	0	0
10/16/2014 15:30	0	0	0	0	0	62	68	-	0	0

The message of the original version of wcloud.py gives back code 429

The message of my protocol gave back code 200

Example of your protocol:

http://api.weathercloud.net/v01/set?wid=xxxxxxxxxxxxxxxx&ver=270&tempin=33.7&temp=32.5&wspd=0.80&wdir=19&hum=67&rainrate=0.00&humin=61&dew=25.6&wspdhi=1.80&heat=40.9&rain=0.00&key=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx&time=1525&bar=1010.72&type=251&chill=32.5

429

Example of my protocol (string typed in by hand):

http://api.weathercloud.net/set/wid/xxxxxxxxxxxxxxxx/key/xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/temp/326/tempin/342/chill/323/heat/394/hum/65/humin/59/wspd/03/wspdhi/06/wspdavg/04/wdir/000/wdir

avg/000/bar/10099/rain/0/ver/1.5/type/201/time=1615

200

Luc

[Lucas Heijst]

Matthew,

You can check my data at WeatherCloud via the following links.

First device (my protocol):

Weerstat. Paramaribo 1

Weather Station X

Other

http://app.weathercloud.net/d0959484454

Second device (your protocol):

Vantage Pro Plus

Module for Weatherlink

http://app.weathercloud.net/d2128164160

Luc

[mwall]

hey luc,

thanks for testing this. 

attached is 0.2rc1.  i have fixed the multipliers and added support for many more fields.  the remaining unimplemented fields are now:

thw (i do not know what this is)
tempagroXX (i do not know how weathercloud defines these)
wspdXX (only useful on weather station with multiple anemometers)
wspdavgXX (only useful on weather station with multiple anemometers)
wspdhiXX (only useful on weather station with multiple anemometers)
bartrend (it is not clear what -60,-20,0,20,60 mean)
forecast (not yet defined by weathercloud)
forecasticon (not yet defined by weathercloud)

this version does a mapping between the fields in the default weewx database schema and the fields defined by weathercloud.

m

[Lucas Heijst]

On Thursday, 16 October 2014 18:13:21 UTC-3, mwall wrote:

thw (i do not know what this is)

Seen on wxforum.net: 

“THW Index” or Temperature-Humidity-Wind Index. This value indicates the “apparent” temperature in the shade due to these factors. 

Luc

[Lucas Heijst]

On Thursday, 16 October 2014 18:13:21 UTC-3, mwall wrote:

attached is 0.2rc1. 

Matthew,

The message to WeatherCloud now gives back code 200.

I have implemented a third device a Vantage Pro2 Plus:

Weerstat. Paramaribo 3

http://app.weathercloud.net/d1608037470

Note: the values of Wspdavg seem too big, see table (I have set the presentation to km/h instead of m/s).

10-min average windspeed is about 1 km/h

10-min average gust speed is about 6 km/h

Luc

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )	Humin ( % )	Hum ( % )	Wspdhi ( km/h )	Wspdavg ( km/h )	Bar ( hPa )	Rain ( mm )
10/16/2014 19:45	31.1	28.2	28.2	24.8	24.2	37.6	32.5	69	79	6.5	6.8	1012.3	0
10/16/2014 19:30	31.2	28.3	28.3	24.8	24.3	37.8	32.8	69	79	6.5	9.4	1012.1	0
10/16/2014 19:20	31.3	28.4	28.4	24.9	24.2	38.1	32.8	69	78	4.3	11.5	1012	0

[mwall]

luc,

a code of 200 means everything is ok. the wspdavg values are off because i forgot to do a units conversion. please try 0.2rc3

m

[Lucas Heijst]

On Thursday, 16 October 2014 21:28:14 UTC-3, mwall wrote:

 please try 0.2rc3

Matthew,

Your Apparent temp (Thw) is 28.3/27.2 Celsius, while the Apparent temp calculation of weewx-wd is much higher: 33.0/32.7 Celsius.

Don't know which calculation is good.

Luc

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )

Thw ( ºC )

Humin ( % )	Hum ( % )

Wspdhi ( m/s )	Wspdavg ( m/s )

Bar ( hPa )	Rain ( mm )

10/16/2014 21:45	30.1	27.4	27.4	24.5	24.1	35.7	30.9	27.2	72	82	1.8	0.8	1013.3	0
10/16/2014 21:35	30.1	27.5	27.5	24.5	24.1	35.8	31.1	28.3	72	82	1.2	1	1013.3	0
10/16/2014 21:05	30.5	27.7	27.7	24.6	24.2	36.4	31.5	0	71	81	4.3	5.1	1013.7	0
10/16/2014 20:55	30.6	27.8	27.8	24.5	24.2	36.5	31.7	0	70	81	2.4	3.9	1013.6	0
10/16/2014 20:45	30.7	27.9	27.9	24.6	24.3	36.8	31.9	0	70	81	3.7	6	1013.3	0
10/16/2014 20:35	30.9	28	28	24.8	24.4	37.3	32.1	0	70	81	3.1	3.3	1013	0
10/16/2014 20:25	31	28.2	28.2	24.9	24.4	37.6	32.7	0	70	80	0.6	0.9	1012.9	0
10/16/2014 20:15	31.1	28.3	28.3	24.9	24.4	37.8	32.8	0	70	80	1.2	1	1012.9	0
10/16/2014 20:05	31.2	28.4	28.4	25	24.4	38	32.9	0	70	79	0.6	0.6	1012.8	0
10/16/2014 19:55	31.2	28.3	28.3	25	24.5	38	32.9	0	70	80	0.6	1.9	1012.5	0

