[zim]

Makes sense. Based on that explanation, the wiring for the rain sensor may be closer to the sensor vs. the controller… another project for a different weekend. (for now, I connected as-is.. and that is good enough to make it work 🙂 .. but long term I want to remove the physical rain sensor and start using the online logic)

Also wanted to document for the sake of other potential users that OS3DC was a drop-in replacement for the Orbit 57592. Swap was very easy, biggest puzzle was realizing the rain sensor wasn’t directly connected to a pin on the Orbit, and finding the pinout for the 12-station version (most search results pointed to a different version)

thanks!
Mike

[zim]

thanks.. not previously familiar with MQTT… reading and learning how to install a local MQTT on my linux box.

[zim]

Perhaps a verbose logging option? ie. would love to see all scheduled events in the log. If they happen to end up at 0%, I’d prefer to see that to help debug that I didn’t make some other error (ie. else I end up wondering why my expected program didn’t run… which appeared to happen with my very 1st attempt 🙂 )

I confess that might be a trick for the graph, but maybe not the table?

[zim]

I reset Wifi and made sure to connect to the specific satellite (ie. specific BSSID). The system now reports Good/Excellent.

I expect that to result in a long term solution.. ie. I’m not moving the Orbi satellites around… and I’m not moving the OS unit around. So with Wifi.Begin always choosing that one satellite (based on BSSID), that should do it.

I dunno.. might be worth a mention in the FAQ for people who have Orbi (or probably any mesh system??) .. ie. the user needs to pick based on combination of SSID and BSSID (it makes perfect sense now, but I have 30+ devices connected to my network and have never needed to think that way before). FWIW, I checked and Orbi doesn’t allow one to pre-select/associate an endpoint to a satellite.. the endpoint has to do the picking.

Other than that, I’m a debug freak and would prefer more verbose logging, but will take that back up in the logging thread I started 🙂

thanks!

[zim]

Broke out the tape measure… the OS unit is placed within 25′ of 3 Orbi satellites (15′, 22′, 25′).

Possibly Wifi device in OS3.3 isn’t going a good job of picking the strongest satellite? (ie. why would the signal quality be varying from Poor to Good… interference/shielding???)

I see mention of the Wifi antenna option on :

https://opensprinkler.com/forums/topic/external-antenna-option-for-os3-2/
https://opensprinkler.com/forums/topic/wifi-distance/

are there details on how to get this? (support ticket), and what kind of antenna it would accept (and where would it plug in?)

thanks
Mike

[zim]

ignore the previous post (says its awaiting moderation and then the forum won’t let me edit b/c of too many redirects (deleted cookies to no avail)

I think there may be a software solution for Orbi-type systems … sanity check me here

it appears that WiFi begin accepts a BSSID argument….

A work-around for this would be to display all BSSIDs on the Wifi page (ie. I’d see 4 entries for my network) and instead of joining by SSID, I join by BSSID (ie. mac address of the particular Orbi station). Both the OS3 and Orbi are fixed.. they aren’t moving around… so that guarantees that the OS3 will connect to the Orbi with the strongest signal.

Plausible?

(and if some of this already exists, apologies, I’m still trying to learn my way around and my CPP is pretty rusty.. I see that some of the start_network calls actually use a BSSID)

[zim]

System installed (by me) Friday. Program added for 630a, 7 days/week. Weather enabled. Time/date info was configured for my time zone, and was correct.

My first automatic program (per above) did not run. and nothing in the logs. No evidence of the program being seen, aborted, canceled by weather, etc. Totally empty log (for that date)… Log did show my manual testing on Friday, but blank for Saturday.

For debug, I disabled weather and changed the time to 1237pm (was debugging ~1230pm) .. program then ran and logs recorded it.

I did not update anything about the “time”.. the time was correct all along. There was a program for 630a (configured the day before) and it did not run, and there was nothing I could find in the logs to indicate why not.

The “stranger” part to me was the total lack of log info (as the manual suggests the logging is pretty verbose). Again, my log was empty for Saturday. Wondering is there any corner case where a program is “seen” and aborted where no log entry is generated?

(sorry, I’m repeating that, but trying to be thorough to each question).

[zim]

I updated the FW to 2.2.1 (0) last night and now the system is reporting “good” for Wifi.

Attached a pic of Wifi signals from my phone (Orbi system with 4 units). Only thing I can think is that the OS is sometimes not picking the strongest satellite (reading up more on that concept).

Please tell me more about the antenna plug? Is that an add-on to the 3.3DC board I just purchased? (I’m also trying to get a schematic viewer and not above breaking out my soldering iron if needed)

thanks!

EDIT: Spoke too soon.. Seemed like it was “good” since the udpate.. and now I just saw Fair and Poor again 🙁

Attachments:

1. 

   Screenshot_2024-08-10-18-44-48-81_a863a16a4498b8f1500a6e0daff4aeb9.jpg
   

[zim]

Additional info

I downloaded WiFi Analyzer (Android) and put my phone right next to the OS3.3. My phone reported -40dB, which from what I read should be very good.

The LCD actually shows 3-4 bars, yet the System diagnostic reports Poor and/or Unuseable.

Is my unit defective?

Attachments:

1. 

   IMG20240810165846.jpg
   

[zim]

I changed the time and disabled the weather adjustment. It ran as expected, at the expected time. So I believe I can rule out time.

So the question that comes to mind is

– if a program is supposed to launch and (insert reason here) it decides not to run, will there be a log entry? ie.
—- my system reports weak Wifi (which is another mystery) .. if it was trying to get weather and could not would it abort? and again, would it log the reason?

EDIT: Also no rain sensor
EDIT2: the UserManual says it logs “zone runs as well as rain delay, sensor status changes, and watering percentage changes.” .. which makes this all the stranger (to me)

[zim]

yes… my router says its connected to 2.4GHz (Orbi does both bands in parallel).

2.4GHz should penetrate walls better.

Which makes it all the stranger (to me) that it reports the signal being so bad despite being within 15′ of the Orbi satellite.

So again I ask, is the reception expected to be this weak? Can it be improved?

[zim]

I finally installed my OS3.0 DC-powered to replace an Orbit 57592 (12-station model, 5 in use).

Only thing that gave me pause is that the Orbit didn’t have a rain sensor input, apparently they may have wired that in to the common bundle (hidden behind a drywall by a previous owner).

So for now I have disabled the mechanical rain sensor, connected the 5x DVF wires to #1-5 and all appears to be working. Next is to start playing with the different rain algorithms which might(?) render a mechanical rain sense as meaningless anyway… along those lines, other than searching forum posts, is there a detailed FAQ to explain Zimmerman vs. ETO vs. ….. (even better with examples). Personally I don’t trust forecasts (they consistently seem to mis-predict rain for us).. but I would be perfectly content in reducing watering b/c the “back-cast” knows it rained last night….

[zim]

I have the DVF100 : https://www.rainbird.com/products/dvdvf-series

Specs at the website say: 24 VAC 50/60Hz (cycles per second) solenoid power requirement: 0.450A inrush current, 0.25A holding current

Which version did you go with?

If both AC and DC work, is one “better” ? (DC appears to be slightly cheaper)

thanks!

[Author]

Posts