Topic: remote APRS WX station project

[RonVE8RT]

I've come to the WXForum looking for guidance, mentoring, and more for a project I'm interested in building.
Because of our range of temperatures, which will go colder than -40 several times during our winters, commercial stations which blank out the temperature at -40 and colder will not do.

The station hopefully will monitor temperature (humidity is optional) wind velocity, rainfall, and have a second temperature input for ground level (or a sub surface permafrost temperature sensor).  Air quality would be a nice feature, especially in summer when forest fire smoke is a problem.

Power would be battery, solar or wind charged.  The station would be remote using a VHF APRS link (it could be tens of kilometres into the bush).   Battery power budgeting and management are a concern.

Looking for help to make this work.   Thanks!  Ron VE8RT

[Mattk]

I'll start by saying any gear running less than -40 measuring especially wind velocity, rainfall will require some serious de-icing (heater) capability and typically that will require 24VDC / 150-250 VA hence the power budget is going to be fairly high. Remote temperature probes to -40 are doable on some systems but typically the quality won't be found on hobby stations. Air quality is also something not commonly incorporated into basic systems.

What is your budget to achieve what you require including the APRS comms?               

[RonVE8RT]

The rainfall / snowfall wouldn't valued data other than from spring through fall.  The data I'd most value would be that gathered/reported during forest fire season and related to fire threat and impact.  That is when the air quality would be wanted.  Year round temperature and to a lesser degree humidity (I don't know if typical humidity sensors would work in the cold).  My decades old experience goes back to Stevenson screens with wet dray and wet bulb thermometers. 

The power requirements you've mentioned are not achievable 24/7.  In winter of course near the solstice we have about three hours of dim low angle sunlight (we may have high winds).  In summer during bad fire years solar is compromised by the persistent (has been July through to the end of September or longer) thick layer of smoke.

The quality and types of data would be that which may be useful for fire progress monitoring, and/or permafrost monitoring, those are my goals.  It would be nice to have a basic station that could be replicated by school students or Maker Space groups in small communities.

[RonVE8RT]

I'm at work and have a few minutes to answer your post.
Radios, to be determined.  Some handhelds have over the years worked well in the cold, I've used HTX202 for APRS and leave them in the car.  The displays may not like the cold but they work.  My Alinco DR-135T didn't work below the manufacturer's specified temperature range, some modifications I did extended the cold range a little.  The receive capability on the handhelds may be of no use, but they're affordable, obtainable, and easy (during winter) to cold test. Not wanting to, or able to I suppose, come up with something from scratch as this will end up as another never completed project. Using an antenna with some gain, and modes available on Direwolf such as FX.25 or IL2P, I'd like to do some range testing, hopefully will improve throughput and range.
Batteries, as there will be times when solar or wind will not be enough; I have a quantity of calendar life expired emergency locator lithium batteries.  They do work at temperatures below -40, as do the Energizer AA lithium cells in my trail cam.  The batteries packed in with the electronics should provide thermal ballast keeping everything from rapid temperature changes and extremes (inside a thermally insulated enclosure).
During winter the temperature, wind, and pressure are most important, I don't know how good humidity sensors are in the cold (haven't looked at a datasheet for one yet).  Air quality might be interesting if the station was located close to a zombie fire.
Additional summer data, it would be nice to also have air quality, and rainfall, and as a bonus a pyranometer (a rough indication of sunlight attenuation from smoke and cloud.
Maybe to start with, and add sensors later, a very basic WX station adapted to monitor sub surface temperatures in areas where permafrost melting is, or is expected to be, a concern.
The hardware is within my range of skills, interfacing, microprocessors, and programming are not.

[TheBushPilot]

RonVE8RT,

A few things I ask myself before starting the design process of a weather station:

• Cost
• Performance requirements
• Power
• Battery Capacity

Given where you live and the requirements at hand, it sounds like you need something quite substantial.

From what I understand you would like this site to operate year round. It will be subject to temperatures at or below -40°C/F each winter on multiple occasions. And you would like it to report via APRS.