10/16/2014 19:45

31.1

28.2

28.2

24.8

24.2

37.6

32.5

0	69	79	1.8	1.9	1012.3	0

10/16/2014 19:30

31.2

28.3

28.3

24.8

24.3

37.8

32.8

0	69	79	1.8	2.6	1012.1	0

10/16/2014 19:20

31.3

28.4

28.4

24.9

24.2

38.1

32.8

0

69

78

1.2

3.2

1012

0

Displaying 1-13 of 13 results in 1 page

[gjr80]

I think we are probably talking about apples and oranges (maybe oranges and lemons!). I understood THW to be Temperature-Humidity-Wind, one of a number of indexes that quantify apparent temperature. Weewx-WD calculates apparent temperature as per the Australian Bureau of Meteorology (BoM) site (http://www.bom.gov.au/info/thermal_stress) using the non-solar formula. I understand that some Davis stations produce a THSW index that also rolls in the solar conditions but I believe their formula is proprietary and not publicly released. BoM also has an apparent temperature formula that includes solar effects, though I am lead to believe the solar radiation value that should be used is not a straight out (instantaneous) value from a solar radiation sensor. When I have looked for THW and THSW info/formulae there appears to be little info around and nothing that I found that I would really feel comfortable hanging my hat on.

Sorry if I have hijacked this thread. :)

Gary

[Lucas Heijst]

Thanks Gary,

I found two pegs to hang your hat on.

First a definition of THW index on ehow: http://www.ehow.com/how_12028315_calculate-thw-index.html

Secondly an interesting Dutch article about weather definitions: http://www.sysanalyser.com/actueel/De%20gevoelstemperatuur.pdf

Jef Simmelink writes in this article about THW index:

4.3. THW en TSHW index (Davis) 

Dit zijn 2 indices die voor zover ik heb kunnen ontdekken alleen door Davis gebruikt worden in hun elektronische weerstations en/of het software pakket daarbij (Weatherlink). Davis stelt dat de THW en met name de TSHW universele indices zijn welke het gehele jaar bruikbaar zijn. Ondanks herhaalde verzoeken, weigert Davis vooralsnog om de achterliggende formules openbaar te maken. De THW index is gebaseerd op dezelfde variabelen als de NET en de Steadman ëshadeí, echter ze geven alle een verschillende waarde. In de winter volgt de THW index nagenoeg ¥perfect¥ de wind chill en in de zomer vrij tot zeer goed de hitte index. Zie de figuren in 6. 

Translated with Google:

4.3. THW and TSHW index ( Davis)

These two indices that as far as I can discover only be used by Davis in their electronic weather stations and / or software package thereto ( WeatherLink ) . Davis suggests that the THW and , in particular, the universal TSHW indices which can be used throughout the year. Despite repeated requests , Davis still refuses to make . Underlying formulas public The THW index is based on the same variables as the NET and Steadman ëshadeí , but they all give a different value . In winter, the THW Index follows almost perfectly ¥ ¥ wind chill and in the summer rather to very good heat index . See the figures in 6 .

The THW index variable of WeatherCloiud refers to the Davis Vantage Pro2 Plus, so the most accurate definition for THW index might be a choice of wind chill and heat index for a switching point based upon temperature (and wind/humidity?)

Luc

[Kenneth Baker]

> hi ken,
>
> i hope i'm not stepping on your toes with this, but here is an
> implementation for a weather cloud uploader. before the weewx 3 release i
> am doing an update to all of the uploaders on the weewx wiki, and this was
> some partially written code in the larder.

Hi mwall,

No toes stepped on, I've been away for a few weeks. I'll catch up and have a look
at your uploader. Thanks!

-- Ken

[mwall]

here is the thw calculation i used in 0.2rc3:

http://physics.stackexchange.com/questions/32857/how-to-calculate-temperature-humidity-wind-index

notice that the units are not consistent:

thw_F = heatindex_F - (1.072 * windspeed_mph)

i can understand dropping insignificant terms to simplify a calculation, but the units need to make sense...

m

[gjr80]

Yes Luc had come across those before. Always tend to take answers from sites that have user contributed 'answers to everything' with a grain of salt unless I can find some other verification. Also had seeen similar comments re the Davis THW/THSW, bottom line is that Davis seems have kept their methods to themselves.

