Topic: Which PC-interfaced wireless systems allow tracking of 7+ temp/RH sensors?

[NeverDie]

For instance, Oregon Scientific's version 3 systems can utilize 10 different wireless channels, and around 8 or so of the 10 can be wireless temp/RH sensors like the THGR810.

However, as noted elsewhere on this forum, OS humidity sensors apparently have had poor accuracy compared to some other brands.  So, I'm wondering what other options I might consider that would be more accurate.

My near-term goal is to track the temp/RH "weather" in my house and in various other structures on my land.  So, for the task at hand, the sensors don't need to be water proof.  However, the attic is one of the areas I want to track, and it can get quite hot there.  How hot?  Well, that's what I aim to find out.

[nincehelser]

This isn't a solution, but an observation that might be developed into one...

I've noticed the AcuLink Bridge spits out data for every sensor it hears.  Although AcuRite has a limit of 3 sensors in their software, apparently that is a somewhat artificial limit.

One user has had up to 4 sensors reporting (using custom software), and the practical upper limit is unknown.

One potential problem is exactly how to identify each sensor.  They transmit a short ID code of what looks like 4 digits, but you would have to figure out what they are by bringing each one up at a time.

The Bridge itself is $80.  Each "tower" temp/humidity sensor is $8.  I use a Raspberry Pi running Linux to intercept the Bridge data, but any Linux machine with a couple of network interfaces and the ability to analyze data packets will easily do.

[johnd]

A system based on a Davis Envoy8X unit would allow up to 8 T/H transmitters to be tracked. This wouldn't be a cheap solution but is certainly feasible.

If you had a cluster of monitoring locations near to one another and needed only temperature at these locations rather than T/H then a variation on the Envoy8X suggestion would to use a Davis 6345 multi-sensor transmitter at that location. You can connect at least four temp sensors to this one transmitter.

[NeverDie]

After sleeping on it, I'm gravitating toward getting a RFXCOM RFXrec433, which can report on 433Mhz sensors from a number of different manufacturers (though, unfortunately, not Davis since Davis is 915Mhz).

There are some digital humidity sensors available now that measure RH with accuracy <=2% and some digital temp sensors have accuracy <= 0.25%C.   From what I've read elsewhere on this forum, I suspect that the OS WMR300 will be using digital sensors.  Most likely the other weather stations will soon come out with digital sensor upgrades as well.  Since humidity sensors can drift as they age, it would be nice if they could be just popped off and easily replaced or upgraded.  However, I'd have to go the Arduino route for that, and that's a rather tall hill to climb just to get good humidity info.

That said, since the AccuWrite wireless temp/rh devices are so inexpensive (around $8 for the whole thing), I'd be willing to try manually upgrading one with a digital sensor if there were a digital drop-in sensor replacement.  However, I don't know if it's feasible, even in principle.  Has anyone succeeded in doing it already?

[nincehelser]

That said, since the AccuWrite wireless temp/rh devices are so inexpensive (around $8 for the whole thing), I'd be willing to try manually upgrading one with a digital sensor if there were a digital drop-in sensor replacement.  However, I don't know if it's feasible, even in principle.  Has anyone succeeded in doing it already?

I've no idea, but here's a look at the inside of one of the AcuRite Tower sensors.  Maybe it'll help.

I know the top board with the spring antenna is the 433MHz transmitter.

The board near the bottom of the ribbon cable is an A-B-C switch.

I assume the white board labeled "28" is one of the sensors.

Please feel free to identify the important components that you can.

[NeverDie]

It looks like board 28 is the humidity sensor.  Most likely it has a hydrophilic coating on it, and the measured resistance changes with humidity.  Given the $8 price tag, I'm not surprised.  As you would probably guess, this is the least accurate RH sensor.

Unfortunately, all the digital RH sensors that I've seen have had more than two leads.  Therefore, I don't see how the Acurite tower  would be easily upgradeable to a digital sensor using a pin-for-pin drop-in replacement.

I've read on this forum that some people have upgraded their Davis rh sensors from a SHT11 by unsoldering it and then soldering in a SHT15.  However, both those sensors are already digital, and it appears they are pin-for-pin the same.  What they're doing is upgrading from plus-or-minus 3%RH digital sensor to a more accurate 2%RH digital sensor.

Having a picture like what you provided helped me think harder about the issue.  Unfortunately, I doubt it will ever be feasible to upgrade from analog to digital by replacing just the sensor element alone, because the rest of the circuitry is still expecting to measure something analog (not digital).

