[polarbearshirt]

Yep, it was the Raspberry Pi that needed it’s time updated. I updated the time zone with raspi-config and used the
sudo date -s "04 APR 2020 16:03:00"
command to update the Raspberry Pi time. Apparently if the time is off by more than a certain amount, it won’t automatically update via NTP. All good now.

[polarbearshirt]

The controllers are OSPi’s connecting via wifi. The App Version is 2.1.7 and the Firmware version is 2.1.9(1).

Emptying out the location is easy enough and I made sure to click submit in order to save, although the “Advanced” tab only gives me three options. Wunderground Key, HTTP Port, and Device ID. I don’t see a “Time” tab or an “NTP sync” section.

I have a feeling the Raspberry Pi is controlling the time and that is why I don’t see the time options. I’ll see if I can login via ssh and update the time via raspi-config.

Attachments:

1. 

   Screen-Shot-2020-04-04-at-2.46.41-PM.png
   

[polarbearshirt]

Thanks for the update Ray. I’m not surprised this is a big task. Maybe someone who has implemented something similar can offer some guidance on how it could be done more easily. Something like how home assistant does user logins. But in the end it is just an irrigation system and the worst case is having to reprogram a controller(s) if someone screws up the system.

[polarbearshirt]

Sorry for the very late response. I’m currently installing the last few OSPi controllers for our HOA irrigation system overhaul.

@dun4cheap Thank you for mentioning your master valve experience. I think I’m going to avoid un-pressurising the system when not watering. In theory it seems like you would save water but it sounds like it isn’t worth it.

It required a new landscaping company and a lot of convincing but I was able to get everyone involved to at least give it a try. I ended up going with a direct replacement of each existing controller (7 old controllers from 2003) with an OSPi controller. I could have combined some of the controllers but the upgrade is going to easier by keeping them seperate. I can set some of the zones up in slave mode if I want to make scheduling easier but all seven will be independant for the time being. I purchased 20 extenders which will be more than enough to cover the community’s ~320 stations.

To start we won’t be integrating flow meter data, although I think this would make diagnosing why we have sky high water usage much easier. It would also be useful for getting an alert if the system sees abnormal flow rates. We’ll see how important this becomes.

Just need to finish the install before the rain stops and the summer heat arrives.

[polarbearshirt]

That’s great! Glad to hear it’s almost ready. Sounds like the Zimmerman method will be replaced with a new adjustment method that is more compliant with WaterSense?

[polarbearshirt]

It looks like OpenSprinkler still isn’t on the EPA WaterSense list. I contacted the WaterSense department and asked if there is anything we can do to get OS added although I haven’t received a reply yet. Does anyone know what needs to be done to get OS EPA WaterSense approved?

[polarbearshirt]

Thank you Mike and D4C for your insight. I just looked through my spam folder and found there were replies to my previous posts.

D4C your setup sounds like a lot like what I ultimately have envisioned. Just larger. Our HOA is around 36 acres. I think your comments will be very helpful in convincing the rest of the board that Open Sprinkler is a workable solution. I like your watering increment solution. We have rivers of water going down the drain currently.

It’s interesting that you’re not a fan of master valves. I would have thought being able to turn off the main inlet and essentially depressurize the system while not watering (at least 12 hours a day) would save a significant amount of water being lost via small pipe or valve leaks. Is this something you would advise against given your experience?

I was also hoping the OS weather based adjustment was going to be effective although it’s good to know it isn’t there yet before implementing the system.

As it happens I’m also in the process of installing security cameras on our front gate as well as a mesh wifi network. If this goes well, we’ll likely add more cameras to cover the rest community. I hadn’t thought of controlling our street lights but that sounds like a good use of the hardware.

Update for where we are now:
I’ve been getting pushback from our current landscaping company, who is saying that they won’t install the OS system because they don’t think it is a “commercial solution”. I’ve asked for a more specific reason although it sounds like our main point of contact lives in the same community as one of the owners of Hunter and won’t consider doing anything else, even if it would be a more capable and less costly option. I don’t think it would be a problem to find another company to do the install but the rest of the board would prefer if our landscaping company did it so they are familiar with how everything works.

That being the case I started requesting proposals from other landscaping companies for the HOA landscaping contract. I plan to also request an installation quote for an OS system to coincide with taking on our landscaping work.

I’ve rethought the design of the system in order to make it easier to manage and require fewer parts. We have 4 POCs (point of connection, aka inlet pipe from city with seperate meter). Two potable and two reclaimed. Seeing as POC A currently has two independant controllers and POC B has three, there are multiple zones open at the same time on one POC. Seeing as this doesn’t appear to reduce water pressure below a usable level, I’m planning on combining zones so that POC A&B have fewer than 72 effective zones. Basically if there are two small turf zones, the plan is to combine them into one zone (parallel wires inside cabinet). Seems like if we’re able to find a good pair for each zone we currently have on POC A&B we’ll only need to control around 50 zones for POC A and around 60 zones for POC B.