Out of interest I looked up 10 sites that were reporting THW and applied the THW formula from physics.stackexchange with some pretty average results, the % variation from the THW value on the site concerned and the calculated one was from 0 to 19% with most around 7 to 8%. As wind speed increased the variation blew out so I am not sure the formula is as simple as portrayed. Perhaps one day when I have nothing better to do I will try to make sense of some of the Steadman documents, they seem to be the authority on (small a) apparent temperatures.

As it turns out the Saratoga weather templates don't use THW they are fed THSW (though am yet to find where it is used/displayed) so THW is no real issue for me. I note some of the Davis stations can report THSW but it is in the LOOP2 packet not the LOOP packet that Weewx uses. Perhaps something to think about later. But then again how many apparent temperatures do you really need.... almost a bit like wind stats :)

Matthew, I agree about the units. The engineer in me says not only should your numbers balance on both sides of an equation but also the units. However, I note in a number of indices this seems to go out the door.

Gary

[mwall]

wcloud v0.4 is now available on the wiki.

thw is disabled until we figure out the correct algorithm.  this release also fixes the timestamp (utc) and makes the wspdhi more correct for the weathercloud specification.

m

[Lucas Heijst]

On Tuesday, 21 October 2014 11:57:31 UTC-3, mwall wrote:

 also fixes the timestamp (utc)

m

Matthew,

At 12:34 local time (GMT-3) I switched from 0.3 to v0.4. I noticed the time parameter in the messages switched from local time (1430) to UTC (1730).

Because of this change you would expect a gap in the data base, because the time in the message switched from 12:45 to 15:45.

But no time gap, see below. You will notice the null values for Thw starting at 12:45.

The graphs show time stamps in UTC. There is no change in behaviour between version 0.3 and 0.4

See also:

Weerstat. Paramaribo 3 

http://app.weathercloud.net/d1608037470

Weerstation Paramaribo (new since today)

http://app.weathercloud.net/d1809683123

Luc

PS. The database for "Weerstation Paramaribo" now show rain rate; I don't understand why this isn't shown for "Weerstat. Paramaribo 3". AFAIK the settings are the same.

Weerstat. Paramaribo 3

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )	Thw ( ºC )	Humin ( % )	Hum ( % )

Wspdhi ( km/h )	Wspdavg ( km/h )

Bar ( hPa )	Rain ( mm )

10/21/2014 18:20	31.9	29.3	29.3	24.1	23.5	37.8	33.6	0	64	71	17.6	4.3	1010.1	0
10/21/2014 18:10	32.1	29.8	29.8	24.6	24.4	38.8	35.2	0	65	73	2.2	1.1	1010	0
10/21/2014 18:00