[NeverDie]

This isn't a solution, but an observation that might be developed into one...

I've noticed the AcuLink Bridge spits out data for every sensor it hears.  Although AcuRite has a limit of 3 sensors in their software, apparently that is a somewhat artificial limit.

One user has had up to 4 sensors reporting (using custom software), and the practical upper limit is unknown.

One potential problem is exactly how to identify each sensor.  They transmit a short ID code of what looks like 4 digits, but you would have to figure out what they are by bringing each one up at a time.

The Bridge itself is $80.  Each "tower" temp/humidity sensor is $8.  I use a Raspberry Pi running Linux to intercept the Bridge data, but any Linux machine with a couple of network interfaces and the ability to analyze data packets will easily do.

Since you already use a Raspberry Pi, you might be interested in this 433Mhz arduino shield for $29:
http://www.freetronics.com/collections/shields/products/receiver#.Ue3AaEYo6-o

[nincehelser]

Since you already use a Raspberry Pi, you might be interested in this 433Mhz arduino shield for $29:
http://www.freetronics.com/collections/shields/products/receiver#.Ue3AaEYo6-o

I was thinking of wiring the receiver module directly to the pins of the Pi like this guy did:

http://ninjablocks.com/blogs/how-to/7506204-adding-433-to-your-raspberry-pi

Mainly because I'm cheap.

I've never worked with an arduino, though....I've just read about them here and there. 

Is there a big advantage to using this shield and arduino rather than wiring directly to the pi?

[NeverDie]

Is there a big advantage to using this shield and arduino rather than wiring directly to the pi?

I don't think I can answer that without getting down to a much finer grain of detail, which it sounds like you are already doing.  If I were to do that (which I'm not going to do anytime soon), I would start by trying to research and compare the performance characteristics of the two different receiver modules.  The other aspect to examine might be which already has better software support that you can leverage.  If saving on material cost is what's driving you, then you *may* get a different answer than if you put a dollar value on your time and consider that as well.  

The reason why I brought it up at all is that I was under the mistaken impression that you were somehow hacking into a Bridge to get the data, not sniffing the RF using another (relatively inexpensive) device, which is what it sounds like you are actually already doing.

Please do let us know how it goes for you and whether, after the fact, you are happy with the path you picked!

[nincehelser]

I don't think I can answer that without getting down to a much finer grain of detail, which it sounds like you are already doing.  If I were to do that (which I'm not going to do anytime soon), I would start by trying to research and compare the performance characteristics of the two different receiver modules.  The other aspect to examine might be which already has better software support that you can leverage.  If saving on material cost is what's driving you, then you *may* get a different answer than if you put a dollar value on your time and consider that as well.  

...

Please do let us know how it goes for you and whether, after the fact, you are happy with the path you picked!

Thanks for the input!

I took a closer look at the Arduino.  Although it's not a full computer system like the Pi, it does look interesting and the software appears more mature.

I think I'll get one and the shield you pointed to for experimentation purposes.  It doesn't hurt to learn something new.

Eventually what I'd like to see is a good 433MHz USB dongle that avoids dependency on particular hardware.  I did find a candidate, but apparently English documentation is non-existent:

http://www.amazon.com/USB-1100PA-433MHz-Wireless-Transceiver-Module/dp/tech-data/B009SIQ7XE/ref=de_a_smtd

[NeverDie]

Eventually what I'd like to see is a good 433MHz USB dongle that avoids dependency on particular hardware.

In that case, you may want to consider a RFXCOM as another option also.  Higher cost, but hopefully less time and effort.  I have a receiver on order now, but RFXCOM makes transceivers too.  From what I've read, the receiver is more sensitive than what is typically found in weather station gear.  Consequently, you should get better range from your weather sensors.  I don't know if it works for AcuRite, but it does work with OS (all three RF versions) and weather sensors made by a number of other brands also.  I don't yet know what weather software (if any) is compatible with it, so if anyone knows of any, please post.

[nincehelser]

Interesting, but it's at a price point much higher than what I had in mind. 

One of my design goals is to get the hardware costs as low as possible...not for myself, but for anyone wanting to build a similar system.   I have no intention of marketing anything, but I still think in terms of keeping production costs low.

rfxcom.com is interesting.  Now I understand why the Acurite Bridge has an undocumented transmitter in it...they must be thinking about entering the home automation market.