[polarbearshirt]

I spoke with Ray last week and it appears the OS 3.0 is limited to 72 zones regardless of those zones being remote or hardwired to the controller. That said he mentioned someone else modifying the firmware for OSPi and running something like 180 zones, via remote zones/controllers. So the new plan for POC A & B is to use use a OSPi for the master controller and OS 3.0 controllers for the in-cabinet zone control. For instance:

POC A
OS Pi as master controller A1 (POC A Master Valve and POC A Flow Meter)
OS 3.0 as remote controller A2 (72 zones)
OS 3.0 as remote controller A3 (22 zones)

POC B
OS Pi as master controller B1 (POC B Master Valve and POC B Flow Meter)
OS 3.0 as remote controller B2 (72 remote zones)
OS 3.0 as remote controller B3 (50 remote zones)

POC C
OS 3.0 as master controller C1 (POC C Master Valve and POC C Flow Meter)
OS 3.0 as remote controller C2 (45 remote zones)

POC D
OS 3.0 as master controller D1 (POC D Master Valve and POC D Flow Meter)
OS 3.0 as remote controller D2 (37 remote zones)

Ray said he’ll likely modify the firmware so that OSPi controllers are able to utilize more than 72 total zones, as the Pi controllers aren’t as limited in ram, CPU, and onboard program storage. It sounds like controlling more than 72 zones might technically be possible with the 3.0 controllers but the need to do so is so rare that it doesn’t make sense to go through the very time consuming work of fully validating the ability. Basically the 3.0 controller has a limit to the number of zones it can control before it runs out of ram, CPU cycles, and/or storage, we just don’t know where that limit is. For our needs the OSPi controller should be just fine as a master controller.

I also asked Ray about the issue people are having with the new weather source data causing run times to be adjusted based on the current temperature instead of the normal Zimmerman method and he said he is aware of the issue. We didn’t go into detail but it sounds like one of the reasons for the mis-adjustment is due to inconsistent data formatting of the received weather information. From experience I know that engineers are extremely reluctant to promise a timeframe for a fix when they don’t know how much time and effort will be required. That said, I would think that this issue is near the top of the priority pile and is likely to be fixed in the coming weeks, not months.

The remaining question is if we can setup OS to monitor flow meter data and create an alarm/event/notification if the flow rate goes outside a predetermined range. For instance, if the normal flow rate for zone 2 is between 8 and 12 gallons per minute, and zone 2 is instead receiving 40 gpm, it would be ideal if we could then skip that zone and send a notification (email, text, etc.) so that we can send someone to figure out what’s going on and make the repair. I’d be interested to know if anyone has implemented any kind of leak detection with their OS.

[polarbearshirt]

I read through the OS manual and it appears one option for POC A and POC B would be to use a single main controller for each and then setup a second controller to act as a remote controller for each additional zone over 72. For instance:
POC A – 72 zones to Controller A1 – 22 remote zones to Controller A2
POC B – 72 zones to Controller B1 – 49 remote zones to Controller B2

Allowing a single flow meter to be used for each POC. That said, the distance between where our controllers are and where the flow meters will be, could require a different setup.

It appears OS can use both non powered two wire flow meters as well as three wire powered flow meters. We have fairly long distances (up to 1700 feet) between where the flow meter & master POC valve will be located and where our current controllers are located. Has anyone had luck with extended flow meter wire lengths?

If flow meter wire length is limited (which seems likely), perhaps a solution would be to locate the master “brain” controller (the one with the flow meter attached) near the flow meter and master POC valve. So instead of having the brain controllers located where all of the wires are, the brain controller would be located near the flow meter and master valve, and the remote controllers would be the controllers physically wired to each zone. For instance:

POC A – Controller A1 (POC A Master Valve and POC A Flow Meter)
POC A – Controller A2 (72 remote zones)
POC A – Controller A3 (22 remote zones)

POC B – Controller B1 (POC B Master Valve, POC B Flow Meter & 71 zones) *POC B is likely close enough for the flow meter to be wired directly*
POC B – Controller B2 (50 remote zones)

POC C – Controller C1 (POC C Master Valve and POC C Flow Meter)
POC C – Controller C2 (45 remote zones)

POC D – Controller D1 (POC D Master Valve and POC D Flow Meter)
POC D – Controller D2 (37 remote zones)

I’ve attached a map which will hopefully make the layout easier to understand. We will be installing a community wide mesh wifi network so wifi connection shouldn’t be an issue.

Unfortunately this design adds more controllers and cost but without something like a wifi flow meter I’m not sure how else to do it.

Is this likely to work or are there any other ideas?

Attachments:

1. POC-Layout.pdf

[Author]

Posts