32.2

29.9

29.9

24.5

24.3

38.9	35.3	0	64	72	8.6	0.7	1010.1	0
10/21/2014 17:50	32.3	30	30	24.6	24.4	39.1	35.5	0	64	72	4.3	1.1	1010	0
10/21/2014 17:40	32.4	30.2	30.2	24.5	24.3	39.1	35.7	0	63	71	4.3	1.4	1009.9	0
10/21/2014 17:30	32.6	30.2	30.2	24.6	24.2	39.5	35.7	0	63	70	4.3	1.8	1009.7	0
10/21/2014 17:20	32.8	30.4	30.4	24.5	24.2	39.6	35.9	0	62	70	4.3	1.8	1009.6	0
10/21/2014 17:10	32.8	30.4	30.4	24.6	24.1	39.8	35.8	0	62	69	6.5	2.2	1009.5	0
10/21/2014 17:00	33	30.6	30.6	24.5	24.1	39.9	36.1	0	61	68	8.6	2.2	1009.1	0
10/21/2014 16:50	33.2	31	31	24.4	24.2	40.1	36.8	0	60	67	4.3	1.8	1009.1	0
10/21/2014 16:40	33.5	31.5	31.5	24.7	24.3	40.9	37.5	0	60	66	4.3	2.2	1009.4	0
10/21/2014 16:30	33.8	31.8	31.8	24.8	24.4	41.4	38.2	0	60	65	4.3	2.2	1009.4	0
10/21/2014 16:20	33.9	32.1	32.1	24.8	24.4	41.5	38.7	0	59	64	4.3	2.2	1009.1	0
10/21/2014 16:10	34	32.8	32.8	24.8	24.7	41.8	40	0	59	63	6.5	2.2	1008.9	0
10/21/2014 16:00	34	32.8	32.8	24.8	24.9	41.8	40.2	0	59	63	8.6	2.2	1009	0
10/21/2014 15:50	34	33	33	24.8	24.9	41.8	40.4	0	59	63	6.5	1.8	1009	0
10/21/2014 15:40	34	32.9	32.9	24.9	24.7	41.9	40	0	59	62	6.5	2.2	1009.1	0
10/21/2014 15:30	34	33.3	33.3	25.1	25	42.1	41	0	60	62	4.3	2.2	1009.1	0
10/21/2014 15:20	33.9	33.4	33.4	25.1	25.1	42	41.3	0	60	62	8.6	2.9	1009.2	0
10/21/2014 15:10	33.9	33.4	33.4	25	24.8	41.8	41	0	60	61	6.5	2.2	1009.4	0
10/21/2014 15:00	33.8	33.4	33.4	24.9	24.8	41.6	40.9	0	60	61	8.6	2.2	1009.5	0
10/21/2014 14:50	33.8	33.6	33.6	24.9	25.2	41.6	41.8	0	60	62	6.5	2.5	1009.7	0
10/21/2014 14:40	33.6	33.6	33.6	25	25.3	41.4	41.8	0	61	62	6.5	2.5	1009.8	0
10/21/2014 14:30	33.5	33.5	33.5	25	25	41.3	41.3	0	61	61	6.5	2.2	1010	0
10/21/2014 14:15	33.3	33.4	33.4	24.8	25.2	40.8	41.5	0	61	62	4.3	1.8	1010.2	0
10/21/2014 14:05	33.2	33.2	33.2	24.6	25	40.4	40.9	0	61	62	4.3	1.8	1010.3	0
10/21/2014 13:55	33.1	33.1	33.1	24.6	24.9	40.2	40.6	0	61	63	8.6	2.2	1010.4	0
10/21/2014 13:45	33	33	33	24.8	25.2	40.3	41	0	62	64	6.5	2.5	1010.5	0
10/21/2014 13:35	32.8	32.1	32.1	24.8	25.2	40.1	39.7	0	63	67	4.3	1.1	1010.7	0
10/21/2014 13:25	32.9	31.6	31.6	25.2	24.7	40.7	38.3	0	64	67	4.3	1.4	1011	0
10/21/2014 13:15	33	32.2	32.2	25.2	25	40.9	39.6	0	64	66	4.3	1.4	1011	0
10/21/2014 13:05	33	32.5	32.5	25.3	25.3	41	40.3	0	64	66	4.3	1.8	1011	0
10/21/2014 12:55	32.9	32.8	32.8	25.2	25.3	40.7	40.8	0	64	65	8.6	2.2	1011.2	0
10/21/2014 12:45	32.7	32.8	32.8	25.1	25.5	40.4	41	0	64	66	8.6	2.2	1011.3	0
10/21/2014 12:30	32.4	32	32	24.8	25.2	39.6	39.5	37.4	64	68	4.3	1.4	1011.6	0
10/21/2014 12:20	32.4	31.8	31.8	24.7	25	39.4	38.9	37.9	64	67	2.2	0.7	1011.8	0
10/21/2014 12:10	32.4	32.2	32.2	24.7	25.1	39.3	39.5	38.3	64	66	2.2	1.1	1011.9	0
10/21/2014 12:00	32.2	32.2	32.2	24.5	25.2	38.9	39.9	37.8	64	67	4.3	1.4	1012	0
10/21/2014 11:50	32.2	31.6	31.6	24.5	24.7	38.9	38.2	35.8	64	67	4.3	1.1	1012.3	0
10/21/2014 11:40	32.2	31.8	31.8	24.5	24.9	38.9	38.8	37.7	64	67	4.3	1.1	1012.4	0
10/21/2014 11:30	32.2	32.2	32.2	24.6	25.2	39	39.7	38.4	64	67	2.2	1.1	1012.5	0
10/21/2014 11:20	32.1	32.3	32.3	24.7	25	38.9	39.6	37.3	65	65	4.3	1.4	1012.6	0
10/21/2014 11:10	32	32.1	32.1	24.6	25	38.7	39.4	37.4	65	66	4.3	2.2	1012.7	0
10/21/2014 11:00	32	32.1	32.1	24.5	25.2	38.6	39.7	34.8	65	67	8.6	3.2	1012.8	0
10/21/2014 10:50	31.8	31.4	31.3	24.6	24.6	38.5	37.7	32.7	66	67	8.6	3.2	1013	0
10/21/2014 10:40	31.8	31.3	31.3	24.6	24.5	38.5	37.7	36.5	66	67	2.2	1.1	1013.1	0
