How to connect PT1000 sensors to Homey

Hello,

I have a Solar water heater system. But unfortunately the controller did yesterday. I was think, maybe I can do the control via Homey. Does some know how I can connect a PT1000 sensor with two wires to a device that can be used in Homey?

Thanks,
Sjors

If you can use an other temperature sensor, then this might be an idea:
https://shop.shelly.cloud/temperature-sensor-addon-for-shelly-1-1pm-wifi-smart-home-automation
Combined with a Shelly 1/1PM and a DS18B20 sensor

Thanks, I will have a look

Typical devices to connect external sensors are the Fibaro Smart Implant (Z-Wave) and the Shelly Uni (WiFi). Both support DS18B20 temperature sensors, but also analog 2- or 3-wire sensors.
Both devices are supported by Homey.
I don’t know, if a PT1000 will work.

The Fibaro Smart Implant works well with analogue temperatur sensors like the PT1000, ie. a common type of Platinum Resistance Thermometer (PRT) with 1000 kilo-Ohm resistance at 0 dg C. I connected a simple 2-wire PRT straighforward according to Section ā€œ4.5: Connection with 2-wire 0-10V sensorā€ of the ā€œOperating Manual - FIBARO SMART IMPLANT FGBS-222 - v1.2ā€:

  • Initially I use 12V DC batteries for power supply, connected as per the wiring diagram.
  • The PRT connects to Input 1 (IN1), with Parameter 20 set to ā€œ5 – Analog input with internal pull-upā€.
  • The ā€œinternal pull-upā€ resistance is simply the temperature-dependent resistance that makes a PRT like this work as intended, I guess… :wink:

However, the output shows up at the Homey Pro smart hub as a voltage, which I then need to calibrate / convert to a temperature (by figuring out the linear V vs T relationship, which should be straighforward enough):

  • with this PRT the output reads 4,22V at 21 dgC. Measured resistance was 12 800 Ohm at 21 dgC (12.8k).
  • That suggests a modest 1.0 mA of current drawn from my (temporary) 8x AAA battery pack, which should provide for useful battery life, before ā€˜upgrading’ to a (permanent) 230V AC to 12V DC adapter.

As far as I understand, while the Smart Implant can handle up to six external Temperature sensors (in addition to the one internal), they must all be digital sensors, of the DS18B20 and/or the DHT22 types.

Does anybody know/ use a better way to capture Temp readings from an existing (built in) PT1000 or similar analogue PRT?
…without any need for ā€˜manual’ conversion from Voltage to Temperature?
…such that this sensor temperature easily integrates with other temperatures, e.g. when trends are analysed on a Homey Pro Insight plot?

I got 10 of these Blebox Tempsensor

I use them for
Hotwater tank/ ventilation/ sauna/ buffer/ pool/ sun heater etc.

Did you take into account that the resistance / temperature relationship is not linear?
Pt1000 Resistance Table.

Thanks for sharing this insight with the link, Adrian,
as I’d just assumed that the Temp vs Resistance relationship was linear and now stand corrected :wink: Much appreciated!

Two follow-up questions, if you don’t mind:

  • I also thought hat a separate ā€œpull-up resistorā€ should not be required when I connect a PT1000 to a Fibaro Smart Implant, in order to measure/ transmit/ record & convert the voltage signal?
  • Given the inevitable voltage drift as my initial battery pack inevitably drains, will there also be drift in the readings of P1000 voltage => temperature ? I had planned to just try the batteries over some time to just see how it perfomed, but perhaps I need to ā€˜just bite the bullet’ and ā€˜upgrade’ to a (permanent) 230V AC to 12V DC adapter, right from the start?

Thanks, Lichtenberg,
That looks like a useful device, which I’ll certainly consider for other applications…

For the case at hand, unfortunately I’m ā€˜stuck’ with an existing PT1000 sensor embedded into a roof-integrated solar PV system, underneath the PV panels high up on the roof - in other words, not easily accessed/ replaced :wink:

…so in the first instance, I’ll try to work with the existing PT1000 sensor, connecting whatever is easiest to the pair of sensor wires coming down from the roof :wink:

This as part of monitoring panel temperature vs solar power yield, perhaps also playing with rinsing & cooling the PV panels…

I’m not sure how you have wired the PT1000 in to L1, but as it is a resistance devices you need to pass a current through it via a series resistor to create a voltage drop. I guess the internal pull-up is providing that function (the 1kOhm in the diagram, but I’m not sure what the actual value of it is).
image

Ideally you want a regulated voltage supply to prevent any drift as you mentioned.

Aha!
Well then hassio for the rescue:

Theres an excellent Hassio to homey integration from the dear @RonnyW

Do notice that an RPI is kinda golden these days…

1 Like