The Weather Wizzard III - for APRS?



I like this station. The old  Davis 7425 Weather Wizard III still keeps on going.   It hangs on the wall in the family room,  displaying the local weather info.

The two projects I always wanted to do with this station is to 1) add a rain gauge, and 2) build a "self-contained" APRS station.  I got my station in 2003 on sale at Fry's Electronics- but even though this little station has been discontinued and is no longer available form Davis, but you can still find it on line.

RAIN!

It has always irritated me the standard 7425 does not come with a rain gauge...  The easy solution to problem  #1 is just buying:

Davis Instruments 7852 Rain Gauge for Davis Weather Wizard & Weather Monitor Stations:

Is it the only solution?

The add on cost is $60 ~ $70 USD.  Since I got my 7425 on sale for $89, it seems a bit odd ~ to my sensibilities anyway ~ that adding this sensor is a significant fraction of the cost of the whole setup.  I think I found a way to get this sensor for 1/2 the cost!

Recently, we re-painted the house and one of my pet peeves is painting cables (don't ask - it's an OCD  thing I think is what  they call it) so I disconnected the sensors to paint.  The Junction box is in a patio area and a pass through hole terminates in a surface mount Wire Mold J-BOX inside makes a nice weather- and mouse-sealed  pass thru for the cable.  ...oh yeah, and wife approved as a family room accessory!

When re-installing the cable thru the hole the RJ-11 plug got hung up in the insulation and became damaged as I fished it through.  The Wind Direction ceased to work on the display. This prompted me into a little investigation...So what mysterious protocol gets the sensor data to the station?   What no serial data to hack? No, they are very simple: analog!

Specifications

Rain Sensor General

Sensor Type . . . . . . . . . . . . . . . . . . . . . . . . . . Tipping bucket with magnetic reed switch

Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Contact closure

Attached Cable Length . . . . . . . . . . . . . . . . . . 40’ (12 m)

Cable Type . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-conductor, 26 AWG

Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . .Modular connector (RJ-11)

Resolution . . . . . . . . . . . . . . . . . . . . . . . 0.01” (0.2 mm)

Input/Output Connections

Black & Red . . . . . . . . . . . . . . . . . . . . . . . . . .Switch terminal

Green & Yellow . . . . . . . . . . . . . . . . . . . . . . . .Switch terminal

Temperature Sensor General

Sensor Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Platinum wire thermistor

Time Constant

In Still Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 seconds

In Liquid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 seconds

Attached Cable Length . . . . . . . . . . . . . . . . . . . . . . . . . . 25' (7.6 m)

Note: There is no absolute maximum cable length. Increasing the cable length above 300' (90 m) causes an increased measurement error at

a rate of approximately +0.06°F (+0.03°C) per 100' (30 m) at 136°F (60°C) and +0.012°F (+0.006°C) per 100' (30 m) at 77°F (25°C).

Cable Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-conductor, 26 AWG

Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modular connector (RJ-11)

Input/Output Connections

Black & Red . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Common

Green & Yellow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Temperature (variable resistance to common); 10KOhm, nominal

Wind Sensor General

Sensor Type

Wind Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wind cups and magnetic switch

Wind Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wind vane and potentiometer

Attached Cable Length . . . . . . . . . . . . . . . . . . . . . . . . . . 40' (12 m)

Note: On Monitor and Wizard stations, cable lengths longer than 140' (42 m) between sensors and console may artificially limit wind speed readings. That

is, beyond that length, maximum recordable wind speed decreases as cable length increases. For example, with a length of 140' (42 m), the maximum

recordable speed exceeds 175 mph. At 240' (72 m), however, the maximum recordable speed drops to less than 140 mph. Below that upper

limit, however, the anemometer's accuracy is not affected.