10/21/2014 10:30	31.8	31.3	31.3	24.6	24.7	38.5	37.8	35.9	66	68	4.3	2.5	1013.1	0
10/21/2014 10:20	31.6	31.6	31.6	24.7	24.7	38.4	38.3	34.3	67	67	11.2	2.9	1013.1	0
10/21/2014 10:10	31.4	31.3	31.3	24.8	24.9	38.2	38	35.1	68	69	4.3	1.4	1013.1	0
10/21/2014 10:00	31.3	30.8	30.8	24.9	25.2	38	37.6	35.6	69	72	4.3	1.8	1013.1	0
10/21/2014 09:50	31.1	30	30	24.8	24.4	37.6	35.5	31.8	69	72	6.5	2.9	1013.1	0
10/21/2014 09:40	31	29.6	29.6	24.8	24.6	37.5	35.1	31.7	70	75	6.5	2.2	1013	0
10/21/2014 09:30	30.9	29.3	29.3	24.6	24.7	37.1	34.8	32.9	69	76	4.3	1.8	1013	0
10/21/2014 09:20	30.8	29.8	29.8	24.4	25.2	36.8	36.1	33.2	69	77	6.5	1.8	1012.9	0
10/21/2014 09:10	30.7	28.7	28.7	24.3	24.9	36.5	33.8	31.6	69	80	4.3	1.8	1012.8	0
10/21/2014 09:00	30.7	28.7	28.7	24.4	24.6	36.5	33.6	29.7	70	79	8.6	3.2	1012.8	0
10/21/2014 08:50	30.6	29	29	24.5	24.9	36.4	34.4	30.3	70	79	6.5	2.9	1012.7	0
10/21/2014 08:40	30.4	28.9	28.9	24.3	25.1	36.1	34.4	30.3	70	80	6.5	2.5	1012.6	0
10/21/2014 08:30	30.3	28.2	28.2	24.2	24.6	35.8	32.8	30.4	70	81	4.3	2.2	1012.5	0
10/21/2014 08:20	30.2	28	28	24.1	24.5	35.5	32.3	29.2	70	81	4.3	1.8	1012.3	0
10/21/2014 08:10	30.1	27.5	27.5	24	24.5	35.3	31.4	30.8	70	84	4.3	1.1	1012.2	0
10/21/2014 08:00	30	26.8	26.8	23.9	23.9	35	29.1	26.6	70	85	6.5	1.8	1012.1	0
10/21/2014 07:50	30	26.2	26.2	23.9	23.5	35	26.2	22.7	70	85	4.3	2.9	1012	0
10/21/2014 07:40	29.9	25.9	25.9	23.8	23.4	34.8	25.9	21.9	70	86	6.5	2.5	1011.9	0
10/21/2014 07:30	29.8	25.5	25.5	23.7	23.3	34.6	25.5	23	70	88	4.3	2.2	1011.8	0
10/21/2014 07:20	29.8	25	25	23.7	23.1	34.6	25	21.5	70	89	6.5	2.5	1011.7	0
10/21/2014 07:10	29.8	24.7	24.7	23.7	22.8	34.6	24.7	21.1	70	89	6.5	2.9	1011.7	0
10/21/2014 07:00	29.8	24.6	24.6	23.7	22.7	34.6	24.6	22.1	70	89	4.3	1.4	1011.5	0
10/21/2014 06:50	29.8	24.6	24.6	23.7	22.6	34.6	24.6	23	70	88	4.3	1.4	1011.3	0
10/21/2014 06:40	29.8	24.7	24.7	23.8	22.6	34.6	24.7	22	70	88	4.3	2.2	1011.2	0
10/21/2014 06:30	29.9	24.9	24.9	23.8	22.6	34.8	24.9	22.5	70	87	4.3	1.8	1010.9	0
10/21/2014 06:20	29.9	25	25	23.8	22.6	34.8	25	23.1	70	87	4.3	0.7	1010.7	0
10/21/2014 06:10	29.9	25.1	25.1	23.9	22.6	34.9	25.1	25.1	70	86	0	0	1010.7	0
10/21/2014 06:00	30	25.2	25.2	23.9	22.7	35	25.2	25.2	70	86	0	0	1010.5	0
10/21/2014 05:50	30	25.1	25.1	23.9	22.6	35	25.1	25.1	70	86	0	0	1010.5	0
10/21/2014 05:40	30	25.1	25.1	23.9	22.6	35.1	25.1	24.4	70	86	2.2	0.4	1010.5	0
10/21/2014 05:30	30.1	25.2	25.2	24	22.7	35.3	25.2	24.8	70	86	2.2	0.7	1010.4	0
10/21/2014 05:20	30.1	25.2	25.2	24	22.7	35.3	25.2	25.2	70	86	0	0	1010.2	0
10/21/2014 05:10	30.2	25.3	25.3	24.1	22.8	35.5	25.3	24.9	70	86	2.2	0.7	1010.3	0
10/21/2014 05:00	30.2	25.3	25.3	24.1	22.8	35.5	25.3	25.3	70	86	0	0.4	1010.3	0
10/21/2014 04:50	30.2	25.2	25.2	24.1	22.7	35.5	25.2	24.4	70	86	2.2	0.4	1010.2	0
10/21/2014 04:40	30.3	25.2	25.2	24.1	22.7	35.7	25.2	24.1	70	86	4.3	0.7	1010.2	0
10/21/2014 04:30	30.3	25.3	25.3	24	22.8	35.6	25.3	23.5	69	86	4.3	1.1	1010.2	0
10/21/2014 04:20	30.3	25.4	25.4	24	22.9	35.6	25.4	25.4	69	86	0	0	1010.2	0
10/21/2014 04:10	30.4	25.5	25.5	24.1	22.9	35.8	25.5	25.5	69	86	2.2	0.4	1010.1	0
10/21/2014 04:00	30.4	25.4	25.4	24.1	22.8	35.8	25.4	25.2	69	86	2.2	0.4	1010.1	0
10/21/2014 03:50	30.4	25.4	25.4	24.1	22.9	35.8	25.4	25.4	69	86	0	0	1010.1	0
10/21/2014 03:40	30.4	25.5	25.5	24.1	22.8	35.9	25.5	25.5	69	85	0	0	1010.1	0
10/21/2014 03:30

