Garden Watering without ground sensors

[s2nrbald]

Does anyone have a good logic for when it is ok to water the garden, without using ground sensors, and likely using the weather? Thanks.

[s2nrbald]

I have decided to use the Loxone Watering example but it still has a downside if you do not have moisture sensors. How do I get the information of the total rainfall amount in the last 24 hours?

[Prof.Mobilux]

You can use the Evapotranspiration method to calculate the watering level. I use this method for quite some time now instead of my moisture sensors and it is very effective. But you need weather data formthat method (Loxone weather service or a LoxBerry with Weather4Loxone or something similar).

https://www.loxwiki.eu/pages/viewpage.action?pageId=70353372

[s2nrbald]

Thank you for this example. I have built it in to my Config. One question since I do not have LoxBerry (I do use the Loxone Weather Service), is there any way to get the 3 input values (Temp, RF, Prec_d) with the Loxone Weather Service?

[Prof.Mobilux]

Yes, absolutely. https://www.loxone.com/enen/kb/weather-service/

You can use Temperature and Rel. Humidity (RF) and use MinMax Block to get today's max values. Precipitation can also be used but must be accumulate manually, e. g. with an analogue memory block.

[Martin1234]

Hi Prof.Mobilux,

Did I understand you correctly that you take the max value for temperature and humidity for the Evapotranspiration calculation (https://www.loxwiki.eu/pages/viewpag...ageId=70353372)? In my understanding it is the average for both values instead of max value.

I am also wondering why the "Moving average" block is after the "Analogue Min Max Limiter". This will lead to stange results in case of heavy precipitation.
Example:

• watering cycle is set to 4 days, so the average from 4 days is calculated
• 3 days of hot wheather without precipitation, result: watering level 200%
• 4. day: precipitation 30mm; result: average of 200%, 200%, 200%, 0% equals to 150% which seems to be strange

Maybe this is acceptable or requested and I understood it incorrectly. Please let me know.

My idea would be to use the moving average over the cycle time of 4 days also on the input values for temperature, humidity and precipitation before using it for the Haude calculation. Another option would be to switch the order of the blocks and put the "Moving average" block before (instead of after) the "Analogue Min Max Limiter".

Any ideas are highly welcomed.

[Prof.Mobilux]

Hi Prof.Mobilux,

Did I understand you correctly that you take the max value for temperature and humidity for the Evapotranspiration calculation (https://www.loxwiki.eu/pages/viewpag...ageId=70353372)? In my understanding it is the average for both values instead of max value.

Yes, it must be the average temperature. Since I have no average temperature from my weather service I used the max. value (as let's say worst case). Since this is no perfect scientific calculation at all, it's "good enough" for me. Of course you can also calculate the real average temperature if you would like to be more accurate. But in my point of view this isn't needed.

[Martin1234]

Hi Prof.Mobilux,
after reading some papers about evapotransoiration and different methods to estimate ET_p I can confirm that for the Haude methode the average temperature and humidity is incorrect and your approach to take the max value from the day is much more accurate. An average temperature would result in very similar values for evapotranspiration on hot, warm and no so warm days, e.g. 150%.

I saw the Haude with the temperature and humidity value taken at 2pm each day, which is pretty similar to the max value. It's also easier in Loxone because you can get rid of the average value.

BTW: Haude seems not to be the best method for estimating ET_p on a day to day level. I will go for the Turc method which also includes the sun radiation and is easier to calculate as well.

[Deepflash]

I am using my own weather station which measures wind, temperature, humidity, solar rad, rain amount.
Those values go into weewx (opensource weather software which runs on my loxberry) which calulates Evapotranspiration very accurately. Based on that I use a multiplication factor that determines how long the watering will last with the watering controller in Loxone. It also decides if water is needed or not and the forecast goes into the watering controller and is the second factor if water is needed or not.