Loading ...
Sorry, an error occurred while loading the content.

Re: Suggestion: Ability to disable wind gust reporting completely

Expand Messages
  • jjantti2
    Oh, sorry about that. I haven t checked the yahoo.com mails lately. I ll get the test version up and running and see how it works out. Once I ve gathered
    Message 1 of 17 , Nov 8, 2009
    • 0 Attachment
      Oh, sorry about that. I haven't checked the yahoo.com mails lately.
      I'll get the test version up and running and see how it works out.

      Once I've gathered enough data, I'll let you know how it worked.

      Thanks for the assistance.

      Juha Laiho
      Wx Station Operator
      Helsinki, Finland
      ---------------------
      CW4473 / IUUSIMAA2
      DW3039 / IUUSIMAA8


      --- In wuhu_software_group@yahoogroups.com, "wuhu_software" <wuhu_software@...> wrote:
      >
      >
      > Juha,
      >
      > I sent you a test copy in email a few days ago.
      >
      >
      > --- In wuhu_software_group@yahoogroups.com, "jjantti2" <jjantti2@> wrote:
      > >
      > > Hi.
      > >
      > > That idea sounds reasonable and practical. If you get this feature up, I'd be more than happy to test it out for you and see what happens to the readings then.
      > >
      > > Best Regards,
      > > Juha Laiho
      > > Wx Station Operator
      > > Helsinki, Finland
      > > ---------------------
      > > CW4473 / IUUSIMAA2
      > > DW3039 / IUUSIMAA8
      > >
      > >
      > > --- In wuhu_software_group@yahoogroups.com, "wuhu_software" <wuhu_software@> wrote:
      > > >
      > > >
      > > > I just had another thought for catching these bad wind readings.
      > > >
      > > > What if we specify a threshold from the average wind reading in order to reject.
      > > >
      > > > For example, if enabled, and the threshold is set to say 20 mph, if a new wind reading is more than 20mph from the 10 minute average it would be rejected.
      > > >
      > > > This would allow the detection threshold to dynamically change (in higher wind speed conditions) rather than being fixed as will the other "questionable" threshold limit that always assumes low wind conditions.
      > > >
      > > > Let me know what you think.
      > > >
      > > >
      > > > --- In wuhu_software_group@yahoogroups.com, "wuhu_software" <wuhu_software@> wrote:
      > > > >
      > > > >
      > > > > Juha,
      > > > >
      > > > > Since the WS-23XX has only one speed reading, and gust is only the highest speed reading over the past 10 minutes, we can be sure that your normal wind readings are also corrupt. It is coincidence that you are not seeing these bad wind readings for the normal speed reading because in the case of WU, the readings are stored only every few minutes. So really, if we deal with only the gust problem, eventually you will see these bad readings under your normal wind speed and average wind readings as well. The only difference between the gust and wind speed is that the bad gust value is sticking around for 10 minutes, increasing the probability of being recorded at the weather servers.
      > > > >
      > > > > As far as CWOP / MADIS goes, I am not sure what criteria that they use to determine bad packets. If I had to guess, they will look at the gust reading to make sure it is as least as large as the wind speed being reported.
      > > > >
      > > > > Are the majority of bad readings above a particular speed? For example, I have mine set to 45mph. Rarely do I see a real 45mph wind but I do see plenty of higher bad wind readings:
      > > > >
      > > > > Fri Oct 30 07:16:09 2009 > Wind speed reading of 114.1 mph detected. Value out of range (0 to 111.8).
      > > > > Fri Oct 30 07:16:09 2009 > Wind gust speed reading of 114.1 mph detected. Value out of range (0 to 111.8).
      > > > >
      > > > > I see those almost every night. Although they are considered out of range, the 45mph limit would have rejected them if they would have been in the valid range.
      > > > >
      > > > > Now if you set your reject limit to something low, say 25-30 mph, do you continue to see random bad readings below that limit? If so, I am not sure how a range will help if the readings are truly random.
      > > > >
      > > > > If there are very specific ranges then adding the range option might work. I am hesitant to invest the time to make the changes if in the end the ranges will not adequately capture the bad readings.
      > > > >
      > > > > An easier option for you might be to force gust to always be the same as the current wind reading. As I mentioned in the first paragraph this only reduces the probability of bad values being recorded, it does not eliminate the problem.
      > > > >
      > > > > Currently there is an attempt to catch bad wind speed readings automatically by using a entirely different method.
      > > > >
      > > > > // These constants are used to detect bogus wind speed readings.
      > > > >
      > > > > // Look for readings above this limit in mph.
      > > > > #define QUESTIONABLE_WINDSPEED_LIMIT 25.0
      > > > >
      > > > > // If average wind speed prior to this event as <= this number, revert to last wind speed sample.
      > > > > #define AVERAGE_WINDSPEED_THRESHOLD 4.0
      > > > >
      > > > > What these constants define is the average wind speed readings over the past 10 minutes, and the "questionable" wind speed limit.
