Measuring Water and Ambient Temperatures

[GDavis01]

I would like to measure the water and air temperatures at my cottage. After a bit of research, I am looking to buy the Universal RS232 2-Way Serial & 1-Wire Interface for ISY to connect to my ISY and a bunch of the DS18B20 (SS encapsulated) sensors.

 

I want to install the bridge in my boathouse which is unheated and can get quite cold in the winter. The main purpose is to determine the temperatures during the warmer months for swimming...etc. However I would also like to track the winter air temperatures. 

 

I have Ubiquiti Nanostations to get my network connection (wirelessly) from the cottage to a switch in the boathouse, which I will connect the bridge to.

 

A few questions... 

 

- I couldn't locate specs on the bridge to determine if it can handle cool temperatures... does anyone know?

 

- The DS18B20 sensors, that I see, seem to be a 1 meter length. Do some come in longer lengths or can I use CAT6 wire to lengthen them?

[Teken]

I would like to measure the water and air temperatures at my cottage. After a bit of research, I am looking to buy the Universal RS232 2-Way Serial & 1-Wire Interface for ISY to connect to my ISY and a bunch of the DS18B20 (SS encapsulated) sensors.

 

I want to install the bridge in my boathouse which is unheated and can get quite cold in the winter. The main purpose is to determine the temperatures during the warmer months for swimming...etc. However I would also like to track the winter air temperatures. 

 

I have Ubiquiti Nanostations to get my network connection (wirelessly) from the cottage to a switch in the boathouse, which I will connect the bridge to.

 

A few questions... 

 

- I couldn't locate specs on the bridge to determine if it can handle cool temperatures... does anyone know?

 

- The DS18B20 sensors, that I see, seem to be a 1 meter length. Do some come in longer lengths or can I use CAT6 wire to lengthen them?

 

Operating Temperature: I haven't seen anything published but would assume it falls in the range of 0 - 40'C.

 

Wire Length: You can use CAT6 cable to extend the sensors reach. Keep in mind CAT6 comes in 23, 24, 26 AWG you will really need to purchase 23 AWG for long runs. No matter what brand of wire you purchase do not buy CCA wire as its not in wall rated or NEC / CEC approved.

 

A better choice is to use 22-4 alarm wire as its even thicker 22 AWG and the best part most of the brands comes with the same wire colors.

 

Lastly, please take a few moments to fully understand which type of wiring topology to use as it will impact the system if long wire runs are undertaken. 

[GDavis01]

Thanks for the quick reply.

 

I have a spool of certified 23AWG CAT6 cable so I should be good to go with that.

 

I guess I will just have to try the bridge and see if it holds up under the colder weather! The good news is that it will be inside the unheated structure.

 

I don't expect to have any long runs... but where would I find information on the types of wiring topology?

[Teken]

Thanks for the quick reply.

 

I have a spool of certified 23AWG CAT6 cable so I should be good to go with that.

 

I guess I will just have to try the bridge and see if it holds up under the colder weather! The good news is that it will be inside the unheated structure.

 

I don't expect to have any long runs... but where would I find information on the types of wiring topology?

 

Pay special attention to how a linear, stub, star, topology can impact the 1 wire network:  https://www.maximintegrated.com/en/app-notes/index.mvp/id/148

[larryllix]

 

If you are only going to run a few feet wiring topology isn't going to matter much whether star connected or daisy chain connected.

But, like Teken says, if you are going to daisy chain a few, or even one over a 100 foot, or more run, then terminations and particular attention to your wiring style may become critical.

From what I have heard the RPi 1-wire board is quite resilient and error free.

Another consideration for you, could be the CAO TAG system. Very accurate and easy to use, plus you can view their cloud srrver that collects data for the last few years on a 1,2,5,10,15,20,30,60 or more update basis. The Pro units hold months of data that can be downloaded after they make contact again with the manager.

For water immersion temperatures you would want one particular unit that takes the same DS18B20 probe.

Downside? CR2032 batteries to change about once per year and a central Tag manager using the cloud somewhat.
Manager: about $40, Tags: about $20 to $35 with temp/humidity, X,Y,Z angle detection, Motion, light intensity and weatherproof.

I have one in my glove box and it detects my wife coming home from almost 2km away on my ISY.

 

[GDavis01]

Before I saw your post on the CAO Tags I went ahead and ordered my Autelis bridge and a bunch of DS18B20 probes.

