Poured most of the night till 3am, but no delay this morning


#1

What is going on with the water delay? Woke up this morning to see that it still ran twice. Last night into the early morning was the worst rainstorm I’ve seen in this area for a long time. 12 minutes for my shrubs and another schedule of 35 minutes for the rest of my zones at 4 am. The same schedules and settings had no problem delaying a few days ago. This is quite frustrating and this delay is the whole reason I got the device.


#2

More info: I wasn’t using a PWS. Can the system check the last 24 hour rainfall as well? It rained late yesterday afternoon at 5 pm as well.


#3

I had 3-4 inches of rain yesterday. Looks like my system would have happily watered this morning except for the fact that I had manually entered a rain delay … Yikes!! If I understand correctly I could create an IFTT recipe to look at past rainfall(?), but … Wow… Shouldn’t looking at rainfall over the last x hours be part of the standard functionality of the device (or am I missing something?).

TK


#4

Quite possibly the bigger storm missed the airport (SLC) that the IRO defaults to.(10 miles) I may just have to go with a PWS nearby my house. I hope it takes into consideration the last 24 hours.


#5

i would still attach a rain sensor to the device. also email support to verify there is not a bug, but it is very very plausible that rain data either didnt make it to rachio (as can happen with micro climates) or it was delayed.


#6

Weather intelligence had an issue starting yesterday around 11am and was not recovered until last night about 11pm, apologize. First time we’ve encountered issues with it. There were no reported failures, so took a while to track down. Building more monitoring into that system :wink:

Everything is working great again, let us know if you have any other questions or feedback.

:beers:


#7

11am Mountain time?


#8

@franz why wasn’t an outage posted to http://status.rach.io/


#9

Yes, looks like database failed over to new instance, weather intelligence didn’t get the memo :wink:


#10

Problem wasn’t discovered till late last night, still trying to uncover root cause,etc.


#11

shouldn’t it still have been posted?


#12

Yes.


#13

Our major storm was about 2am mountain time. 3 hours after the system was fixed that was mentioned above.


#14

Ok, if you reach out to [support@rachio.com] they can help review your watering time. Also, there should be a notification in your event history as to the observed + predicted precipitation from the station, curious as to what that is. We go back 24 hours, and get the current forecast and sum the values. It might come down to the weather station reporting.

:beers:


#15

when i have had the same problem, it was exactly this…it rained like crazy here but not at the station. pretty typical here in my area, we will get .5 inch of rain about 100 yards to my north west, but not here. i could switch to a different weather station but looking at the histories of my local stations i still feel the one i have chosen is the better of the 4.

which is why im all kinds of fired up about the new ifttt events you all are adding.

i suggest an external rain sensor and an ifttt rule to impose a 24/48 hour rain delay if it trips as that will handle some valid/natural anomalies. my sensor is set to ~1/8th inch rain and that is where my weather intelligence is set as well so i feel secure with imposing an artificial restriction.

personally, it the complexity was not approaching “np”, i would like the sensor tripping to impose a configured amount of percip in the system for my area when computing felxies. but that will cause the code to grow bat shit crazy in a hurry…


#16

I’m struggling with weather station accuracy in the Salt Lake valley as well. I live in South Jordan, so KSLC weather is never going to be right for me. I also happen to live near a river and sit down in a little gully so my temps are always off by 2-10 degrees (usually cooler). I definitely feel like moving away from Forecast.io api was a loss in terms of accuracy in my particular situation.

I’m currently testing MID_E3770 which seems to have decent accuracy and is much closer to me.


http://www.findu.com/cgi-bin/wx.cgi?call=EW3770

I’ll also go get a rain sensor today, and hack together a safety net using IFTTT (which, frankly - shouldn’t be needed). I was hoping to avoid a PWS, but it may be the only way to get the accuracy I want… If I can’t start capturing value (primarily Iro reducing water consumption due to weather) it will be difficult for it to ever “pay itself off”. Not to say I’m not happy, I love the product as a gadget - but I’m also interested in the financial and environmental benefits of Iro. :smile:


#17

just try to keep 2 things in mind, there are so many things out of rachio’s control which can jeopardize the their accuracy and as complexity grows the need for fail safes increases. ifttt is one way to provide any number of fail safes.


#18

These can really help with micro climates, and if you can find one near your house that you trust, even better.

:beers:


#19

After my post this morning I switched to the one closest to me to the east. MID_AT921 I still went ahead and notified support with all the requested information. What rain sensors are you looking in to? I too am looking into a rain sensor as a backup to these stations.


#20

This is what we use for internal testing, don’t know how well it works in the wild, but the price is right.