Cable Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-conductor, 26 AWG

Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modular connector (RJ-11)
Input/Output Connections
Black. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wind speed contact closure to ground
Green . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wind direction pot wiper (360° = 20 kOhm)
Yellow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pot supply voltage
Red . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ground

Based on these specs form Davis I came up with the following schematics:

So what about that rain sensor?  It is just a reed switch and magnet. There are plenty of tipping bucket sensors on the market for less than $70! that use these. For example:

16095  Acu-Rite 00899 Wireless Rain Gauge
Price: $34.99

http://www.klockit.com/products/dept-64__sku-16095.html

OK, so it's not the best one we ever saw.  Typically, the made in China electronics in these types of products will likely fail and stop working in most of these after a month or so - and the plastic is probably not UV stabilized like the Davis 7852 so it may not be the best for harsh environments but  it is readily available and - oh yeah, I have a broken one!

I got mine from the Home Depot to measure the irrigation sprinkler watering on my newly sodded lawn.   It has a reed switch and magnet...which attaches to a transmitter (you know...the part that no longer works).  I think the main failure mode of these is the soldering of the circuits, not the switch and magnet assembly. I think the soldering is done by hand and I am sure not much time is allowed for the assemblers for each unit...

This is an easy hack!  Disassemble, Rip out the transmitter, get an RJ-11 cable, solder the RED & BLACK wires to one side of the reed switch and the GREEN & YELLOW wires to the other side (or what ever colors you have), re-assemble, and ..done! 

...Except for calibration. The interesting thing is the specs say 0.02" resolution but the buckets look the same as the Davis' buckets. It may be the display waits for two tips to register a count or the buckets take a little bigger drop to tip.  This Acu-Rite model happens to come with calibration screws and a method to calibrate...so making it match the 7425 should not be an issue, in fact we can easily re-scale in the count in my APRS_WX modem firmware as well if needed even if we can't fix value displayed.

BTW-- They no longer carry it in Home Depot, I found these specs for mine online in a clockmaker's catalog.  Does the successor's sensor look strangely familiar? Yes, it looks exactly like the $99 sensor hummmm maybe I ought to check it out?

...and from the comments it too stops working after a few months!

The Weather Channel Wireless Rain Gauge

Price: $24.99

Model # WS-9005TWC-IT 

Internet # 100599154

Ships FREE with $45.00 Order

Strangely, This item cannot be shipped to the following state(s): AK,HI

http://www.homedepot.com/buy/outdoors/outdoor-living/the-weather-channel/wireless-rain-gauge-6153.html

APRS!

The major problem with #2 is getting the weather data out of the console.  The unit does send data serially but, again it has a cost. ( no, I am not cheap, really.)  The only way I saw to do this for years was to add the Davis Weatherlink For Windows 7862 software package (~$150) which comes with a little interface device called the "Data Logger"  So... since I don't want a PC in the loop the software would be useless to me anyway in my APRS setup. 

Well, useless after the command protocol was hacked ~ but that would be simple matter with a seral data analyzer snooping at the comm port on the PC as it interfaced with thier software. But as it happens, the command protocol that was not readily availble is now ~ Davis has released their SDK for Weatherlink! ☻ 
Because of this released Davis documentation, and some intrepid hackers ( thanks, DeKay )

http://madscientistlabs.blogspot.com/2011/01/davis-weatherlink-software-not-required.html

It has been also shown that this device is little more than a RS-232 level converter!  The keyword is; convert - you don't want the full ±12 Volts going to the weather station.  This APRS modem I designed a while back should be able to be easily modified to interface with this directly. The GPS dongle should still work normally as well.

This PCB  was developed from these schematics a few years back, what I did was take an existing design, modified slightly and made a PCB.  In fact,  I am currently using two of these boards in my APRS stations.  It works great and is very compact.  ...and I own the source code!  Yes, I will post it.

The two connectors on the right are on a snap off option board I made when I fabricated this PCB to allow the GPS info to get tapped into, so I could add a remote GPS display. It was free due to the small size of the TNC board ( 2" x 3") and it provides a sturdy interface to the GPS device. ( OK ~ maybe let's just say I am frugal?)