 

My plan is to update the variables on the ISY regularly and then push them down to my RPI where I will create a database of the readings.

 

With respect to the wiring topology, I will not be running any wires more than 10 - 15', but reading through the link provided by Teken helps me understand the concerns if I decide to do more later.

[Teken]

With such a short distance you will be more than fine. Going the Autelis route will negate ever having to worry about a battery change or a device failing due to extreme temperatures.

 

The Autelis Bridge should be flashed to 1.3.2 firmware by the time you receive it. If not please ensure you do so to enjoy all the bug fix's and features I pushed for.

[builderb]

Before I saw your post on the CAO Tags I went ahead and ordered my Autelis bridge and a bunch of DS18B20 probes.

 

My plan is to update the variables on the ISY regularly and then push them down to my RPI where I will create a database of the readings.

 

With respect to the wiring topology, I will not be running any wires more than 10 - 15', but reading through the link provided by Teken helps me understand the concerns if I decide to do more later.

I've got bash and Python scripts that will read sensors, push the variables to the ISY, and MySQL and/or rrd databases, if you need anything along those lines. I also have some rrd_graph scripts that can pull from those databases. Depending on your setup, you may want a pull-up resistor somewhere along your wiring. A 4.7k ohm resistor is the default.

 

Oh, and before I dropped one of the 'waterproof' sensors into my fish tank, I got a can of PlastiDip and coated the sensor such that I could submerge it several inches before the water could find a seam to penetrate.

[GDavis01]

 

 

I've got bash and Python scripts that will read sensors, push the variables to the ISY, and MySQL and/or rrd databases, if you need anything along those lines. I also have some rrd_graph scripts that can pull from those databases. Depending on your setup, you may want a pull-up resistor somewhere along your wiring. A 4.7k ohm resistor is the default.

 

Hmmm! Your setup sounds more complicated than I was planning. I don't expect to get into graphing. I have not yet looked into setting up a database on the RPI so I may come back to you on that at some point.

 

I am also a bit confused by your suggestion... I plan to use the Autelis bridge to move the sensor readings to the ISY and then pass those variables down to the RPI. Whereas you seem to be suggesting accumulating the sensor readings on an RPI (?) and then passing them up to the ISY! Or am I missing the point?

 

I would appreciate it if you would explain your comment about the pull-up resistor... I didn't realize that I would need one! I was planning to daisy chaining 4 sensors to the Autelis bridge which would be connected to the LAN via a switch in my boathouse. Where would I place the resistor in this setup and why would I need one.

 

 

 

Oh, and before I dropped one of the 'waterproof' sensors into my fish tank, I got a can of PlastiDip and coated the sensor such that I could submerge it several inches before the water could find a seam to penetrate

 

Interesting thought...  did you spray the entire sensor cable including the stainless portion or just the cable portion? I would think that the coating over the stainless portion might affect the sensor readings. I will be dropping 2 of the sensors into a lake so this may be a very good idea.

[builderb]

Hmmm! Your setup sounds more complicated than I was planning. I don't expect to get into graphing. I have not yet looked into setting up a database on the RPI so I may come back to you on that at some point.

 

I am also a bit confused by your suggestion... I plan to use the Autelis bridge to move the sensor readings to the ISY and then pass those variables down to the RPI. Whereas you seem to be suggesting accumulating the sensor readings on an RPI (?) and then passing them up to the ISY! Or am I missing the point?

 

I would appreciate it if you would explain your comment about the pull-up resistor... I didn't realize that I would need one! I was planning to daisy chaining 4 sensors to the Autelis bridge which would be connected to the LAN via a switch in my boathouse. Where would I place the resistor in this setup and why would I need one.

 

 

Interesting thought...  did you spray the entire sensor cable including the stainless portion or just the cable portion? I would think that the coating over the stainless portion might affect the sensor readings. I will be dropping 2 of the sensors into a lake so this may be a very good idea.

 

You are correct, I actually don't have the Autelis portion of the system, I use a 1-wire controller to monitor my network of sensors, and in my case, push the data to my databases and to my ISY. But you could just as easily adapt the code to read from the Autelis, or from the ISY, and stash those values in the database. 

 

Personally, I have found that numerical temperature data just isn't very useful in its raw form. Unless I can see trends, spot highs and lows, and correlate them to other sensors, it's all just a jumble of numbers. Humans are pattern-recognizing creatures, not number crunchers. Thus the time I've spent toying with databases and graphs. 