      > > > >
      > > > > For example, if the average wind speed is < 4.0 over the past 10 minutes, then a reading above 25mph is detected, this wind reading is considered suspect and is rejected and replaced with the average over the past 10 minutes. This is applied to normal wind and gust readings. An entry is posted to the log when this happens:
      > > > >
      > > > > "Wind gust speed reading of X.X mph detected when wind is only gusting to Y.Y mph (over past 10 minutes)."
      > > > >
      > > > > I am wondering if in your case, if I put these values on the tweak screen so that you could change them, you might be able to tune them in for your particular problem.
      > > > >
      > > > > Let me know if you think you might be able to use that method over the others.
      > > > >
      > > > >
      > > > > --- In wuhu_software_group@yahoogroups.com, "jjantti2" <jjantti2@> wrote:
      > > > > >
      > > > > > Hi.
      > > > > >
      > > > > > I'm aware of the fact that WS-23xx series stations do not send separate gust values.
      > > > > >
      > > > > > At first I tried to add the erroneus values to "reject known bad gust readings" field, but that proved as a neverending project since the values were not the same usual values.
      > > > > >
      > > > > > Example: I see an erroneus value of 105.0 as a gust reading. So put the value to the bad readings list. Then, another erroneus value kicks in; this time it's 106.7. So put that in the list too. Then, a thrid value comes along at 101.1. Putting that in to the list as well. And then arrives the fourth value of 105.2. Since it's 105.0 that's in the list, 105.2 passes through as a good value.
      > > > > >
      > > > > > To clarify things more (hopefully), what I had in mind is a possibility to have an option in WUHU that zeroes out all wind gust values regardless what is sent from the station.
      > > > > >
      > > > > > In this particular case (WS-23xx), the gust value is calculated from the normal wind speed value over a set period of time and as such, is not really a separate value. With the option enabled, the gust values are not calculated and only a gust value of 0 is reported instead.
      > > > > >
      > > > > > For stations that do report a separate gust value (like WS-36xx), this option, when enabled, would simply ignore the reading and report a gust value of 0 instead.
      > > > > >
      > > > > > Having this option would give the station operator the possibility to overcome a situation similar to mine where normal wind speed is reported by the station but gust values are off the scale. One could keep this option enabled until the situation with the wind sensor is rectified.
      > > > > >
      > > > > > There may be a caveat here. Since I'm not an expert in dealing how the wind data should look like when it's sent to WU, CWOP and others, I don't know what will happen, if the receiving end gets a wind data that has continous wind speed of x and gust value of 0: Will that be counted as a legitimate wind reading or simply ignored as an invalid value?
      > > > > >
      > > > > > If this option is not a good idea, then comes "Plan B": How about the possibility to add a multiple ranges of bad wind gust values to the "reject known bad gust readings" field instead. Would that prove as a better alternative?
      > > > > >
      > > > > > Best Regards,
      > > > > > Juha Laiho
      > > > > > Wx Station Operator
      > > > > > Helsinki, Finland
      > > > > > ---------------------
      > > > > > CW4473 / IUUSIMAA2
      > > > > > DW3039 / IUUSIMAA8
      > > > > >
      > > > > >
      > > > > > --- In wuhu_software_group@yahoogroups.com, "wuhu_software" <wuhu_software@> wrote:
      > > > > > >
      > > > > > >
      > > > > > > Juha,
      > > > > > >
      > > > > > > The problem is that there is only a single wind reading that comes from the console, there is no separate gust value. You can eliminate gusts altogether (set it to wind speed for example, or even zero), but then you are just uploading bad wind reading values without bad gust readings. Since WUHU cannot control which records are stored at the server, the problem continues.
      > > > > > >
      > > > > > > There is no way to look at the wind data from the console and determine if it is a good reading or not.
      > > > > > >
      > > > > > > The only way to deal with these bogus readings is to cap (max allowed reading) them or to take out specific readings (reject known bad readings).
      > > > > > >
      > > > > > > In order to accept some readings and reject others, you have to establish some sort of criteria to do so.
      > > > > > >
      > > > > > > Apparently I am not understanding what your idea is.
      > > > > > >
      > > > > > > Thanks.
      > > > > > >
      > > > > > >
      > > > > > > --- In wuhu_software_group@yahoogroups.com, "jjantti2" <jjantti2@> wrote:
      > > > > > > >
      > > > > > > > Hi.
      > > > > > > >
      > > > > > > > Yes, that's precisely what I thought was going on, since I remembered only WS-36xx stations have that feature. What's funny is that I still get normal wind speed readings, but the gust values tend to go off the scale, without the wind speed ever reaching that off the scale value.
      > > > > > > >
