Correct weewx.conf settings for Inspeed Anemometer and Vane

236 views
Skip to first unread message

maf....@gmail.com

unread,
Feb 20, 2015, 10:34:41 AM2/20/15
to weewx...@googlegroups.com
I am trying to get my station up and running and am wondering what the correct setting should be for weewx.conf since I see somewhat conflicting suggestions. I am using the Hobby Boards Anemometer board which has the DS2438 A/D to determine wind direction and the DS2423 counter for wind speed. On the counter, both input channels are connected together by a shorting resistor so that counter.A and counter.B should read the same (more on this later). The anemometer is a Inspeed OPTICAL unit with 8 pulses per revolution connected to the DS2423 counter inputs. This is different than their other models which have 1 or 4 pulses per revolution. The wind direction is an Inspeed E-VANE connected to the DS2438 A/D.

My current weewx.conf is:

##############################################################################

[OWFS]
    interface = u
    driver = user.owfs
    [[sensor_type]]
        windSpeed = inspeed_windspeed
    winDir = inspeed_windvane
    #rain = counter
    [[sensor_map]]
        outTemp = /uncached/28.D5A7C6000000/temperature
        outHumidity = /uncached/26.ADD891000000/humidity
    windSpeed = /uncached/1D.BCBF0D000000
    windDir = /uncached/26.5C6EE7000000/VAD
    #rain = /uncached/1D.496E0F000000/rain
##############################################################################

The main problem is that the windDirection is almost always right around North even though the vane moves around quite a bit and the windSpeed is generally close to 1 - 2 mph even though it is blowing much more.

Should the windSpeed entry have a "/counter.A" appended (even though the example does not have it)? Are the values computed using "inspeed_windspeed" and " inspeed_windvane" correct for the 8 pulse per rev and the Inspeed E-Vane or does something have to be changed in the code? If so, where? Other entries I have seen have "average" for the windSpeed sensor_type. Does "inspeed_windspeed" do an average?


Using .../owfs.py --sensors All of the 1-Wire devices are detected properly. Using .../owfs.py --readings returns the following for the A/D used for wind direction:

_usePath: /26.5C6EE7000000
_type: DS2438
_path: /26.5C6EE7000000
_useCache: True
  /26.5C6EE7000000/B1-R1-A/: gain,offset,pressure
  /26.5C6EE7000000/family: 26
  /26.5C6EE7000000/offset:            1
  /26.5C6EE7000000/locator: FFFFFFFFFFFFFFFF
  /26.5C6EE7000000/VDD:         4.94
  /26.5C6EE7000000/HIH3600/: humidity
  /26.5C6EE7000000/r_locator: FFFFFFFFFFFFFFFF
  /26.5C6EE7000000/id: 5C6EE7000000
  /26.5C6EE7000000/HTM1735/: humidity
  /26.5C6EE7000000/pages/: page.ALL,page.0,page.1,page.2,page.3,page.4,page.5,page.6,page.7
  /26.5C6EE7000000/HIH4000/: humidity
  /26.5C6EE7000000/temperature:      -2.8125
  /26.5C6EE7000000/disconnect/: date,udate
  /26.5C6EE7000000/MultiSensor/: type
  /26.5C6EE7000000/endcharge/: date,udate
  /26.5C6EE7000000/type: DS2438
  /26.5C6EE7000000/EE: 1
  /26.5C6EE7000000/IAD: 1
  /26.5C6EE7000000/CA: 1
  /26.5C6EE7000000/VAD:         4.92Why are there listings for various Humidity sensors?
  /26.5C6EE7000000/address: 265C6EE70000004E
  /26.5C6EE7000000/date: Wed Dec 31 23:49:10 1969
  /26.5C6EE7000000/r_address: 4E000000E76E5C26
  /26.5C6EE7000000/S3-R1-A/: current,gain,illuminance
  /26.5C6EE7000000/vis:     0.002441
  /26.5C6EE7000000/crc8: 4E
  /26.5C6EE7000000/udate:        17350
  /26.5C6EE7000000/humidity:      126.822
  /26.5C6EE7000000/r_id: 000000E76E5C

Why are there listings for various Humidity sensors?


For the wind speed the following is returned:

BCBF0D000000
_usePath: /1D.BCBF0D000000
_type: DS2423
_path: /1D.BCBF0D000000
_useCache: True
  /1D.BCBF0D000000/pages/: count.ALL,count.0,count.1,count.2,count.3,count.4,count.5,count.6,count.7,count.8,count.9,count.10,count.11,count.12,count.13,count.14,count.15,page.ALL,page.0,page.1,page.2,page.3,page.4,page.5,page.6,page.7,page.8,page.9,page.10,page.11,page.12,page.13,page.14,page.15
  /1D.BCBF0D000000/family: 1D
  /1D.BCBF0D000000/counters.ALL:         4105,       33848
  /1D.BCBF0D000000/counters.A:         4105
  /1D.BCBF0D000000/counters.B:        33848
  /1D.BCBF0D000000/locator: FF80FFFFFFFFFFFF
  /1D.BCBF0D000000/type: DS2423
  /1D.BCBF0D000000/crc8: 04
  /1D.BCBF0D000000/memory:
  /1D.BCBF0D000000/r_id: 0000000DBFBC
  /1D.BCBF0D000000/address: 1DBCBF0D00000004
  /1D.BCBF0D000000/mincount:         4105
  /1D.BCBF0D000000/r_address: 040000000DBFBC1D
  /1D.BCBF0D000000/r_locator: FFFFFFFFFFFF80FF
  /1D.BCBF0D000000/id: BCBF0D000000

