Forum Replies Created
-
Posts
-
That’s a good suggestion, but for now I have to stick with the USB type B connector. The injection molded case is already in production, so I can’t change the cutout size now. Also, one supported feature is to power a WiFi adapter through OpenSprinkler’s USB port (http://rayshobby.net/?p=4233), and I worry that a micro USB connector would be too fragile and easy to break if you insert adapter to it.
Did you look at my second reply above? I still suspect that maybe your controller is not getting NTP sync signal.
One easy test to try is to take out the RTC battery (the CR1220 coin battery), wait for a minute, and then put the battery back and restart the controller. If it doesn’t get your current time, it means the controller is not getting NTP signal, that’s why it cannot update its time.
There were some efforts to make use of a Raspberry Pi to serve the Javascripts:
http://rayshobby.net/?p=5279
Basically the Raspberry Pi runs a small webserver (something as simple as a Python HTTP server is sufficient), and then you can modify the OpenSprinkler code to use Javascripts stored on your local server.If you have no Internet, the webpages won’t render correctly, so you can’t change any settings through the web interface. But you can still send http command. Basically the Javascripts are used to render the webpages, and to construct an HTTP GET command to send to the controller. If you know how the HTTP GET commands are constructed, you can send it without the webpages at all. The list of commands are documented here:
http://rayshobby.net/?page_id=730#httpget
It is very technical, so I don’t expect anyone to use these commands manually (unless if you write a program to do it).You only need Internet to set the programs (because some of the Javascripts required to render the webpage are stored on a server). Once the programs are set, the controller runs on its own without requiring Internet.
DS1307 and DS1302 have different pin layouts, and use different code. So you could use it but there will be a bunch of changes you have to make.
Well, it’s not the crystal body that attracts noise, but it’s the trace between the crystal and the DS1307 needs to be surrounded by ground trace. You can even solder the crystal directly onto the lower left two pins of DS1307, to hopefully reduce the trace length.
This means the RTC is not advancing time. In light of my blog post yesterday (http://rayshobby.net/?p=5962) one thing to try is to adjust the position of the 32.768kHz crystal, and resolder its pins to make it as close as possible to the PCB. If this doesn’t work, there may be an issue with the crystal, and I can send you a replacement crystal if you email me your mailing address. Thanks.
I am fairly confidence I’ve figured out the source of the problem, and reported it in this blog post:
http://rayshobby.net/?p=5962Anyone who has received the 2.0 development board and is having issue with it can email me at [email protected] for a free replacement. Thanks for your patience!
So you are looking for ‘pumps’ not ‘valves’. I don’t know much about pumps, but if it works with 24V AC, and the operating current is below 800mA, then it should work with OpenSprinkler.
The only control feature on the main unit is the self-test, which turns on each station for a specified amount of time one by one. This is activated by pressing pushbutton B1 while starting the controller.
I am curious what kind of controls you want your landscaper to have. If you want the landscaper to be able to set a complete set of programs on the unit, you probably should look at other options. The OpenSprinkler is designed for web-based programming, and its 3-button interface is not suitable for setting programs manually on the unit itself.
Yes, one wire from 24VAC transformer is the COM (common) wire, and the other wire should be tied to ground. AC transformer is no-polar, so it doesn’t matter which wire you pick as ground.
The code does not use RTC at all.
Well, I don’t have an estimate for how long it will take for DIY 2.0 to be ready. I don’t even have prototype DIY 2.0 yet. So it’s up to you whether you want to wait. Also note that the initial 2.0 firmware will be functionally the same as the current 1.4.
The backlight is LED-based, and 10mA is enough to make LEDs look really bright.
2.0 will come out soon as assembled (we’ve already distributed a development version to recent users who ordered 1.4s). However, DIY 2.0 will not be available soon, because it has to go with the new injection molded enclosure and we are still waiting for them to come. The current enclosure is too small for DIY 2.0 (which will adopt ATmega644 that is significantly bigger than ATmega328).
Also, if the time doesn’t change, that means it’s not getting NTP sync signal. Check if you have a valid Internet connection, and make sure the NTP port (123) is not blocked by your router.
The user manual, and also the one-page instruction you received has this line:
Setup Options:
***Once you are done, hold B3 again, until the controller restarts. Your changes are now saved and applied.If you don’t do this, the changes will not be saved.
NOE is the shift register Output Enable (active low) pin. A logic high disables output, and a logic low enables.
If you don’t need 24VAC->5VDC conversion, and you don’t need RTC, then you can get the OpenSprinkler Zone Expansion Board, which requires a minimum of 6 pins (3.3V, GND, and four shift register pins) to interface with Pi.
Unfortunately there is not enough code space to add such a feature to the 1.4 hardware. If you need to change the Javascript path, you can modify the source code and re-flash the microcontroller.
It’s difficult to pinpoint the exact reason without seeing your board and soldering quality. Here is a reference image of the soldered board:
http://rayshobby.net/wordpress/wp-content/uploads/2013/04/svc142_build_back.jpgThere are a few tests you can start with:
– measure the resistance between VIN and GND, and also VCC and GND. If any of them is small (less than kilo-ohm level), there is probably a short somewhere.
– since your IC1 is burned, you should also check the related elements, particularly D1, D2, D3, and see if they have been damaged in any way. To check these diodes, use a multimeter to measure the forward drop voltage. D1 and D3 should have 0.5-0.7V, and D2 should have around 0.2V. If the forward drop voltage is shown to be close to 0, the diode is burned and needs to be replaced.
– Try to power the circuit board through USB and measure the voltage between VIN and GND, and also VCC and GND, see if they are roughly 5V and 3.3V respectively.Regarding the three 220uF capacitors, I apologize for the issue. Just to make sure: since the 220uF and 10uF capacitors are very similar in appearance, did you check the label on the capacitor body to see they are actually all 220uF?
Thanks for pointing out the missing Rain Sensor step. Will add that to the instructions later.
The backlight is hard wired so there is no way to turn it off. It may seem a waste but the power consumption is actually less than 10mA.
The upcoming 2.0 has backlight controlled by PWM so the brightness can be adjusted.
One quick question: everyone who encountered issues with 2.0 development board: could you check the PCB and let me know if you have a black PCB or green PCB? They have slightly different schematics. If I can identify it’s always the green version that leads to the issues, that can help me narrow down the problem. Thanks.
