Vantage Console Firmware Release History B.P. 1/8/03

 

 

Version History

 

REV B January 8, 2003

            Corrected

            -------------

Elevation entered in meters was not converted correctly.  3.048 ft/meter was used instead of the correct 3.281 ft/meter.

 

The transmitter reception diagnostics did not account for missed packets when the console was in the “L” state. 

 

Average solar rad and average uv at 1 minute intervals dashed occasionally if no packets were received in that interval.  Now, if no packets are received in the 1 minute interval for one or both of those two variables the current value of that variable is used for the average.

 

The high rain rate value for an archive interval would sometimes be 0 if no packets were received in the interval (generally at 1 minute intervals).  Now the current rain rate is used for the high in this case.

 

 

Changes

-----------

When setting the transmitter type in the setup screen users where accidentally changing the transmitter period by pushing the Hi/Low key.  Now, the transmitter interval changes on the Hi/Low key press only if the Temp key is down also.

 

The reality filter for rain rate was changed from .09 in/hr to .35 in/hr to improve our ability to capture the high rain rate. 

 

 

REV B August 12, 2002                   (Updater Only)

            Corrected

            ---------

Zero threshold for UV was raised from .3 index to .5 index.  Any readings < .5 index are forced to 0.

 

Previous code did not allow a user to run the newer Leaf/Soil station (6343) with the earlier Soil Moisture station. 

 

 

REV B April 24, 2002

            Corrected

            ---------

The user could enter up to .25” for rising and falling barometer alarms, but only worked up to .12”.  The alarms now works up to .25”.

 

Fixed problem entering altitude in meters.  Can now set altitude up to 4572 meters (15,000’).

 

Old code did not correctly handle rain packets coming from a Soil station.  The packets are used to communicate low battery but should not of been processed as rain packets.  Now they are not.

 

Display of outside temps between 0 and .9 in tenths was changed from .0 to .9 to 0.0 to 0.9.

 

The legend for the 15 minute rain graph in mm was off by 10.  For example, it would  show .01 when it should of shown .1.

 

The “low battery on station” message would not show up for a Temp or Temp Hum station run as an ISS in the setup.

 

The “low battery on station” message could be cleared inadvertently for an ISS run as a Temp or Temp Hum in the setup.

 

            In the High/Low screens the flashing graph point sometimes did not change to the new position right away when you switched to a new graph.

 

Changes

            ---------

Added support for the new leaf/soil board.  You can pick leaf/soil for a tranmitter type in the setup.

 

            Leaf and soil packets will not be recognized if a leaf, soil, or leaf/soil station is not defined in the setup.

 

Rain alarms are checked every hour in addition to every click.  Previously, the user had to clear the rain day alarm, but now it will clear when the last 24 hour total is less than the alarm threshold.

 

REV B firmware uses a new Weatherlink record.  Added high UV and high Solar Rad in the period.  Number of leaf wetness sensors logged is 2.  Increased extra temps by 1 to 3 with 52nd byte that wasn't being used.  Added record type which will be 0 now - old record was 0xff.  Now only measured rain rate is logged.

 

NEW RECORD

struct StandardRec

{

   unsigned dateStamp;           // Year | Month | Day   7 bits | 4 bits | 5 bits

   int timeStamp;                    // Hour * 100 + Min

   int tempOut;

   int hiTempOut;

   int lowTempOut;

   int rain;

   int rainRate;

   int bar;

   int solar;

   int nWindSamples;

   int  tempIn;                  // .1 degrees F

   byte humIn;

   byte humOut;

   byte windSpeed;

   byte hiSpeed;

   byte dirOfHiSpeed;

   byte domDir;

   byte uv;

   byte et;                      // .001 in

   int  hiSolarRadInPer;

   byte hiUVInPer;

   byte forecast;

   byte leafTemp [2];

   byte leafWet [2];

   byte soilTemp [4];

   byte recordtype;

   byte extraHum [2];

   byte extraTemp [3];

   byte soilMoist [4];

};

 