As you can see from the schematic, it has a built in level converters for RS232 so we have access to both the ±12V and +5V sides.  ( Note: This circuit can also run @ 3V)  This circuit board also contains all the components and software for a complete self-contained APRS modem. 

I may revise the board at some point to allow for an internal temperature and battery voltage (for telemetry) as well as barometric pressure, and relative humidity sensor interfaces on-board to fill in the missing WX data.  I also will probably change to a better connector vs. the audio jacks I originally used. BUT - I still have two of these boards waiting to get built!

All we have to do is modify the FIRMWARE to give the correct commands, add a couple of jumpers and connector for low-voltage serial data (the RJ9 4-pin connector which interfaces to the 7425 seems common but finding one that fits and has all 4 conductors is a bit of an issue as I found out), then receive the raw WX data from the station, format the WX data unto APRS packets, and shazam! APRS-Wx packets start flying!  More to follow on this project as well...STAY TUNED!

~ ~ ~

$99 Weather Station - Working

Rain event 0.21" recorded. 2/3 - 2/4 2011

Added some bracing material ~ sorry Ms. Spider, I had to relocate your nest so that it would not get glued! 

Wind instrument is semi-permently still attached to the TV mast:

The Davis wind instrument is still chugging away:

Here are some epic wind gusts recorded by it:

42 MPH 04:47 01/27/08
49 MPH 01:29 01/29/08
32 MPH 08:55 03/04/08
35 MPH 04:19 12/15/08
34 MPH 08:48 01/10/09
29 MPH 11:19 01/29/09
45 MPH 13:34 02/09/09
26 MPH 12:19 03/22/09
32 MPH 13:11 04/15/09
23 MPH 04:38 08/21/09
41 MPH 04:16 12/07/09
55 MPH 14:19 01/19/10
46 MPH 07:25 01/21/10
34 MPH 07:53 02/27/10

Snow level: 1000FT

Every few years we see some snow in So. Cal.  Not that it was a huge blizzard but worth noting I guess:

 It rained most of the day on 26 Feb 2011  but after about 10:00PM local time, real snow started falling!  We are at about 1650 FT above sea-level here.

Mt Woodson

 We have seen this before but it is unusual to have the level drop so low and stil have a dusting in the morning.  This is Mt. Woodson just to the west of our house. It is about 2000 Ft. AMSL.

Mt Laguna - East of San Diego

 Mt. Laguna to the east, from my front yard.  It is not unusual to see snow up there this time of year, but to see white all the way down into Ramona is very cool.

Snow Tastes Good!

Well, it tastes cold anyway!

Sue and Johhny Deere

What is going on here dad?

Just a dusting here, snow on the palm trees! The temps were back up to the 50's by afternoon and all traces down here were gone. 

$99 Missed Rain Event #2

OK ~ so we don't get many weather events here in So Cal.  and this rain guage works great...as long as it does not rain!  The first part of this weekend's storm  it properly reported ~ .12-Inches. 

The main rain came Friday night and it was reading 0" since midnight ( it had been steadily falling throuought the night) Again, I noted a large ( +3" ) reading on the previous day's graph.  What is going on here? 

OK I have a theory now!  When I went back up on the roof and inspected ~ this time no spiders would be spared~  I noticed some interesting things.  First the RG was still there and not blown over in the wind as I feared. Second it did not seem to be sitting level? 

I built a mall leveling platform out of a small rectangle of plywood, with another strip at one end to prop up the platform, since this is on the  "flat roof" part over a porch the slope is very small and needed only a 1/4" - 1/2" rise on one side to make it level. After I built this platform I primed and painted it and screwed the RG into it. It is not attached in anyway to the roof.

So here is what I think is going on:  As the platform gets wet, the bottom side stays wet and the top side dries causing the board to warp. As it warps, it finally starts to move the tipper so it is at a critical angle that causes the sensor to begin flickering causing the big totals.  This goes on until the angle become so great the tipper will no longer move past the sensor.

It gets worse...The gauge had some water in it, AND it no longer transmits as well.  SO here is what may have happened:  Since the tipping bucket was now shifted to one side the rain was deflected along it in a fully tipped position, some of the rain did not properly exit the device and accumulated inside it.