Does anyone have an idea on why counter.A and counter.B have different values? They have the same input on the Hobby Board. It is always the case that counter.B has a much higher value than counter.A. What has to be changed to use a 8 pulse per rev anemometer?

Thank you for any help or suggestions.

Mike

Ralph Underwood

unread,
Feb 22, 2015, 6:32:56 PM2/22/15
to weewx...@googlegroups.com
I have the new Inspeed and have it connected to /counters.A   I don't have any connections to the counters.B. I have a Hobby Boards anemometer board with two counters, one is used. My WindDir connects to the OWFS direct.


Michael Form

unread,
Feb 22, 2015, 9:14:38 PM2/22/15
to weewx...@googlegroups.com
Ralph:

Which Inspeed anemometer do you have? They make a reed switch, a hall effect and optical type units with one, four and eight pulses per revolution. That is one of the reasons I was trying to figure out where the formula is to calculate wind speed is located to make any changes necessary. The Hobby Board Anemometer Board specifies a formula to calculate wind speed based on the number of counts but this would obviously be different for the three different types of outputs (1, 4 and 8 pulses per rev).

Which unit do you have? I have the optical 8 pulse per rev model.

I am not sure what you mean by "My WindDir connects to the OWFS direct". I have connected the output of the Inspeed E-Vane to the A/D input of the DS2438 on the Hobby Board. For some reason the value displayed by Weewx is always NORTH even though the A/D voltage varies. I would like to figure out how it calculates the value displayed. Does it do some sort of an average and over what period of time? I know the formula to calculate the direction but how does Weewx take into account the wildly varying directions as the vane spins around? Where is this determined?

This is one reason I was trying to determine the correct settings in the weewx.conf file.

Any help that someone can shed on this would be appreciated.

Mike


On 2/22/2015 6:32 PM, Ralph Underwood wrote:
I have the new Inspeed and have it connected to /counters.A   I don't have any connections to the counters.B. I have a Hobby Boards anemometer board with two counters, one is used. My WindDir connects to the OWFS direct.


--
You received this message because you are subscribed to a topic in the Google Groups "weewx-user" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/weewx-user/GuR6mzhB1vs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Ralph Underwood

unread,
Feb 22, 2015, 11:17:36 PM2/22/15
to weewx...@googlegroups.com
Sorry, I was wrong about the wind vane, I was following the wrong wires. Mine is connected as described in the HB documentation.  There is no connection to the counters.b terminal on the HB anemometer board.

I don't know which anemometer I have, it is new (within 6 months). Is there a marking I should look for?


My wind speed and direction are running in a test environment in my basement.  I have been trying to sort out the HB barometer when I have time - mine is basically a random number generator.

maf....@gmail.com

unread,
Feb 23, 2015, 6:04:29 PM2/23/15
to weewx...@googlegroups.com
I am not sure what the markings are as mine is up in the air about 20 feet and it is 12 degrees (and falling) with 20 mph winds with several feet of snow and ice all around so I don't think that I will go up and look. :)

On the other hand, you can just rotate the anemometer one revolution and count the number of pulses it generates. You can do this with a DVM if you do not have a scope or counter. If you have 8 pulses per rev you have the optical unit, if you have 4 per rev you have the Hall effect type and if there is only one pulse per rev you have the reed switch.

As an aside I do not think that wewwx or owfs handles the Inspeed anemometer correctly. I spoke with the manufacturer of the units (he lives in the next town) and from what he said it seems that the reported wind speed is about 25% low. I basically simulated the anemometer pulses on the bench and the reported wind speed is consistently below what it should be. I am not sure where the problem lies but it is almost (but not quite) linear so I can probably take care of it in the calibrate section of weewx.conf, although it would be better to actually fix it. AT this point I am not familiar enough with the software to attempt this so I will just fix it "after the fact".

Mike

maf....@gmail.com

unread,
Feb 23, 2015, 11:33:34 PM2/23/15
to weewx...@googlegroups.com
I should correct myself. After looking at the numbers, the windSpeed is not low by 25% it is high by a factor of 8 (approximately). I assume that the calibration was for the single pulse per rev rather than the 8 pulses per rev of the optical Inspeed. I took care of this in the calibrate section of weewx.cont.

Mike
Reply all
Reply to author
Forward
0 new messages