 

Based on what you've said, you're probably OK without the resistor. The reason you add one is because on long cables, the time it takes for the rising side of the signal waveform to reach read voltage levels exceeds proper read times, usually due to excessive length, but also when signal reflections are traveling down longer branch lines in a star topography.  Connecting a 4.7k ohm resistor across the data and positive lines allows the positive line to "pull-up" the signal cable more effectively. I'm no electrical engineer, and I'm sure there are others here who can give a much better explanation than I can, but that's the nub of it.

 

Also, it is possible to do this without the Autelis or a 1-wire controller. You can connect directly to the w1 interface through the Raspberry Pi's GPIO pins, and read from there. I wouldn't count on that kind of setup going very far, or having too many devices on it, but you can read temperatures with a Pi for well under $100, including the cost of the Pi.

 

Edit: forgot about the PlastiDip. No, I used the kind you dip the sensor into. This was a year or so ago, but I believe I dipped it once a day for about 4 days to get a nice thick coat on it. It doesn't affect the sensor reading, probably just takes a little longer for the temperature to stabilize, but once it's submerged it's all going to be the same temperature eventually.

[larryllix]

Simply put...

 

When you send very sudden change of electrical voltage down a long wire and the impedance changes suddenly you get reflections. A resistor at the far end can absorb some of the energy at that end and soften the impedance change.

 

As an analogy...

you can fly through the air and you can glide through the water but... try diving off a high-diving board into water on your flat stomach. A pointed hand or toe helps gradually soften the sudden change in density between the air and the water.

[builderb]

Simply put...

 

When you send very sudden change of electrical voltage down a long wire and the impedance changes suddenly you get reflections. A resistor at the far end can absorb some of the energy at that end and soften the impedance change.

 

As an analogy...

you can fly through the air and you can glide through the water but... try diving off a high-diving board into water on your flat stomach. A pointed hand or toe helps gradually soften the sudden change in density between the air and the water.

 

I'll put my "I'm no EE" disclaimer out there again, but my understanding is that, in this case, the resistor can go anywhere along the way, beginning, end, middle. I reviewed this page during my research, which seems to show the resistor at the beginning in some examples. Is it better at the end? My main network isn't that big, so I don't know if my lack of problems is due to total network weight being low, or being helped by the inclusion of a resistor at the beginning of the run.

 

https://www.maximintegrated.com/en/app-notes/index.mvp/id/148

[larryllix]

I'll put my "I'm no EE" disclaimer out there again, but my understanding is that, in this case, the resistor can go anywhere along the way, beginning, end, middle. I reviewed this page during my research, which seems to show the resistor at the beginning in some examples. Is it better at the end? My main network isn't that big, so I don't know if my lack of problems is due to total network weight being low, or being helped by the inclusion of a resistor at the beginning of the run.

 

https://www.maximintegrated.com/en/app-notes/index.mvp/id/148

Yes. Best at the very far end or both ends depending on what the input to the Autelis interface is.

 

I am using the WC8 for 1Wire stuff and after lots of problems I have found with a scope that the 1wire devices control their slew rate (how sharp the signal edges are) but the WC8 cheap interface doesn't. As a result signal from the sensors don't tend to echo in the wires but the signals from the WC8 cause echoes constantly confusing itself.

 

The furthest you could easily mount a resistor would be at the cable junction to the last sensor. One at the Autelis end may work just fine when you aren't running 100s of feet of cable too. Teken is familiar with that equipment.

[Teken]

- The Autelis Bridge does not require a pull up resistor to be used.

 

- If you purchased the hermetically sealed stainless steel 1 wire sensors you don't need to encapsulate them with anything else.

 

- Given your 1 wire network will be so short and small I don't forsee there being any issues.

 

- If you're going to submerge a few sensors into freezing water ensure the the entire cable is protected with a temperature rated jacket. As many of the cheaper cabling is not weather rated and will snap and break like glass.

 

- Place the Autelis Bridge on a GFCI outlet to ensure any possible fault voltage protects you and others since you are indeed submersing the sensors into water.

[GDavis01]

- The Autelis Bridge does not require a pull up resistor to be used.

 

- If you purchased the hermetically sealed stainless steel 1 wire sensors you don't need to encapsulate them with anything else.