0

0

0

0

0

0

0

0

0

0

0	0	0	0
10/21/2014 03:20	30.6	25.5	25.5	24.2	23	36.2	25.5	25	69	86	2.2	0.7	1010.1	0
10/21/2014 03:10	30.6	25.5	25.5	24.2	23	36.3	25.5	25	69	86	2.2	1.1	1010.3	0
10/21/2014 03:00	30.6	25.6	25.6	24.2	22.9	36.3	25.6	25.3	69	85	2.2	0.7	1010.2	0
10/21/2014 02:50	30.7	25.6	25.6	24.3	22.9	36.4	25.6	24	69	85	4.3	1.4	1010.4	0
10/21/2014 02:40	30.7	25.7	25.7	24.3	23	36.5	25.7	24.2	69	85	4.3	0.7	1010.5	0
10/21/2014 02:30	30.8	25.8	25.8	24.4	23.1	36.7	25.8	24.3	69	85	4.3	1.1	1010.6	0
10/21/2014 02:20	30.8	25.8	25.8	24.3	23.1	36.6	25.8	25.8	68	85	0	0	1010.5	0
10/21/2014 02:10	30.8	25.8	25.8	24.2	22.9	36.5	25.8	25.8	68	84	0	0	1010.5	0
10/21/2014 02:00	30.9	26	26	24.3	23.2	36.8	26	25.7	68	85	2.2	0.4	1010.7	0
10/21/2014 01:50	30.9	26	26	24.3	23.2	36.8	26	26	68	85	0	0	1010.6	0
10/21/2014 01:40	31	26	26	24.5	23.1	37.1	26	26	69	84	0	0	1010.7	0
10/21/2014 01:30	31	26	26	24.6	23	37.3	26	26	69	84	0	0	1010.8	0
10/21/2014 01:20	31	26.2	26.2	24.5	23.1	37.2	26.2	26.2	69	83	2.2	0	1010.9	0
10/21/2014 01:10	31	26.3	26.3	24.4	23.2	37.1	26.3	26.3	68	83	0	0	1011.1	0
10/21/2014 01:00	31.1	26.4	26.4	24.5	23.3	37.3	26.4	26.4	68	83	0	0	1011.2	0
10/21/2014 00:50	31.1	26.4	26.4	24.5	23.3	37.3	26.4	26.4	68	83	0	0	1011.2	0
10/21/2014 00:40	31.2	26.6	26.6	24.6	23.3	37.5	26.6	26.6	68	83	0	0	1011.3	0
10/21/2014 00:30