OLD RECORD

====================================

struct StandardRec

{

   unsigned dateStamp;           // Year | Month | Day   7 bits | 4 bits | 5 bits

   int timeStamp;                // Hour * 100 + Min

   int tempOut;

   int hiTempOut;

   int lowTempOut;

   int rain;

   int rainRate;

   int bar;

   int solar;

   int nWindSamples;

   int  tempIn;                  // .1 degrees F

   byte humIn;

   byte humOut;

   byte windSpeed;

   byte hiSpeed;

   byte dirOfHiSpeed;

   byte domDir;

   byte uv;

   byte et;                      // .001 in

   byte soilTemp;

   byte soilMoist [4];

   byte leafTemp [4];

   byte leafWet [4];

   byte extraTemp [2];

   byte extraHum [2];

   int reedClosed;              // reedClosed

   int reedOpen;                // reedOpen

};

 

Bar trend is now stored where "P" was stored in loop record. 

 

            If no rain was recorded in the archive period, the hiRainRate logged is the rainRate at the end of the period.  This should more accurately reflect the decaying rain rate.

 

 

REV A Oct 25, 2001

            Corrected

            ---------

            The low dew point graph for the month did not draw the points in the correct horizontal position.  You could still view the low dew point values for past months in the high low screen however.

 

            The high monthly and yearly dew point, heat index, and thsw index will be dashed unless the user does an “all clear” when he first sets up the station or clears them manually from the high low screen.

 

            Temperature correction for soil moisture was backwards.  The tempco is negative with a magnitude of 1% for every degree away from 75°F.  The reading is too high at low temps and too low at high temps without correction.  Our correction accentuated the error instead of reducing it.

 

            For Manufacturing

            -----------------

            When time has not been entered the clock was set to 12:00 pm, but in the burn in it is better if a new day does not occur in 12 hours so the new default time is 12:00 am.

 

 

REV A Oct 11, 2001

            Corrected

            ---------

            The 2 hour archive interval archived data every minute every other hour.

 

 

REV A Oct 8, 2001

            Corrected

            ---------

            The graph legend for bar in mm was 10x what it should have been.

 

            The 15 minute bar graph in mm, hpa, or mb incorrectly showed the inch legend too.   You could see more than one “1” sometimes or an extra “.”.  Note, the default graph is 1 hour so people have to zoom in to find this problem.

 

            We discovered not all consoles could detect a power adapter missing (every try) when the lamps were on.  The new code requires over 750 counts to detect a power adapter is present.   Also, we will change the EXTV voltage divider network to 10k’s instead of 1M on future console boards which will also fix the problem.

 

Added

            --------

            Interpolation for wind speeds of 4, 9, and 15 on large cups.  For example, when speed goes from 3 to 5 we show a 4 first and vice versa.  Note, small cups have same problem and will not show 3, 5, 12, 19 for speeds less than 20.  No code space left to help out small cups.

 

 

REV A Sep 7, 2001

            Corrected

            ---------

You can now enter negative numbers for elevation.

 

First download after buffer wrap required a second download to get most recent data. 

 

High heat index times for the day were not saved correctly.  If you had solar rad the times displayed were for THSWI and if not, they were dashed.

 

The THSWI high times for the day were not saved at all.

 

If power was removed after entering the Setup screen you could still lose up to 4 of your most recent Weatherlink records.  The most recent 0 to 4 records in the Weatherlink are stored in FLASH RAM which is volatile.  These records are now saved to FLASH every time you enter the Setup screen so no data is lost when power is removed.

 

The code version with a single digit day scrolled with a hitch on boot up because more than one space separated two words (ie. Sep  9).  Now strings with more than one space between words will scroll correctly although I don’t think we had any other than that code revision string.

 

            Added

            --------

            New NWS wind chill calc.

            THSWI modified to take into account the new wind chill formula.

 

 

            Removed

            --------

            A receive debug mode that you could go into when in the first Setup screen by pressing the RAIN key.  No longer needed or used.


 

 

REV A July 18, 2001

            Corrected

            ---------

Archive record for dominant direction would always choose North if there was no wind speed in the period.  Now a 255 is written to the record indicating dashes to the software.

