I just had to set Dec, Jan, and Feb to 0. The rest were 0.XX. But I couldn’t submit the form until I changed Dec - Feb to 1, then back to 0.
hmm. that’s interesting I created an issue here in GitHub to track this: https://github.com/jeroenterheerdt/HAsmartirrigation/issues/32
I’ve got the implementation running voor some days, first days it went pretty well but last days it is raining (netherlands) but also de irrigations is running. Using latest version. Looks like there is no update
The daily adjusted runtime is being calculated to be greater than the base index. Looking at the math I think I may have found what’s wrong. Doing the math with the values provided by my entities:
ART = W * BSI
W = | P - ET | / ETpeak
P = 0.01
ET = 0.17
ETpeak = 0.17
W = | 0.01 - 0.17 | / 0.17 = 0.941176471
BSI = 1175.3
ART = 0.941176471 * 1175.3 = 1106.2
But daily ART was calculated at 1292.0 and hourly ART was calculated at 2001.
The equation W = | P - ET | / ETpeak
seems incorrect to me. If P > ET, then W > 1. For example, if I were to get 2 inches of rain, then W = | 2 - 0.17 | / 0.17 = 10.764705882. I think this is what @bsmeding might be experiencing.
If I assume this is what’s causing my W to be > 100%, then my P would need to be ~ 0.356879945. In reality, my P is 0.1. However, if I convert 0.1 inches to mm, I get 0.254. I’m wondering if you forgot to convert mm to inches when performing this calculation?
W should be the difference between P and ET divided by ETpeak. I think the calculations are fine, but I will check them again. If P > ET the behavior should be different indeed. What is causing this I am not sure about yet, but I will look into it. It works fine on my installation so not sure.
@bsmeding without details such as screenshots of your sensors and attributes as well as version number of the addon I cannot help.
@trrevvorrloo I took a look at the code and the maths and the good news is that their are in sync, except for one important detail: If P > ET then B > 0 and W is set to 0 in the code, since there is no need for irrigation. That is not shown in the maths though, I will update it.
The component stores everything in mm and other metric units internally, not in inches. It is only converted in inches if you want to show it that way in the UI.
Given your data in the screenshots I did all the calculations myself - the calculations seem fine, but maybe a little off. I am getting a water budget of 170.59%, where the system returns 170.29%. I am getting an hourly ART of 2005 seconds, where the system provides 2001 seconds. I think this is just rounding errors, so looks good to me.
With the bucket being -0.2
as in your screenshot I get a water budget of 118% which is off a bit from what you have in your daily ART screenshot. Calculating with 118% I get a daily ART of 1392 seconds (or 23 minutes) which seems fine by me.
Calculating with the water budget that you have in your screenshot (109.11%) I end up with exactly the right numbers (daily ART= 1292 seconds).
So, it seems like the water budget is not correctly calculated based on the bucket, but even if it was a conversion error the water budget would be way off compared to the discrepancy we have now…
so I don’t really see the issue, other than that there are probably a lot of rounding errors compounding here because of the rounding in the metric system and then the rounding that happens when converting from imperial to metric and back.
Interesting, I was thinking the daily adjusted run time should never be greater than the base schedule index but it sounds like that’s expected?
Hi Jeroen, this morning the valve was running again and hit the maximum limit of one hour i have set. That while it is a rainy day. My version is behind now i see, I’m running 0.0.31.
I will upgrade now to latest version
Yes, the BSI is based on ETPeak (max of reference ET values). It represents the amount of time the system needs to run to compensate for a loss of ETPeak amount of moisture. Whenever your real evapotranspiration (as calculated by netto precipitation) > ETPeak your W > 100% so your ART > BSI. Totally expected.
All - please take note of this - we have found a major issue with the change percentage feature and have decided to remove it. I have just release version 0.0.40 which is version 0.0.29 but with added initial_update_delay
and bug fixes. It disables the change percentage feature.
This should bring us back to a working state. From there we can work to add change percentage back in. We apologize for any inconvenience caused.