0

0

0

0

0

0

0

0

0

0

0	0	0	0
10/21/2014 00:20	31.2	26.7	26.7	24.6	23.3	37.6	28.1	28.1	68	82	0	0	1011.5	0
10/21/2014 00:10	31.3	26.8	26.8	24.7	23.4	37.8	29.4	29.4	68	82	0	0	1011.4	0
10/21/2014 00:00	31.3	26.8	26.8	24.7	23.4	37.8	29.4	29.4	68	82	0	0	1011.5	0
10/20/2014 23:50	31.4	26.8	26.8	24.8	23.4	38.1	29.6	29.6	68	82	0	0	1011.4	0
10/20/2014 23:40	31.4	26.8	26.8	24.7	23.3	38	29.4	29.4	68	81	0	0	1011.5	0
10/20/2014 23:30	31.4	26.8	26.8	24.6	23.3	37.9	29.4	29.4	67	81	0	0	1011.5	0
10/20/2014 23:20	31.5	26.8	26.8	24.6	23.3	38	29.4	29.4	67	81	0	0	1011.6	0
10/20/2014 23:10	31.6	27	27	24.7	23.5	38.1	29.8	29.7	67	81	2.2	0	1011.7	0
10/20/2014 23:00	31.6	27.1	27.1	24.7	23.5	38.3	30.1	30.1	67	81	0	0	1011.7	0
10/20/2014 22:50	31.6	27	27	24.7	23.5	38.3	29.9	29.9	67	81	0	0	1011.5	0
10/20/2014 22:40	31.7	27.1	27.1	24.8	23.5	38.5	30.1	30.1	67	81	0	0	1011.6	0
10/20/2014 22:30	31.7	27.2	27.2	24.8	23.4	38.7	30.2	30.2	67	80	0	0	1011.6	0
10/20/2014 22:20	31.8	27.2	27.2	24.9	23.5	38.8	30.3	30.1	67	80	2.2	0	1011.5	0
10/20/2014 22:10	31.8	27.3	27.3	24.8	23.5	38.7	30.4	30.4	67	80	0	0	1011.5	0
10/20/2014 22:00	31.9	27.3	27.3	24.7	23.6	38.7	30.5	30.5	66	80	0	0	1011.6	0
10/20/2014 21:50	32	27.4	27.4	24.8	23.6	38.9	30.6	30.6	66	80	2.2	0	1011.7	0
10/20/2014 21:40	32	27.4	27.4	24.8	23.6	39	30.6	30.5	66	80	2.2	0	1011.8	0
10/20/2014 21:30	32	27.3	27.3	24.9	23.5	39.1	30.4	30.4	66	80	0	0	1011.8	0
10/20/2014 21:20	32.1	27.4	27.4	24.9	23.5	39.3	30.6	30.1	66	79	2.2	0.4	1012	0
10/20/2014 21:10	32.2	27.6	27.6	25	23.8	39.5	31.1	31.1	66	80	0	0	1012	0
10/20/2014 21:00	32.2	27.5	27.5	24.8	23.6	39.2	30.9	30.9	65	79	2.2	0	1011.9	0
10/20/2014 20:50	32.2	27.6	27.6	24.8	23.6	39.3	31	31	65	79	0	0	1011.8	0
10/20/2014 20:40	32.3	27.6	27.6	24.9	23.6	39.5	31	30.9	65	79	2.2	0	1011.7	0
10/20/2014 20:30	32.4	27.8	27.8	24.8	23.7	39.6	31.4	31.4	64	79	2.2	0	1011.7	0
10/20/2014 20:20	32.4	27.9	27.9	24.7	23.7	39.4	31.6	30.9	64	78	2.2	0.4	1011.6	0
10/20/2014 20:10	32.4	27.9	27.9	24.7	23.7	39.4	31.6	31.3	64	78	2.2	0.4	1011.5	0
10/20/2014 20:00	32.4	27.9	27.9	24.7	23.7	39.4	31.5	30.6	64	78	2.2	0.7	1011.4	0
10/20/2014 19:50	32.5	27.7	27.7	24.8	23.6	39.6	31.2	30.9	64	78	2.2	0.7	1011.2	0
10/20/2014 19:40	32.5	27.8	27.8	24.8	23.6	39.6	31.3	30.6	64	78	2.2	0.7	1011	0
10/20/2014 19:30	32.5	28.2	28.2	24.6	23.5	39.3	31.9	30.8	63	76	2.2	0.7	1010.7	0
10/20/2014 19:20	32.5	28.4	28.4	24.6	23.6	39.3	32.3	30.7	63	75	4.3	1.1	1010.4	0
10/20/2014 19:10	32.4	28.7	28.7	24.7	23.6	39.3	32.7	31.1	64	74	4.3	1.1	1010.3	0
10/20/2014 19:00	32.3	28.9	28.9	24.3	23.6	38.8	33	31.6	63	73	2.2	1.1	1010.4	0
10/20/2014 18:50	32.4	29.1	29.1	24.3	23.5	38.8	33.3	32	62	72	2.2	1.1	1010.3	0

Displaying 1-141 of 141 results in 1 pages. 

[Lucas Heijst]

On Tuesday, 21 October 2014 19:05:38 UTC-3, Lucas Heijst wrote:

PS. The database for "Weerstation Paramaribo" now show rain rate; I don't understand why this isn't shown for "Weerstat. Paramaribo 3". AFAIK the settings are the same.

My fault. For "Weerstat. Paramaribo 3" I had set Rain rate to OFF. It is now set to ON.

[Lucas Heijst]

On Tuesday, 21 October 2014 11:57:31 UTC-3, mwall wrote:

wcloud v0.4 is now available on the wiki.

m

Matthew,

Parameter wdiravg have values of 0 and 360 degrees. Is this correct?

Luc

Weerstation Paramaribo

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )

Humin ( % )	Hum ( % )

Wspdhi ( km/h )	Wspdavg ( km/h )	Wdiravg ( º )

Bar ( hPa )	Rain ( mm )

Rainrate ( mm/h )
10/21/2014 19:50	30.6	27.3	27.3	22.2	21.6	34.3	29.5	61	71	2.2	0	292	1011.2	0	0
10/21/2014 19:40	30.6	27.2	27.2	22.3	21.6	34.4	29.3	61	72	2.2	0	360	1011	0	0
10/21/2014 19:30	30.7	26.9	26.9	22.6	21.2	34.7	28.7	62	71	2.2	0.4	61	1011	0	0
10/21/2014 19:20	30.8	27.4	27.4	22.6	21.7	34.9	29.7	62	71	2.2	0	0	1010.8	0	0
10/21/2014 19:10	30.9	27.7	27.7	22.5	21.7	34.8	30.1	61	70	2.2	0.4	172	1010.6	0	0
10/21/2014 19:00	30.9	28	28	22.5	22.1	34.9	30.7	61	70	6.5	1.4	323	1010.6	0	0
10/21/2014 18:50	31.1	28.2	28.2	22.7	22.1	35.3	31	61	70	8.6	3.2	293	1010.4	0	0
10/21/2014 18:40	31.4	28.6	28.6	23.2	22.7	36.2	31.9	62	71	8.6	1.8	315	1010	0	0
10/21/2014 18:30	31.6	28.7	28.7	23.7	22.9	37	32.3	63	71	6.5	1.4	337	1009.7	0	0

Displaying 1-9 of 9 results in 1 pages. 