      > > > > > > > I was asking this particular "disable gust reporting" feature to make the wind speed readings look normal on all weather services (FinWX, WU and CWOP) until this situation can be rectified. I'm not too keen on getting any converters or signal amplifiers to overcome this situation, since I believe this could be accomplished more easily by software.
      > > > > > > >
      > > > > > > > Can you add this feature to some future refresh of WUHU? It would be highly appreciated.
      > > > > > > >
      > > > > > > > Best Regards,
      > > > > > > > Juha Laiho
      > > > > > > > Wx Station Operator
      > > > > > > > Helsinki, Finland
      > > > > > > > ---------------------
      > > > > > > > CW4473 / IUUSIMAA2
      > > > > > > > DW3039 / IUUSIMAA8
      > > > > > > >
      > > > > > > > --- In wuhu_software_group@yahoogroups.com, "wuhu_software" <wuhu_software@> wrote:
      > > > > > > > >
      > > > > > > > > Juha,
      > > > > > > > >
      > > > > > > > > The Gust values are really just the highest wind speed seen over the
      > > > > > > > > past 10 mins on a WS-23XX station as the station does not report a
      > > > > > > > > separate wind speed and wind gust value.
      > > > > > > > >
      > > > > > > > > I am wondering if the voltages between the remote thermo and the sensor
      > > > > > > > > are TTL level (5v?) If so, my guess is that there would be some type of
      > > > > > > > > "extender" out there to handle that distance. Has anyone investigated
      > > > > > > > > the signals on a scope? I am pretty sure that the remote thermo supplies
      > > > > > > > > the GND, 5V (or 3.3V), and the sensor transmits serial data back to the
      > > > > > > > > remote thermo.
      > > > > > > > >
      > > > > > > > > I am thinking something like these:
      > > > > > > > >
      > > > > > > > > http://www.commfront.com/TTL-RS232-RS485-Serial-Converters/RS232-TTL-Con\
      > > > > > > > > verter.htm
      > > > > > > > >
      > > > > > > > > http://www.commfront.com/TTL-RS232-RS485-Serial-Converters/RS232-TTL3.3V\
      > > > > > > > > -Converter.htm
      > > > > > > > >
      > > > > > > > > Just to give you an idea of what type of device I am talking about.
      > > > > > > > > There may be something cheaper available out there, but you get the
      > > > > > > > > idea.
      > > > > > > > >
      > > > > > > > > Note that the TTL/485 converter can go 4000ft.
      > > > > > > > >
      > > > > > > > >
      > > > > > > > >
      > > > > > > > > --- In wuhu_software_group@yahoogroups.com, "jjantti2" <jjantti2@>
      > > > > > > > > wrote:
      > > > > > > > > >
      > > > > > > > > > Hi all.
      > > > > > > > > >
      > > > > > > > > > My second weather station (La Crosse WS-2308) has a wind sensor
      > > > > > > > > installed in a "long-wired" configuration. What this means is that the
      > > > > > > > > standard wind sensor wiring is extended using two 10m/~30ft flat phone
      > > > > > > > > extension cables using line extenders (those blocks with female RJ11
      > > > > > > > > sockets on both ends of the block), giving a total of 30m/~100ft of line
      > > > > > > > > lenght from the sensor to T/H sensor.
      > > > > > > > > >
      > > > > > > > > > Needless to say this is not the optimal configuration. Due to the
      > > > > > > > > nature of the cable the tolerance to any interference is nonexistent and
      > > > > > > > > has resulted in hurricane-force wind gust readings usually during the
      > > > > > > > > evening hours (why only then remains unknown to me). Despite these crazy
      > > > > > > > > wind readings the continous wind readings are all fine.
      > > > > > > > > >
      > > > > > > > > > I've tried to contain these wind gust readings by using the "Reject
      > > > > > > > > all wind speed readings above XX MPH" but with little or no luck. I've
      > > > > > > > > set the MPH limit to 36 MPH, but instead of speeds getting slashed to
      > > > > > > > > zero (or given an averaged value) the gust speed is now slashed at 36
      > > > > > > > > MPH and are posted in that particular value.
      > > > > > > > > >
      > > > > > > > > > Gust values don't seem to have any static values, so typing them all
      > > > > > > > > to the "ignore these gust values" list in WUHU is out of the question.
      > > > > > > > > >
      > > > > > > > > > It doesn't hurt me in any way to have the gust values removed since
      > > > > > > > > the wind sensor is reporting continous wind speeds without any problems.
      > > > > > > > > >
      > > > > > > > > > So I decided to turn to you and ask if there is a possibility to
      > > > > > > > > disable all wind gust reporting completely but still retain the
      > > > > > > > > continous wind speed values. Can this feature be added to WUHU?
      > > > > > > > > >
      > > > > > > > > > Best Regards,
      > > > > > > > > > Juha Laiho
      > > > > > > > > > Wx Station Operator
      > > > > > > > > > Helsinki, Finland
      > > > > > > > > > ---------------------
      > > > > > > > > > CW4473 / IUUSIMAA2
      > > > > > > > > > DW3039 / IUUSIMAA8
      > > > > > > > > >
      > > > > > > > >
      > > > > > > >
      > > > > > >
      > > > > >
      > > > >
      > > >
      > >
      >
    Your message has been successfully submitted and would be delivered to recipients shortly.