I will dry everything out and make a new platform.  I thought the paint would protect the wood, but looks like it needs another type of material or better protection.

$99 Weather Station..."Part 2 - Deux "

I noticed that after the rain reset incedent, the outdoor temp was not cycling...oh yeah, I tried a re-scan, you need to push the channel switch, ahhh there it is!  Once more to have it cycle indoor and outdoor...all is well! 

Hey wait, how did i miss that the outdoor Temp has a RH% as well!  OK so I only need to read the pressure locally.   Ok, I confess the bi-lingual manual was sitting next to the console I would not have figured it out so fast - and- about the channel search: the manual says if  the device picks up a second channel, ignore it?  Interesting.

$99 Weather Station..."Part Deux"

I see they are now in the $80 range ...oh well for once I am an early adopter?  I am thinking that this was made for Celestron by LaCrosse.  It has that look and feel to it.  There are another bunch of cheap wx stations appearing in various catalogues.  I picked up a wireless Rain Gauge at the Home Despot ...for about 30 bucks when I put in my new lawn to monitor the watering  last year.  It stopped working after about a week.  

Anyway, I hope to see the familiar 44-bit packets as shown in  Jean-Paul ROUBELAT - F6FBB (a Fench Ham) post at : http://www.f6fbb.org/domo/sensors/tx3_th.php 

The wireless snoop will give me a second source of wind speed and direction, and an extra outdoor temp located in that sensor as well.  Pressure and RH%  are most likely locally measured in the console. 

The pressure and RH% were real easy to read on the PIC weather station using an A/D converter and a couple OpAmps.  I should have no problem porting these  to the new APRS WxServer. 

I am a little cautious of relying entirely on this station's wind sensor it seems more flimsy than the Dallas Semi 1-Wire, which started having issues after about a year of operation, the speed sensor would "stick" and when it broke free again it would sometimes give a huge false wind spike (50+ MPH) When I took it down I found that the plastic around the bearings simply went out of round causing the axial to wobble.

It will be a happy day if I see these as there is a Practical Adrino Post by Marc Alexander and Jonathan Oxer ~that should do most, if not all, of the heavy lifting on the decode! 

Celestron $99 Wx Station

The $99 Celestron wx station from OPT has been up and running since Sept.  I have been wanting a rain guage for the Davis but it was almost as much as this whole thing!  It has been working great. Until.... last weekend we had a small rain but I noticed the rain total was not counting.   Then I nioticed  a huge  rain total eariler in the week ( a huge High Pressure system with 8" rain we did not notice  ~I think not)  I went up on the roof and took a look... visions that it got blown over or something.

There was a huge spider living under the leveling platform but every thing seemed OK I tipped it a couple times as well.  I checked  the batteries and they were amost still new ( 1.49V) This guy must have been running  back and fourth on the tipper or a bird was trying to get him or something! Too funny.

Now I need to hack the wireless link to get rain and other data for the APRS. The beauty of wireless hack is the display gets the data as well and I don't have to come up with a fancy display.  Yesterday,  I   found a pretty interesting looking little receiver from Digikey that will likely be able to hear this simple RF stream. (simple = not spread spectrum) :

The  ALPHA-RX433S looks real flexible. It will need to be programmed via SPI ..no dip switches...but it will be hooked to the Arduino anyway ...and for $6 bucks?  Yeah, worth a try.  I also found a helical antenna... ANT-433-PW-LP  sure  a wire will do just fine for receive but this is too cool: (for $2 - why not?)  If this dosen't work maybe  I'll get the transmitter as well and will have a new toy to play with...sending data and stuff...

This station is probabily made by another manufacturer and  branded  by Celestron so this will most likely work with other wireless stations.  Stay tuned....

PS. No spider was harmed during this repair ~ Ilive and let live I say~  But I am pretty sure it was his web that caused the bucket to stop tipping.  but  the extra rain count is still a mystery....