@bsmeding please update to v0.0.40 or higher. I am confident the weird problems you are having will disappear.
Release notes:
v0.0.40 - disabled change percentage since it introduced bugs. Versions since v0.0.30 introduced a number of bugs. As a result, we have reverted to v0.0.29 and implemented bug fixes only from there. Versions from v0.0.40 onwards should work correctly. Change percentage feature has been removed while we work on bringing it back later. You will see references to it in the code as it has been temporarily removed from the options screen and any calculations.
Hi @jeroenterheerdt thanks i have updated the integration and will monitor what will be set at 23:00 let you know tomorrow
Your bucket is - 35. Please reset using ‘reset_bucket’ service. That should do the trick. Monitor the netto precipitation. Your bucket should be updated to: bucket=bucket+netto precipitation. If negative value in netto precipitation then bucket will of course decrease. When it is below zero you should see a daily adjusted run time as well which tells you to irrigate. Don’t forget to call reset_bucket again afterwards. See the readme.
Thanks dat dit the trick. 0 seconds now
perfect, please monitor netto precipitation
and the bucket
going forward to check it works correctly. since your bucket
is now 0, you should see your bucket = netto precipitation
whenever that daily value is calculated (by default at 23:00 or if when you call it manually if you disabled auto-updating). After that the bucket will increase/decrease with whatever netto precipitation
is.
@jeroenterheerdt Ive just started using your component this week. The framework on it is awesome and very versatile. However, I am having an issue that I am hoping you can help me out with it. It seems that the sprinkler is running every day it seems and just this morning, it went off again and it rained all last night. The yard was still pretty wet at 6am so I would have thought daily adjusted run time would not have gotten above 0 11pm last night, but it did so it could be I am understanding how it should work or I have some bad values. Any help is apprecitaed.
Paul, happy to help. Please open an issue on the github page and make sure to include the information as instructed there. Thanks.
@jeroenterheerdt, great work you have done here. Let me add my thoughts and questions:
I now set it up 4 times for 4 different zones. 2 are lawn with sprinklers, 2 are drip hoses. Later were a bit difficult to calculate as it’s not really an area that is watered, but a certain length of hedge. But I got the times almost right I guess.
.) Adjusting those values afterwards is still not possible I guess, and if you find out a full cycle needs 20% longer irrigation you have to remove/reinstall the integration, correct?
Second I noticed that all the units are missing. Looking at screenshots in this thread, and your Wiki I saw those units. They are very helpful as that way you can be sure that you don’t see imperial values when you expect metric. And debugging is much easier as well. EDIT: Noted that it can be checked in OPTIONS
Still an idea for enhancement: SI units would be totally enough: So l/min instead of liter/minute or mm/h instead of mm/hour would totally suffice, make it better readable, less cluttered and safe some space.
Third, I got a value of 0.24 for rain. Per hour? Per day? I guess it needs clarification in the lovelace card as well. Also maybe it’s a good idea to shorten the numbers after the comma for eg. “netto precipitation”. Two numbers would be enough (like all other values). 16 are just messing it up I think But this is cosmetic only of course.
Last but not least I will need to find out how long my “english lawn”'s roots needs to adopt from bi-daily irrigation to weekly irrigation (we got clay here, not sand) without drying out. And tips? I will go from every second to every third, to every fourth day etc. But how long should that transition last? I guess some experts are here for sure.
@iz3man thanks for letting me know. If you want to add a fixed % to a cycle you can easily do that in your automation, right? Or you can use the lead_time
option to add a fixed number of seconds to your irrigation.
Thanks for the suggestions on making things easier to read, I will incorporate them in our next major release - currently we are in beta with some modifications, so when we release that the suggestions you made will be there as well.
For rain - this is total for the day - should be clear from your config.
Not sure where the 16 digits come from, need to look into that.
About your last question - I am no botanist or gardener at all, so I cannot really answer that.
Thanks for taking the time to provide this feedback!