Probably go commercial instrument route if budget isn't an issue. To keep things simple, get an R. M. Young alpine wind monitor, temperature/relative humidity sensor (-50 to +50°C measurement range), Setra/Vaisala/Young barometer, and Texas Electronics or Young tipping bucket rain gauge. Apogee Instruments or Eppley Laboratory solar radiation sensor. Soil temperature can be measured with an epoxy sealed thermistor or PRT from Campbell Scientific or Apogee Instruments. Campbell Scientific data logger, more recent CRBasic variant that can be calibrated for a certified operating temperature range of -55 to +85°C.

Not sure what the current draw of the radio would be but optimizing the station battery usage could get it down to mere milliamps. Get a pair of 250Ah sealed lead-acid batteries and the capacity alone assuming no source of recharge would get you through the winter. 200 watt solar panel with an MPPT charge controller.

I know my suggested way may be the costliest and most involved but it would probably last the longest in those conditions. Silver lining here is you can often find all of these instruments on eBay for a fraction of what they cost originally. Most expensive on that list being the data logger.

Hope this helps.


Cheers

[RonVE8RT]

This is helpful.  250 AH battery or two might keep it from being blown over or some curious person trying to cart it away

I was given, unexpectedly, a salvaged HOBO battery powered weather logging station (ran on a set of AA cells).  An amateur radio friend in Norman Wells saved it from being scrapped and put it on the C46 back haul, I didn't know about it until they unloaded the plane.  It has a wind speed (no direction) sensor, temperature, humidity, rain, and ground temperature sensor, pyranometer but not barometric pressure.  Somewhere I downloaded Onset HOBO information a few years back.  At the time I wanted to replace the data logging, or keep it and use the real time output so it would interface to a modem & radio for APRS.  I got hung up on the interfacing part and the project stalled.  It is well constructed and had been in use, maybe for years.

The project as you've described would be the ultimate goal, especially if it was taken up as a school project and could get some financing.  I'm interested in working as a volunteer post retirement (should be retired already) with schools and communities if I can be of service.

I'm taking the recommendations of assembling a higher end station seriously, although what I thought I wanted was a lightweight inexpensive hobbyist weather station.  The hobby version would serve me well maybe as an educational platform that stayed close to home.  For the long run though, anything funded and expected to perform, the higher budget more robust station will be the goal.  I'll be looking into costs and availability of the system components you've suggested.

[havtrail]

I have seen pictures of HOBO stations being used in Antarctica, so if you have that many components already it may be worth exploring. Lately though, Onset, makers of HOBO, have been downplaying weather monitoring and focusing on water and environmental research monitoring.

Rich K.

[RonVE8RT]

If I had some guidance / help to add onto the HOBO system (pressure and wind direction) I'd dig out the datasheets.  They're not on my work computer or I could list off the component model or part numbers.  This would be the quickest and most affordable way forward initially and give me a backyard test platform for cold weather evaluation of a data through to RF system and power management.
The challenging part for me and what stalled the project earlier is sorting out how to access the live data output and interface it to the APRS modem.
What an opportunity, to take the HOBO which has been supporting a bird feeder for years, and get data from it    FWIW I've been delighted to have it as a bird feeder stand if nothing better, this year we've been thrilled to have boreal chickadees show up.

[RonVE8RT]

Maybe a useful FYI wrt to HOBO sensor wiring.  Our local squirrels easily chew through the sensor wiring as I found out on the air speed sensor.  They chew on RG58 coax too, but I have found that they leave things alone if I don't disturb them when they eat the bird food.  Contrary to the expectation that the number of squirrels would increase if they're fed it looks like the population has dropped significantly, maybe because of the foxes living in town (habitat loss from the fires last year?) and other predators.  The wind speed sensor wire will be replaced with mil spec wire and the undamaged wires I'll see if spiral wrap will protect them.  Anyone else having rodent or other wildlife damage issues?