 

- Given your 1 wire network will be so short and small I don't forsee there being any issues.

 

- If you're going to submerge a few sensors into freezing water ensure the the entire cable is protected with a temperature rated jacket. As many of the cheaper cabling is not weather rated and will snap and break like glass.

 

- Place the Autelis Bridge on a GFCI outlet to ensure any possible fault voltage protects you and others since you are indeed submersing the sensors into water.

 

I have no idea if the sensors I bought on eBay (China!) are " hermetically sealed stainless steel 1 wire sensors"!! I ordered a bunch of them so if I find that I'm having a problem then I will try something like PlastiDip.

 

Given that I won't be doing a lot swimming in the winter months   I will be pulling those sensors out of the lake and only using the air temperature ones.

 

The bridge, as you suggest, will be on a GFCI outlet.

[apostolakisl]

If you plan on leaving the one wire sensors in the water during winter months, I would say you are likely to have issues.  If the sensors themselves freeze into solid ice, including the wire to them, they will be subject to all kinds of shearing forces as the ice cracks, shifts, thaws, re-freezes, etc.  

 

Perhaps consider encasing the entire wire up to and just including the top portion of the sensor in a pvc pipe filled with epoxy or some other similar material. 

[GDavis01]

 

 

If you plan on leaving the one wire sensors in the water during winter months, I would say you are likely to have issues. 

 

I agree... that's why I said that I would be pulling those sensors out of the water for the winter.

[GDavis01]

I received my Autelis bridge some time ago, and today I received my Chinese sensors (DS18D20's).

 

As I start to do the installation and testing, I am struggling with the connection of the sensor!

 

(There may be an obvious answer to this question, but I'm not seeing it! ) How do I physically connect a sensor to the bridge?

 

The sensor has 3 wires, Yellow, Red and Black. 

 

The 1-Wire Pinout on the Autelis bridge has 8 pins 

 

Pin 1 - 1-Wire Data
Pin 2 - 1-Wire Data
Pin 3 - Do Not Connect
Pin 4 - Ground
Pin 5 - Do Not Connect
Pin 6 - Do Not Connect
Pin 7 - +3.3V Output
Pin 8 - Do Not Connect

 

Does this make sense? Black to ground (pin 4); Red to power (pin 7) and Yellow to 1-wire data (pin 1 or 2)?

 

Are there polarity issues? 

 

What is the difference between pins 1 and 2?

[builderb]

Can't help you with the Autelis, but your wiring is correct. Yes, there are polarity issues. If you wire up the power backwards, you'll fry the sensor. Thankfully, they are cheap!

[GDavis01]

With such a short distance you will be more than fine. Going the Autelis route will negate ever having to worry about a battery change or a device failing due to extreme temperatures.

 

The Autelis Bridge should be flashed to 1.3.2 firmware by the time you receive it. If not please ensure you do so to enjoy all the bug fix's and features I pushed for.

 

My Autelis bridge  came with firmware v1.2. The wiki has the latest fw as v1.3.1 not v1.3.2!

 

I have upgraded to the v1.3.1, but where do I find v1.3.2?

[Teken]

I received my Autelis bridge some time ago, and today I received my Chinese sensors (DS18D20's).

 

As I start to do the installation and testing, I am struggling with the connection of the sensor!

 

(There may be an obvious answer to this question, but I'm not seeing it! ) How do I physically connect a sensor to the bridge?

 

The sensor has 3 wires, Yellow, Red and Black. 

 

The 1-Wire Pinout on the Autelis bridge has 8 pins 

 

Pin 1 - 1-Wire Data

Pin 2 - 1-Wire Data

Pin 3 - Do Not Connect

Pin 4 - Ground

Pin 5 - Do Not Connect

Pin 6 - Do Not Connect

Pin 7 - +3.3V Output

Pin 8 - Do Not Connect

 

Does this make sense? Black to ground (pin 4); Red to power (pin 7) and Yellow to 1-wire data (pin 1 or 2)?

 

Are there polarity issues? 

 

What is the difference between pins 1 and 2?

 

Can you offer a screen capture of the top of the unit with the sticker? Mine indicates pin 1 is 5 VDC it makes no mention of pin 2 at all.

 

My Autelis bridge  came with firmware v1.2. The wiki has the latest fw as v1.3.1 not v1.3.2!

 

I have upgraded to the v1.3.1, but where do I find v1.3.2?

