Measuring the temperature of a Hot Water Cylinder for use in Home Assistant

Forums › Gadgets, robotics, home automation, electronics (including wearables) › Measuring the temperature of a Hot Water Cylinder for use in Home Assistant

prob

222 posts

Master Geek

#312220 27-Mar-2024 11:18

I have a Bluetooth / ZigBee controller to turn the hot water cylinder on and off. For the moment this is just to make use of off peak power.

I am looking to get solar and then I can hopefully use solar power during the day.

It would be helpful to know the temperature of the water. I expect that there is a resistive sensor (old school) somewhere that regulates the temperature, but what is he easiest way to get a temperature reading from the cylinder that I can import into Home Assistant?

Thanks in advance.

SteveXNZ

58 posts

Master Geek

Lifetime subscriber

#3212006 29-Mar-2024 09:10

Coming back to this post after a couple of days hoping to read all the answers as I have the same question!

In my case I have a standard HWC element/thermostat in the base of my cylinder which I turn on and off with an Aeotec heavy duty Z-wave switch (ZW110-B09). I want to have a greater understanding of the HWC water temperature at both the base and top exit point so I can program some more intelligent decisions based on tariffs and solar.

My approach has been to use a DS18B20 sensor thermally cemented to as close to the temperature source as I can get. I feed this into a Fibaro Smart Implant (FGBS-222) which can accept up to six temperature input channels. Note if wiring more than one you need a 4.7k resistor as a pullup from the DATA to VCC line.

While I have got this working it’s fiddly, not reliable, and it’s difficult to get the sensor close to the source. Another approach is to drill a hole through the outer case of your HWC and use a cylindrical DS18B20 pressed against the internal cylinder. Unlikely to resort to that.

Of course what would be great is to have the thermostat/element in the base of the cylinder integrated with a Z-wave/Zigbee/Thread controller so it can be incorporated into smart home systems. I’ve searched but haven’t found anything – I’m puzzled why the technology around heating hot water, which is a significant energy consumer, seems to be buried in the dark ages. Illumination required!

AliExpress

Shop now on AliExpress (affiliate link).

Goosey

2792 posts

Uber Geek

Subscriber

#3212008 29-Mar-2024 09:16

How does your monitoring and control fit in with this guidance from “WHO”…

The optimal temperature for Legionella proliferation in water varies between 32°C and 35°C, but it can easily proliferate at temperatures of up to 45°C. Usually, there is no growth above 55°C, and a temperature of over 60°C has a bactericidal effect. Thus, the WHO recommends that water be heated and stored at 60°C (3).

SteveXNZ

58 posts

Master Geek

Lifetime subscriber

#3212013 29-Mar-2024 09:42

Goosey:

How does your monitoring and control fit in with this guidance from “WHO”…

The optimal temperature for Legionella proliferation in water varies between 32°C and 35°C, but it can easily proliferate at temperatures of up to 45°C. Usually, there is no growth above 55°C, and a temperature of over 60°C has a bactericidal effect. Thus, the WHO recommends that water be heated and stored at 60°C (3).

It's always good practice to keep the thermostat at 60°C for long term storage. Having temperature sensors allow you to:

Monitor the rate of temperature decline when there's heavy hot water draw (ie the teenagers are in the shower again)
Monitor recovery times
Allow for the HWC to be used as an energy storage device. Eg when tariffs are cheap or solar plentiful, heat the water to the upper limit, say 70+°C. So long as a tempering valve prevents scalding and the HWC is well-insulated, this can be an effective energy storage mechanism, not dissimilar in concept to a solar battery.

It all comes back to data yielding knowledge yielding wisdom - in this case home energy management. When we hit the button on our heat pumps for household heating and cooling we expect to be able to set a desired temperature, and see how the room temperature is tracking towards it. If we can make it happen for air, why not water?

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3357848 27-Mar-2025 14:43

Just wondering if there's any updates on this, looks like I need to get a new HWC and if I need to spend $$$$ on a new cylinder I'd like to have a temperature-monitoring capability for it. It looks like you can get heating elements with thermostat pockets but that's for standalone elements, not the whole unit, are there HWCs that you can easily insert temperature probes into? And for the thermostat pockets, will they be for the HWC's own thermal control or can you drop in a DS18B20 for your own use?