[RonVE8RT]

Having found my HOBO files I remember why I shelved attempting to adapt it for use with APRS.
The sensors are described as being "smart sensors" but I did not find out what that term means or if it changes from sensor to sensor.  The sensors that are on the system include:
Logger H21-001
TMA-M006 8 bit temperature (grass level) temperature sensor (-40 to +100 C)
THA-1 temperature and humidity sensor
WSA-M003 wind sensor
RGx-M002 rain sensor
LIA-M003 light sensor

The manual for the logger mentions optional sensor inputs for third party sensors but gives no details.
I did find documentation for a VIA-CM14 0-5 volt adapter, the interconnect diagram suggests it adapts 1 wire sensors to the logger but that is not stated.
I have a copy of the now public domain Windows software to download and configure the logger and the now public domain Box Car Pro activation key.

I'm hung up not able to move forward, I don't know what of this WX logger I can use.

[TheBushPilot]

Never been a fan of Hobo based on the fact that you're locked into their ecosystem. Despite using decent measurement equipment, their data logging stuff looks junk. For how much you pay I think a much better system could be made with significantly better parts.

I'd take what parts are universally measurable and move away from the hobo system.

I think if you get a CRBasic Campbell Scientific data logger with PPP you can send APRS data over TCP connection with your radio ID. Not sure what cellular reception looks like up there but you can get one of their low power OEM cell modems compatible with their data loggers and send data to the web that way(?) Not sure what coverage looks like up there but I'm sure there are also proper ways to convert the data to radio signal like you originally had looked into doing.

Assuming you know the protocol to connect and send data it should be pretty straight forward integrating that into the code.

Hobo is likely a dead end and more difficulty than it's worth given the closed architecture and limited customizability.


Cheers

[havtrail]

I think the "smart" sensor designation simply means, as one Onset description reads, "All sensor parameters are stored inside the smart sensor, which automatically communicates configuration information to the logger without any programming, calibration or extensive user setup." You can plug any sensor into any receptacle in the logger, and the sensor will identify itself and things will self-configure. They also have a serial number built in.

The HOBO sensors are pretty high end compared to the typical hobbyist-grade weather equipment, and their prices reflect that. The S-RGx rain bucket, with a tipper mechanism, is larger (6 inch diameter and 9 inches deep) and I think far better than Davis, Ecowitt or others, from what I read on this forum. My weather station uses a "smart"-ized version of the Davis wind speed/direction unit, which Onset characterizes and offers as a lower-budget alternative to their own higher-priced unit.

Your S-LIA light sensor is a special one to measure light in the photosynthetically active radiation (PAR) range rather than overall ambient light levels. That, along with the S-TMA temperature probe in a stainless housing usable in soil or water as well as air, suggests that the unit you have was geared toward plant or agricultural use, rather than general weather monitoring.

I'm not familiar with the H21 logger and the older Box Car software. The weather monitoring station I manage uses phone data transmission to Onset's HOBOlink remote server, but I also happen to use their HOBOware Pro software on a laptop in conjunction with Onset's small temp/humidity loggers (like the UX100) to measure and log conditions in an old stone house used by our local historical society to house its papers and artifacts.

I am not at all familiar with APRS. If you can download the data from your logger to a PC using their software, you may also be able to peek into the data file and try to figure out its formatting. That's well above my level of knowledge, but others here may be able to help you puzzle through it.

Rich K.

[RonVE8RT]

If I'd joined this list a few years ago the bird feeder would have been put up somewhere other than the HOBO tripod    Apart from the squirrel damaging a cable it's all there and in (as best I can tell) good condition.  I did find the links on your signature file helpful - maintenance and inspection of weather stations in particular.
The HOBO sensors specifications are at best down to -40, but the accuracy of the data wasn't that important, what I did not want to have happen was for the information to be blanked out (one manufacturer I contacted told me that there product does that).  FWIW interestingly all of the automotive temperature displays I've seen do read and reasonably accurately) below -40  .
The APRS data can be set to report at any interval, for my purposes even once or twice an hour is enough.
Here is a link to an APRS site that I've selected the remote station (at our ski club) VE8SKI which has continued to operate without servicing for several years (and its made with inexpensive components).  VE8TEA on the map is my home APRS IGate (internet access point for APRS data).  Its not my station or I may have made a clone of it.
I'm encouraged by the help I've received here, maybe over the coming months my own APRS WX data will be online.
Thanks to all!

[RonVE8RT]

Seen this happen so often before, here's the missing link.
https://aprs.fi/info/a/VE8SKI