Update 4.9s

What's hot...

Moderator: Mattk

User avatar
N5XL
Senior Boarder
Senior Boarder
Posts: 45
Joined: Wed Dec 28, 2011 6:04 am
Location: Scottsdale, Arizona

Re: Update 4.9s

Post by N5XL » Sat Jan 05, 2013 7:53 pm

El,
When I looked back at this email picture, I see you had selected th0 for your sensor for the Envoy. Should it have been th1 or is this because you still do not have your ISS yet?

Image

User avatar
N5XL
Senior Boarder
Senior Boarder
Posts: 45
Joined: Wed Dec 28, 2011 6:04 am
Location: Scottsdale, Arizona

Re: Update 4.9s

Post by N5XL » Sat Jan 05, 2013 10:06 pm

El,

While we wait for you to look at this again this afternoon, it does look to me as something odd is going on. What I cannot determine is if this oddity is associated only with the Envoy or if it’s a problem when multiple stations are attached.

What I do know is that the sea level pressure calculations I performed on your station appear to be close enough to assume that the WS2300 pressure data is ok.

I also know that if I look at “actual thb1 sealevel hpa” from the Envoy it does not change as you change the altitude. Furthermore, the values recorded are approximately the same what the WS2300 reports as console pressure. This is consistent with unadjusted console pressure and implies that values labelled as “actual thb1 sealevel hpa” for the Envoy are really “actual thb1 pressure hpa”. Not sure whats going on there.

I hope the problem goes away with only the Envoy by itself!

Dave

el-rain
Junior Boarder
Junior Boarder
Posts: 27
Joined: Tue Aug 14, 2012 11:10 am

Re: Update 4.9s

Post by el-rain » Sun Jan 06, 2013 12:36 am

Hi David,
just back home.

I unplugged the old weatherstation physically from USB Port of the iConnect, deleted it from weather station menu and reload meteohub. But problem is still the same.

actual_thb1_height_m 75
actual_thb1_press_hpa 1015.4
actual_thb1_sealevel_hpa 1024.2

When I compare with local stations, Airport FMO (about 13km North,FMO on wetteronline.de) and local university station (about 3.7km South, show only local pressure, Institute of Landscape Ecology of the University of Muenster) the value shown as sealevel here seems to be the local pressure.

When I checked the data I loaded from the logger to the Davis Weatherlink software it show only the sealevel value which was at about 1031 this morning when the station altitude was still on 75m in the Envoy and goes down to about 1024 when I change it to 0m.

When I read you last post it seems you see what I find suspect here now as well. Looks to me like somehow the local presure is used as sealevel and the local one is calculated from that one by using the station altitude.

Even more strange to me is that it don't madder if configure in the weather station menu if the sealevel claculation should be "based on station's altitude" or "use reading from weatherstation" the values shown in the sensors tab or on <meteohub URL>/meteograph.cgi?text=all pressure values are the same for both.

To answer your question with the th0, this is greyed out for the used options and metehub switch is always back to th0 and currently there is even no th1 and the ISS isn't installed yet.
It is only selectable if you use "based on station's altitude and temperature.

Is there any possebillity, maybe via SSH, to do a reset of all weatherstation data without deleting the network settions, as my iConnect is connected via WLAN and I have no ethernet I need to remove it from location to an other room to do a full reset, as I assue that the reset under maintenace will reset as well the network settings. This might the last option I can do, as the Envoy is still shown as station1 and there seems to be some data form station0 (WS2300) in the system, maybe worth to get rid of all of them. If this will not fix this, I'm totally out of ideas.

vy73
EL

el-rain
Junior Boarder
Junior Boarder
Posts: 27
Joined: Tue Aug 14, 2012 11:10 am

Re: Update 4.9s

Post by el-rain » Sun Jan 06, 2013 2:08 am

Hi,
when searching around about this a bit in the web I reviewed the readme file of VPLive and there I found in a statement which might explain all this.
Calculates the raw station pressure and the altimeter pressure from the console's reported sea level reduced pressure
After that I searched on the Davis Webpage and take a look in the latest Vantage Pro and Pro2 Communications Reference from 24. August 2012 and there I see that in the normal LOOP packet are only one Barometric value is reported. Together with the statement in the VPLive readme this might explain my behaviour. As in that case the only way Metehub can handle this is to take the received value as sealevel and recalculate the local pressure based on the altitude.