 

            Added

            --------

User now has a choice of logging average inside and outside temp, or  there value at the time of the archive.  Setting EEPROM byte 4092 to 0 will cause the temps to be averaged.  A nonzero value will sample them at the time of the archive.

 

REV A June 19, 2001

            Corrected

            ---------

            A small number of the longest forecast strings would not scroll correctly. 

 

 

REV A May 31, 2001

            Corrected

            ---------

            If  you remove the power adapter, the lamps still operated as if the

power adapter was plugged in.

 

            The Hum Cal did not work properly.  Although it works for humidity,

the calculated values of dew point, heat index, and thswi do not use the

corrected value.   

 

            When scrolling back days the UV med totals were 1/2 of what they

should be. 

 

            Forecast improvement for hot weather and for California.  If the

current temp is greater than 87, it will never forecast rain.

 

            Comments in the status area which can stay a long time now only come

up when the appropriate weather variable(s) are selected.   The new conditions

for each message are shown below.

 

      if ( wind10Min >= 25 && wind10Min < 40 && currentGraph == 0)

                                              // HOLD ONTO YOUR HAT

         PushCommentNo (10);

 

      if ( wind10Min >= 8 && wind10Min < 18 && currentGraph == 0 &&

           windChill < 80 && windChill > 60 && GetRainStorm (CURRENT, 0) == 0

&& LightOutside ())

         PushCommentNo (9);                   // KITE FLYING WEATHER

 

      tmp = GetCurrentVar ();

      if ( windChill <= -30 && windChill > -50 && wind10Min >=10

              && (tmp == WIND_SPEED_VAR || tmp == CHILL_VAR || tmp ==

TEMP_OUT_VAR))

           PushCommentNo (1);                // LOW CHILL ADVISORY

 

      if ( heatIndex >= 105 && heatIndex < 115 &&

(tmp == TEMP_OUT_VAR || tmp == HUM_OUT_VAR || tmp == HEAT_VAR || tmp ==

THSW_VAR) )

         PushCommentNo (5);                      // HEAT ADVISORY

 

 

            May 15 code always returned 33% for inside hum.        (1 unit shipped and has been recalled)

 

 

REV A May 15, 2001

            For Manufacturing

            -----------------

            You can now turn on all segments in the first Setup screen and the

receive diagnostic screen by pressing 2nd and then the "+" key.

 

 

 

REV A May 4, 2001

            Corrected

            ---------------

            LOOP command at lower baud rates gave CRC errors when used with

Windows Weatherlink software.

 

            Kite flying weather message now only comes up during daylight hours.

 

            All versions below this line were beta versions. 

===============================================================

 

 

REV A  April 30, 2001

            Corrected

            ---------------

            Inside Temp alarm would not go off once alarm condition ceased.

           

            For Manufacturing

            --------------------------

            CRC error variabale is now cleared each time you do an RXTEST.

 

 

REV A  April 11, 2001 15:59:42

            Corrected

            ---------------

            Display memory not updated after download in one infrequent case.  LCD

momentarily shows incorrect random segments on.

            Fixed same problem in Setup screen.

            Now account for daylight savings transitions for sunrise and sunset

times.

            Diagnostic screen shows "---" when batteries drop below 3V.  Can't

really tell what it is at < 3V.

            Low battery moved to 517 counts (3.0V) for new 3V regulator.

            Lamps shut off if battery voltage drops below 517 counts (3V).

            Soil moisture overflowed when sensor was very dry and showed 0.

      Changes/Fixes made for PC software.

      ----------------------------------------------

            Now clear 10 minute wind speed alarm bit when checked if alarm is not

set.

            Same for high speed alarm.

            Removed unused commands SETA and GETA.

            Added PutRain (0) after EEWBR 0x82 write so rain alarms on console

would update right away.

      For Manufacturing

      --------------------------

            When on initial test fixture the baud rate in the EEPROM is

ignored and 19,200 is used - caused many initial test failures (ERROR 1).

 

 

REV A March 15 2001 7:35:51

            Corrected

            ---------------

            High rain rate for year not cleared on All Clear.

            UV not always zero at night.  (< 0.3 now zero)