 

I would have thought 1.3.2 was listed in the download section but gather not. If you scan the Autelis forums you will find the 1.3.2 firmware release. Keeping in mind it was introduced for a couple of break fix's and feature enhancements I wanted to see. Not having it won't impact the system from offering the basic data to the ISY Series Controller.

 

Let me know if you have further questions . . .

[GDavis01]

Interesting! Thanks for the quick reply...

 

 

 

Can you offer a screen capture of the top of the unit with the sticker? Mine indicates pin 1 is 5 VDC it makes no mention of pin 2 at all.

 After reading your post, I looked at the sticker on top of the unit and I now see the proper connections!

 

1 = 5 volt

2 = ground

5 = data

 

The quick guide that came with the device had the following ""See the wiki for connection details" for 1-Wire! So I looked at the wiki and not at the sticker! The wiki makes no mention of the connections on the sticker!!

 

With respect to the firmware, I will leave the device as is for now and eventually look for the the v1.3.2.

 

It would be great if Autelis would include a readme file that explained the enhancements between the different versions! I have the same issue with my Autelis Shade Control for Somfy RTS device. I see that there is a new firmware available but I am reluctant to make the upgrade without knowing what it does. The device works great as is, but perhaps there is some new functionality that I am missing out on! 

 

Now back to the set up...

[Teken]

Interesting! Thanks for the quick reply...

 

 After reading your post, I looked at the sticker on top of the unit and I now see the proper connections!

 

1 = 5 volt

2 = ground

5 = data

 

The quick guide that came with the device had the following ""See the wiki for connection details" for 1-Wire! So I looked at the wiki and not at the sticker! The wiki makes no mention of the connections on the sticker!!

 

With respect to the firmware, I will leave the device as is for now and eventually look for the the v1.3.2.

 

It would be great if Autelis would include a readme file that explained the enhancements between the different versions! I have the same issue with my Autelis Shade Control for Somfy RTS device. I see that there is a new firmware available but I am reluctant to make the upgrade without knowing what it does. The device works great as is, but perhaps there is some new functionality that I am missing out on! 

 

Now back to the set up...

 

I agree if companies offered just a little bit of *What's New, Release Notes* etc it would answer lots of questions. If you would like a reference thread which tracks my efforts with this device let me know and I shall forward the resource link.

[GDavis01]

If you would like a reference thread which tracks my efforts with this device let me know and I shall forward the resource link

 

Thanks... I have been reading your journey with the device (found the reference in another thread)... and it  has been very helpful.

 

A few questions, if I may:

 

 - What is the "ISY 1-Wire Variable Offset" option, on the bridge, which is defaulted at '0'?

 

-  Is it possible to pass the data to the ISY in a decimal format (22.2C versus 222C)?

 

[Teken]

Thanks... I have been reading your journey with the device (found the reference in another thread)... and it  has been very helpful.

 

A few questions, if I may:

 

 - What is the "ISY 1-Wire Variable Offset" option, on the bridge, which is defaulted at '0'?

 

-  Is it possible to pass the data to the ISY in a decimal format (22.2C versus 222C)?

 

The offset is poorly named and labeled - essentially all it means is what is the *First* variable ID number in use is. So for example lets say you have your first sensor number 1. But you already have existing variables in use so the *Newly* created variable for sensor number 1 is say ID:32.

 

Your off set would be 32 . . .

 

If your using 4.XX firmware like I am by default the system doesn't support (precision) decimal values like it does for the 5.XX branch. But you can do some basic math to offer that if you like by doing this:

 

Whole Value Test - [iD 02D5][Parent 02D4]

 

If

        $s.Autelis_Living_RM_Temp < 100

 

Then

        $s.Autelis_Kitchen_Whole_Temp  = $s.Autelis_Living_RM_Temp

        $s.Autelis_Kitchen_Whole_Temp /= 10

        $s.Autelis_Kitchen_Frac_Temp  = $s.Autelis_Living_RM_Temp

        $s.Autelis_Kitchen_Frac_Temp %= 10

        Send Notification to '1' content 'Test Whole Value'

 

Else

   - No Actions - (To add one, press 'Action')

 

I never did this because I knew 5.XX supported this basic features. Plus, given I had hundreds of variables in place I wasn't going to spend an entire year replacing the very same.

 

That would have been a huge time suck I wasn't going to allow.