There is as well a LOOP2 packet but as I see in some other threads I found in the internet this seems to be very new. The document above is from 30. July 2012, and the section for LOOP2 was added in that revision.

In LOOP2 are multiple additional Barometric values:

Code: Select all

Barometric Reduction Method: 0 - user offset 1- Altimeter Setting 2- NOAA Bar Reduction. For VP2, this will always be 2.
User-entered Barometric Offset: Barometer calibration number in 1000th of an inch
Barometric calibration number: Calibration offset in 1000th of an inch
Barometric Sensor Raw Reading: In 1000th of an inch
Absolute Barometric Pressure: In 1000th of an inch, equals to the raw sensor reading plus user entered offset
Altimeter Setting: In 1000th of an inch
But as this seems very new I'm not sure if this is supported by meteohub.

I take a look as well into the metehub manual there is stated that only a subset of Davis Vantage2 funktions but there is no comment about the behaviour of the pressure in.

I believe the only person can answer this is Boris, as he know what is supported and how this is impemented.

I will go to station location and set altitude back into the Envoy, this seems to be at least a workaround to get good pressure values, even if I can't change them via meteohub.

vy73
EL

User avatar
N5XL
Senior Boarder
Senior Boarder
Posts: 45
Joined: Wed Dec 28, 2011 6:04 am
Location: Scottsdale, Arizona

Re: Update 4.9s

Post by N5XL » Sun Jan 06, 2013 2:35 am

El,

Thinking about this still. If we go on the theory that:
1. the Envoy is only sending one pressure value that is adjusted by setting elevation in the Envoy and
2. if you set the Envoy elevation to zero, you send console pressure (regardless of what Meteohub thinks it really is), and
3. it appears that Meteohub applies sea level adjustments to this value

what would happen if you set the Envoy to zero meters to force the Envoy to send console pressure to Meteohub, then set a negative altitude in Meteohub? In other words, would Meteohub apply the correct calculation if you set Envoy to zero meters and Meteohub to -75 meters? It may be that Meteohub would adjust "actual thb1 pressure" to be the correct sea level pressure.

Dave

el-rain
Junior Boarder
Junior Boarder
Posts: 27
Joined: Tue Aug 14, 2012 11:10 am

Re: Update 4.9s

Post by el-rain » Sun Jan 06, 2013 12:45 pm

Hi David,
I tried this, but with negative altitude it looks like both pressure and sealevel are recalculated as both are changed to the same value.
The other question is how this impact the Altimeter pressure as I will send this data as well to APRS.

actual_thb1_height_m 0
actual_thb1_press_hpa 1020.5
actual_thb1_sealevel_hpa 1020.5
actual_thb1_altimeter_hpa 1020.5

actual_thb1_height_m 75
actual_thb1_press_hpa 1011.7
actual_thb1_sealevel_hpa 1020.5
actual_thb1_altimeter_hpa 1020.7

actual_thb1_height_m -75
actual_thb1_press_hpa 1029.4
actual_thb1_sealevel_hpa 1029.4
actual_thb1_altimeter_hpa 1020.3

actual_thb1_height_m -150
actual_thb1_press_hpa 1038.2
actual_thb1_sealevel_hpa 1038.2
actual_thb1_altimeter_hpa 1019.9

to compare:
Local AirPort sealevel: 1028
University pressure: 1019.7

vy73
EL

User avatar
N5XL
Senior Boarder
Senior Boarder
Posts: 45
Joined: Wed Dec 28, 2011 6:04 am
Location: Scottsdale, Arizona

Re: Update 4.9s

Post by N5XL » Sun Jan 06, 2013 8:25 pm

El,

This oddity appears to be caused by the inability of the Envoy to send a proper console pressure, coupled with the way that Meteohub is interpreting the values sent by the Envoy. In other words, it appears there are two different issues going on. For Meteohub to work correctly, it must have local console pressure from the Envoy to perform its altimeter and sea level pressure calculations and this is not happening.