CJC

54 posts

Master Geek

#3358225 28-Mar-2025 13:36

My setup is a little basic, but I have a Zigbee remote temperature probe (probably a DS18B20 on the end of it) that's tucked up against the output pipe of the cylinder, under the pipe insulation. It's normally in the range of 55-62 degrees, so accurate enough for me.

hairy1

3331 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#3358236 28-Mar-2025 14:00

I have a DS18B20 attached to an ESP8266 tucked in the base of the cylinder behind the insulation running esphome. It works really well and I have it triggering if the water gets too cold. It has definitely saved me heating the water up many times during the night as the HWC thermostat cycles. It always get heated to over 60 degrees once a day.

My views (except when I am looking out their windows) are not those of my employer.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3358247 28-Mar-2025 14:59

Looks like the Rheem 180L and up have pockets for temperature probes, but not the 135L, sigh. Currently trying to decide whether I go for the annoyingly large 180L or the 135L and thermal-epoxy it to the output pipe like @CJC.

hairy1

3331 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#3358251 28-Mar-2025 15:25

Here's my last two days of my hotwater cylinder temps. I add the yellow graph which shows when the cylinder is heating for awareness. I move all my heating out of peak power times. I always do a heat in the morning before 6 am for WAF. The cylinder is allowed to heat during shoulder period during the day with solar.

My views (except when I am looking out their windows) are not those of my employer.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3358253 28-Mar-2025 15:29

I'm on free power 9pm-midnight so a dumb solution in this case would be to just enable power to the HWC during those hours and let it run down during the day.

SepticSceptic

2170 posts

Uber Geek

Trusted

#3358375 28-Mar-2025 23:18

Would the copper pipework at the top of the cylinder be a convenient mounting point for a temperature sensor ?

Given that copper is an excellent temperature conductor and the pipe sits at the top of the heated water column, any reading should be pretty close to actual ?

Would save any drilling, etc

hairy1

3331 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#3358395 29-Mar-2025 08:27

I tucked mine in beside the heating element access panel. I decided the bottom was better as it has much more variance than the top.

My views (except when I am looking out their windows) are not those of my employer.

SpookyAwol

626 posts

Ultimate Geek

#3366526 22-Apr-2025 15:34

My system is fairly basic, i just use the Shelly Temp add on with 2 probes. The probes are inserted between the insulation and the tank ( at 2 levels) and gives a fairly good indication of temps.
Obviously the control temp is provided by the tank thermostat, but you can easily monitor temps and switch tank on and off with free power / solar peaks

fastbike

202 posts

Master Geek

#3377111 26-May-2025 22:07

I'm just starting on this project.

Background: We have a Thermocell solar hot water system - which uses evacuated flat panels with a working fluid (glycol) to transfer accumulated heat into a heat exchanger and thence to the the HWC via a low powered pump.

I've just added a Shelly Plus Uni with 5x DS18B20 sensors. The top and bottom sensors are inserted into the tubes that house the sensors for the solar setup (top is a read only, bottom is used for pump control) and the other three are inserted into 3x 40mm diam holes drilled through the outer galv steel housing (removed the insulation) and glued the sensors to the stainless inner, before foaming the hole in the insulation.

I can now see the gradient of the cylinder temps: today was pretty overcast with a short fine spell around midday before high cloud resumed. Here's the read out from the 5 sensors at 4pm

I have a relay to control the bottom 3kW element, and a Triac type control on the top 3kW element - which allow a variable amount of power to be fed in which is useful to use as a solar diverter. I've set the bottom thermostat to 50 degrees, and the top thermostat to 70 degrees. Interesting this cylinder has thermostats which only make contact on the outside of the stainless inner cylinder. The previous one (rotted out by chlorine) had a tube that a thermostat wire seated into.

I need to set up some automation rules to make sure the water heats overnight but only as required i.e. if the cylinder has sufficient water @approx 55 degrees then no supplement is required. With a weekly cycle to get above 65 degrees for control of bacteria etc.

Otautahi Christchurch