[Lucas Heijst]

On Tuesday, 21 October 2014 11:57:31 UTC-3, mwall wrote:

wcloud v0.4 is now available on the wiki.

makes the wspdhi more correct for the weathercloud specification.
m

Matthew,

I would expect for values of wspdhi of not null the wspdavg also wouln't be null, but sometimes they are.

Luc

Weerstation Paramaribo

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )

Humin ( % )	Hum ( % )

Wspdhi ( km/h )	Wspdavg ( km/h )	Wdiravg ( º )

Bar ( hPa )	Rain ( mm )

Rainrate ( mm/h )

[mwall]

i see a few issues with weathercloud api:

- they do not consistently distinguish between NULL/None and 0.  i was sending non-null thw for awhile, then i started sending None for thw.  now when i view the database, i see 0 where i sent None.  on the other hand, i see dash (-) for heatin, and i presume that means NULL/None (even though i have been sending non-None values).

- they require a 'time' field, but that has no date.  the api doc does not specify utc, but email from weathercloud says it should be utc.  it would seem that the 'time' field is not, in fact used.  it *should* be the time (and date) of the observations.  email from weathercloud says it should be 'exact time of upload to server'.

- the api doc says wind directions should be in [0, 359].  but what about 359.9?  i should think it would be either [0,360] or [0,360)

- user authentication is done with cgi args, but https is not required.  this is simply insecure and exposes them to vulnerabilities in the future.

m

[mwall]

On Tuesday, October 21, 2014 7:04:35 PM UTC-4, Lucas Heijst wrote:

On Tuesday, 21 October 2014 11:57:31 UTC-3, mwall wrote:

wcloud v0.4 is now available on the wiki.
makes the wspdhi more correct for the weathercloud specification.
m

Matthew,

I would expect for values of wspdhi of not null the wspdavg also wouln't be null, but sometimes they are.

the spec says wspdhi is 10 minute average of gust.  however, 'gust' is ill-defined.  some hardware has gust (which in this case means 'max wind in a sampling period', typically because the hardware samples the wind infrequently, e.g., fineoffset), some does not (ultimeter).  ultimeter does not need it since it reports instantaneous values as fast as every 1/2 second.  so i implemented wspdhi as the 10 minute average of the max of windSpeed and windGust.  using this approach can lead to a disconnect between wspdhi and wspdavg.  happy to entertain alternatives.

imho, weathercloud is making many of the same mistakes other services have made: ill-defined timestamps; schema based too much on davis instruments; authentication in the args; ill-defined wind speed/gust; use of forecasts-from-station-hardware.  however, they appear to be learning and making favorable changes.

at one extreme you have thingspeak or xively (completely unconstrained), at the other you have windfinder or weatherbug (over constrained).

m

[Lucas Heijst]

On Tuesday, 21 October 2014 20:09:03 UTC-3, mwall wrote:

- they do not consistently distinguish between NULL/None and 0.  i was sending non-null thw for awhile, then i started sending None for thw.  now when i view the database, i see 0 where i sent None.  on the other hand, i see dash (-) for heatin, and i presume that means NULL/None (even though i have been sending non-None values).

m

Matthew,

On Oct 21 03:30:21weewx sent a record with only a few parameters to wcloud:

http://api.weathercloud.net/v01/set?wid=xxx&ver=270&wdiravg=225&rain=0&key=xxx&time=0330&wspdavg=0&type=251 

The database show null values:

Date	Tempin ( ºC )	Temp ( ºC )	Chill ( ºC )	Dewin ( ºC )	Dew ( ºC )	Heatin ( ºC )	Heat ( ºC )

Thw ( ºC )

Humin ( % )	Hum ( % )	Wspdhi ( km/h )	Wspdavg ( km/h )

Bar ( hPa )	Rain ( mm )	Rainrate ( mm/h )

10/21/2014 18:20

31.9

29.3

29.3

24.1

23.5

37.8

33.6

0

64

71

17.6

4.3

1010.1

0

0

10/21/2014 03:30

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Note, the parameter wdiravg (=225) is not shown in the table (see header).

In the plotted data (last 24 hour), the null values are not shown, see attachment.

Luc

[mwall]

it would appear that weathercloud is applying some sanity checking to the values it receives.

the wcloud uploader 0.4 was calculating interior dewpoint and heatindex assuming metric values, even for databases using US units.  the result was very high heatin, and weathercloud turned those into dashes.

wcloud 0.5 fixes heatin and dewin calculations.

m

[Lucas Heijst]

On Wednesday, 22 October 2014 10:51:22 UTC-3, mwall wrote:

wcloud 0.5 fixes heatin and dewin calculations.
m

Thanks Matthew,

v0.5 still produces good heatin and dewin values for my database using EU units.

Luc 

[Simon]

Thanks for your hard work Matthew. seems to be working fine here:

http://app.weathercloud.net/d1185736460

[Simon]

Just for your information....

"Today we’re very pleased to announce that Weathercloud is now also compatible with WeatherCat and WeeWX, thus bringing to 8 the total number of weather station programs supported so far."

http://blog.weathercloud.net/?p=867