The first problem appears to be that the Envoy is not sending unadjusted console pressure to Meteohub but is instead sends something else. The second problem appears as if Meteohub may be labeling certain data fields from the Envoy incorrectly. As I do not know the format that the Envoy sends data and I do not know how Boris is interpreting this data, I can only speculate. I would expect that Davis uses the same data protocol for all of its equipment and not just something different for the Envoy, so others should have this problem too but I don’t know this for a fact.

What I do know is that according to the weather service, the current altimeter for airport EDDG (FMO) is 1028.0 hpa. If you are close to FMO and the weather is stable, your Meteohub reported altimeter should also be close to 1028.0. For this to be the case at your 75 meters station elevation, your raw console pressure needs to be 1019 hpa. It does not appear that setting your Meteohub to 0 meters gives you values of 1019 hpa however. Assuming that the Envoy is also at 0 meters and no SLP offset, I honestly don’t know what to try next other than perhaps a negative elevation in the Envoy to force it to send the correct console pressure.

You might also look at the WS2300 to see what values show up for console pressure there (actual thb0 press hpa)…it should be very close to 1019 hpa.

Boris, we are probably at a point where we could use your insight about the data format from the Envoy and what might be happening.

73
Dave

el-rain
Junior Boarder
Junior Boarder
Posts: 27
Joined: Tue Aug 14, 2012 11:10 am

Re: Update 4.9s

Post by el-rain » Sun Jan 06, 2013 9:35 pm

Hi Dave,
Protocol of Davis is documented, you can find this on the Davis Webpage as stated before in this thread.
As well they have updated this, needed Firmware Version 1.9 on Console/Envoy, which add a new dataformat this include as well the raw pressure value. The 1.9 was published in 2010 but the document describe LOOP2 was published August 2012.

What Meteohub currently get from Davis Equipment, only LOOP dataformat or as well the new LOOP2, which include raw pressure, and how Meteohub will work with that can only be anserd from Boris.
To me it looks like the LOOP is used, which seems to report sealevel, and the workaround to get local pressure is to calculate this with the altitude, but this is speculation only.

What I did now is to set altitude in the Envoy and in Metehub in a way it match the values I have from the airport EDDG (FMO) and the local station from our University which is only half of the distance I have to the EDDG. As the EDDG in North and the University is south and there pressure closely match at the moment this should give an acceptable pressure for a privat station.

Maybe Boris will implement LOOP2 somewhere in future so this workaround I did isn't needed anymore but since then I can live with that.

Currently:
WS2300
actual_thb0_height_m 75
actual_thb0_press_hpa 1020.1
actual_thb0_sealevel_hpa 1028.9
actual_thb0_altimeter_hpa 1029.2

Envoy
actual_thb1_height_m 75
actual_thb1_press_hpa 1020.2
actual_thb1_sealevel_hpa 1029.0
actual_thb1_altimeter_hpa 1029.3

University: 1020.1 hPa
EDDG: 1029

Thank you for your great support here to help me to figure out my problem and the suggestions to identify the root cause.

vy73
El
DG7YEO

User avatar
N5XL
Senior Boarder
Senior Boarder
Posts: 45
Joined: Wed Dec 28, 2011 6:04 am
Location: Scottsdale, Arizona

Re: Update 4.9s

Post by N5XL » Mon Jan 07, 2013 4:17 pm

El,
What Meteohub currently get from Davis Equipment, only LOOP dataformat or as well the new LOOP2, which include raw pressure, and how Meteohub will work with that can only be anserd from Boris. To me it looks like the LOOP is used, which seems to report sealevel, and the workaround to get local pressure is to calculate this with the altitude, but this is speculation only.
I read the communication protocol that Davis published and I agree with you. It appears that Meteohub may be requesting the "Barometer" value from a "Loop 1" format packet. This appears to be the only way to get any barometer value from the old Davis equipment. Because of this, it looks like Meteohub has to "undo" the Davis calculation to calculate a raw console pressure. The problem with this approach is that there are numerous parameters that affect the barometer readings on Davis equipment (Elevation, Dew Point, Virtual Temperature, C (humidity correction factor), R (correction ratio), Barcal (linear offset), Gain, and Offset (calibration parameters. I dont know how Meteohub deals with all of these variables from Davis equipment as changing any one of them will force a different value to be given by the "Barometer" command. This is a bad situation for Meteohub.

