Forum Replies Created
-
Posts
-
Since the solenoid runs AC current, it’s a bit tricky to measure (more difficult than measuring DC current).
Which version of hardware do you have? The fully assembled OS 2.1, or DIY 2.2? I suspect this has to do with writing into uSD card. I will import your program and see if the issue is reproducible.
The inductor value is chosen based on the maximum current output. The higher the current output the lower the inductance. Although LM2596 is capable of 3A maximum output current, OpenSprinkler is only designed to provide up to 1A, so on the inductor selection graph of the datasheet, we are looking at the upper-left corner of the graph, and that’s 100 to 150uH.
In the meantime, this is a very entertaining video to watch:
https://www.youtube.com/watch?v=84aWtseb2-4
can’t help laughing towards the end.
For the moment can you use Chrome? Also, if you can report your Safari’s version and your OSX version, that can help us reproduce the issue.
I am working with Samer to figure out the issue, it seems to have to do with Javascript engine as vinny suggested.
The UI is supposed to use the device time (i.e. what the firmware reports) and should not be using the browser or computer’s time, for the simple reason that if you are accessing OpenSprinkler remotely from a different time zone the UI should still show OpenSprinkler’s time.
@sgruby: I tried Safari on OSX and iPad, both showing the time correctly. So unfortunately I can’t reproduce the issue…I wonder if this might be a browser cache issue. Can you try a different browser, or on a different computer and see if it makes any difference?
As something interesting I just learned yesterady: within the US, Hawaii and Arizona actually don’t observe DST — but Arizona is more complicated, because there are regions in Arizona that do observe DST: http://phoenix.about.com/cs/weather/qt/timezone.htm. I don’t know why DST is still a thing, in my opinion we should stop using DST 🙂
I just checked my OS and it picked up the new time correctly. I suggest keeping NTP on, and just reboot the controller. I don’t think this is a firmware bug.
Sorry, deleting my previous post — just re-read your message, you said the LCD displays the correct time, that means the firmware is working fine. It’s potentially a UI issue. Can you refresh the webpage, or if using app, quit the app and restart it and see if it gets the correct time?
Sorry to join late on this thread. @2wet: I believe the WiFi issue has been solved after the OSBo pre-configured image has been updated back in August. OSBo will continue to be offered and there is no plan to discontinue it at all.
@denis.french: all P8 and P9 pins are mapped out on OSBo. You can check the user manual page 9 for details:
http://rayshobby.net/docs/osbo11_manual.pdfThe WiFi adapter is placed on the side with P8 pinout, covering about 10 pins. So that region will not fit standard pin headers, but the adapter is well above the PCB, you can definitely solder wires from the PCB to connect external components.
@Ant: thanks for the update. Glad to hear that it finally worked.
I just tried it on my testing OS, and I think it’s setting time correctly based on your location. To check:
1) make sure Use NTP is flag on
2) set location to Melbourne,Australia
3) wait for the controller to update time
4) if the time looks wrong, check http://x.x.x.x/jo?pw=your_password, where x.x.x.x is your OpenSprinkler’s IP address. Look for the ‘tz’ parameter, it should be 92 (according to your current time zone and DST).@Alan: thanks for reporting it. At the moment we don’t test the uSD card slot on DIY kit, which means an issue like this can slip out of our testing. Sorry about that. We will improve the testing procedure to include uSD card test. Thanks.
The support page has some information about how you can do it: https://opensprinkler.freshdesk.com/support/solutions/articles/5000165136-extend-opensprinkler-s-hardware-capabilities
You will need to connect a PIR motion sensor and modify the source code. There are plenty of tutorials online about Arduino and PIR motion sensor, so it should be fairly straightforward.
If I remember correctly, Firmware 2.1.0 stores log files in a subdirectory named ‘LOGS’ whereas previous versions store log files in the root directory. I suspect the issue you encountered may have to do with creating the LOGS folder for the first time, and deleting it when it doesn’t exist. I will check if this is a firmware bug. But I assume after the first log file writes it should be ok. If the issue persists, I would recommend taking out the microSD card and reformatting it.
Before I answer your other questions, let me first ask you: does your OpenSprinkler have a microSD card installed? Logging requires a microSD card. Without the card, logging should return no data. If you are not sure, check the LCD screen to see if there is a microSD icon (see page 4 of user manual https://opensprinkler.freshdesk.com/solution/categories/5000022938/folders/5000099564/articles/5000403746-opensprinkler-user-manual-firmware-2-1-0-)
Then, my next question is: is the issue repeatable? Does it still persist, or is it a one-time thing?
What you described would involve pretty significant changes to the program data structure and UI. It’s better addressed by using an external script. You can write a script in Python or any suitable language that sends HTTP commands to OpenSprinkler to turn on zones in any sequence with any delay time. You can create as many scripts as you need. Please check the API documentation, particularly Section 10. Test Station (previously manual override) command:
https://opensprinkler.freshdesk.com/support/solutions/articles/5000405359-opensprinkler-firmware-2-1-0-api-documentationThere is a ‘station delay’ option. You can set station delay to 2 minutes and that will create a 2-minute delay between two consecutive running stations.
Just to quickly point out that I’ve found a fully signed USBasp driver that works in all versions of Windows. So no more messing with driver signature enforcement any more. Details are in this blog post: http://rayshobby.net/?p=9912
Ok, the long closing time often has to do with low water pressure. For example, if you have multiple valves running at the same time, it may take longer than usual to close a valve.
So far the opensprinkler.com blog posts are a subset of rayshobby.net blogs which have to do with OpenSprinkler and are what we’d like to announce to all users. Other posts (including technical topics) only appear on rayshobby.net. So if you want to see the complete set of posts, just follow rayshobby.net. Otherwise, if you are only interested in the important posts relevant to OpenSprinkler, follow opensprinkler.com
Right now the OSPi program (written in Python) is separate from the OS program (written in Arduino programming language), so the changes are not easy to migrate between the two.
While developing the sprinklers_pi program. Rich Zimmerman made a clever design that allows his program to be cross-compilable on both. That’s something I am hoping to take a look at soon, so that we can make the OSPi and OS software more consistent in the future.
That being said, the OSPi Python program brings a lot of benefits in easy extension and plugins. That’s one of the major advantages of OSPi.
So it looks like your closest personal weather station (PWS) is pws:INORTHER38. You can set this as the location and it should return the correct time zone and sunrise/sunset times.
In general, you can type in your city in Wunderground.com and click on ‘Change Station’ underneath the city name. There it shows a list of the nearby PWS stations. The first one is usually the closest. The name at the end (in parentheses) is the id. You can then use pws: followed by the id as the location name.