On newer Vantage Pro2 firmware 1.9 and Envoys, that is no longer the case. Meteohub can instead request a Loop 2 packet and specifically get the "Barometric Sensor Raw Reading" which is the exact parameter Meteohub needs without having to calculate anything. This is the ideal situation for Meteohub.

Boris still probably needs to verify the barometric calculations used for "Loop 1" equipment, but It looks like Boris could benefit by adding "Loop 2" support for the newer firmware.

I am glad you have a temporary solution El ! This is not an easy issue to solve.

73
Dave
N5XL

wfpost
Platinum Boarder
Platinum Boarder
Posts: 587
Joined: Thu Jun 12, 2008 2:24 pm
Location: HONSOLGEN
Contact:

Re: Update 4.9s

Post by wfpost » Tue Jan 08, 2013 5:47 pm

Boris,

anything you can check, why email is broken now ??
I don´t want to go back to 4.9q

Thanks
wfpost wrote:just realized, that email stopped after updating to version 4.9r

Always getting segfault errors, even with the testmail function
just updated to 4.9s, but still the error with not sending emails


Jan 4 00:15:20 meteohub kernel: sendmail[3263]: segfault at 00000000 eip 00000000 esp bfa8acac error

Dec 30 09:00:42 meteohub kernel: sendmail[13089]: segfault at 00000000 eip 00000000 esp bfb1935c error 4
Dec 30 23:55:10 meteohub kernel: sendmail[26211]: segfault at 00000000 eip 00000000 esp bf8b853c error 4
Dec 30 23:59:41 meteohub kernel: sendmail[26833]: segfault at 00000000 eip 00000000 esp bfc7a0dc error 4
Dec 30 23:59:45 meteohub kernel: sendmail[26844]: segfault at 00000000 eip 00000000 esp bfc372ec error 4
Dec 31 09:00:47 meteohub kernel: sendmail[31099]: segfault at 00000000 eip 00000000 esp bfa6327c error 4
...
Jan 4 00:06:59 meteohub kernel: sendmail[10558]: segfault at 00000000 eip 00000000 esp bfbc96bc error 4

User avatar
admin
Platinum Boarder
Platinum Boarder
Posts: 5025
Joined: Mon Oct 01, 2007 10:51 pm

Re: Update 4.9s

Post by admin » Thu Jan 10, 2013 11:16 pm

Strange, email send works ok on my x86 with 4.9s. And I also did not touch the sendmail itself by one of the latest updates.
Do you have done any debian updates?

wfpost
Platinum Boarder
Platinum Boarder
Posts: 587
Joined: Thu Jun 12, 2008 2:24 pm
Location: HONSOLGEN
Contact:

Re: Update 4.9s

Post by wfpost » Thu Jan 10, 2013 11:19 pm

No, I have not done any debian updates.
Hmm, very strange. If I try to go back to 4.9q will I get Demo Licence issue failures?

Or is it possible to have a look into my machine? Thanks!

User avatar
admin
Platinum Boarder
Platinum Boarder
Posts: 5025
Joined: Mon Oct 01, 2007 10:51 pm

Re: Update 4.9s

Post by admin » Thu Jan 10, 2013 11:21 pm

I can try, would need access on ports 22 and 7777.

wfpost
Platinum Boarder
Platinum Boarder
Posts: 587
Joined: Thu Jun 12, 2008 2:24 pm
Location: HONSOLGEN
Contact:

Re: Update 4.9s

Post by wfpost » Thu Jan 10, 2013 11:33 pm

both port are now open.
There was a cgi so start, or shall I email you the details??

I´ve called the help.cgi
open ssh tunnel to meteohub support
done

Post Reply