Forum Replies Created
-
Posts
-
“avrdude: no programmer has been specified on the command line or the config file Specify a programmer using the -c option and try again”
Can’t think of any reason why you should see this error. I am traveling today but I will get back tomorrow to check it on my Mac.
I have expanded the zip, but see no obvious exe to launch the GUI.
In the unzipped folder, do you see a subfolder named application.windows or something like that? Go to that folder and you should find an executable program there.
No, there is nothing wrong: the icons have been removed since an early 2.0.x version and have not been added back since then… My laziness, not your fault 🙂
The first thing you want to try is to open a browser, and type in:
http://x.x.x.x:1080
where x.x.x.x is your IP from your ISP. If this doesn’t work, the app won’t work either. Can you type down the first two fields (i.e. first two x.x.) of your IP? From that we can probably tell if this is an external IP or internal IP.It’s the second way you described: the manual rain delay simply skips programs that are scheduled to run during the delay period. The issue with pushing everything till the rain delay ends is that it significantly complicates the algorithm: first, the controller will need a way to ‘cache’ the program data during the rain delay period; second, if the rain delay is very long (like several days), it will need to ‘cache’ a lot of data, and it keep pushing programs that are supposed to run normally. This is not easy to implement, and probably causes more confusion.
If the rain delay missed an infrequent program, you can use the mobile app to do an ad-hoc run of the program. Also, note that firmware 2.0.6 now supports logging, so it’s easier to find out what programs / stations have run during the past several days.
Yup, it’s being considered. The plan is to have the program headers multiplexed to support two types: a program can be either a interval type, where you specify the start time, repeat interval, and repeat count, or a multiple start time type, where up to 7 start times are supported. The interval type is useful for situations where the program needs to run many times, for example, every few minutes; and the multiple start time type is useful in situations as you described.
At the moment you can still achieve the multiple start times by setting up multiple programs. It’s less convenient, I admit, but it does allow you to set arbitrary start times.
I’m guessing that based upon some previous posts that is the lcd gets past the ‘connecting’ prompt that the Ethernet controller has initialized? Sooooo maybe something wrong with the clock signal? hehe
That’s correct. If it can get past the ‘connecting’ screen, that means the Ethernet controller can initialize successfully. The LEDs on the Ethernet jack not lighting up is definitely an issue — the green light should light up, and the orange light should blink occasionally.
About the continuity testing: the RJ45 jack is an isolated type (it has internal transformers): that means the PCB pins are not connected directly to the Ethernet cable wires.
Any new findings? I’ve seen one previous case where the 25MHz crystal (for Ethernet controller) is defective. This is very rare, but it’s a non-zero chance.
Prior to firmware 2.0.5, the algorithm used to do what your rainbird controller does: when rain sensor is activated, it shuts off the valves, and then when the rain sensor is deactivated the valves will be open again, assuming the program hasn’t terminated. So if you would like the same way, and don’t need the ‘per-station ignore rain’ feature available in firmware 2.0.5, you can flash the controller back to firmware 2.0.4.
Alternatively, instead of connecting the rain sensor to the rain sensor port, you can connect the rain sensor in series with the COM (common) wire. This way when rain is activated, it will electrically break the COM wire, but will not terminate the programs. This is a more common way to connect rain sensor on controllers which don’t have dedicated rain sensor port (and hence have no software handling of rain sensor).
Land sensor and flow sensor are not supported in the firmware. You will need to modify the source code to add support. For flow sensor, search the forum and you will find a few threads where people have discussed how to add it.
Do you mean OpenSprinkler 2.0? There is no version 2.8 yet 🙂
Yes, the station 1 triac seem to have burned, and it can certainly be replaced. If you want to do self-diagnosis, here are my suggestions:
1. use a multimeter to test the resistance between VIN-GND, and also between VCC-GND. Check the resistance values, they should be at least a few kilo-ohms. If they are below a kilo-ohm, something is shorted and probably more chips have to be replaced.
2. use a diagonal cutter to cut the top three pins of the damaged chip (it’s a triac). this chip is likely to have shorted, and cutting the leads will isolate it from the rest of the circuit.
3. try to use a USB cable to power the controller, see if the LCD lights up and displays messages.
4. if the above passes, power the controller with 24V AC and check if everything is working fine.The triac can be replaced with a soldering iron — even though it’s a surface mount component, it’s fairly large, and can be easily desoldered and re-soldered. If you need replacement parts, please send an email to [email protected].
If you’ve recently upgraded the firmware, keep in mind that each firmware upgrade will trigger a system reset and all settings are reversed to factory default. So if you’ve set static IP before, make sure you set it again after firmware upgrade.
You are right, I accidentally typed an extra 17. I’ve corrected it. 17 (3.3V) is free. Actually it doesn’t matter because all 3.3V pins are internally connected so there is no issue with them being free or not.
The 5V fan you linked to seems to draw 400mA of current. If you use it, I would recommend using a separate 5V power supply, instead of drawing power directly from the OSPi board.
the dubious was
to be obliged to use the pins in the same column 3.3VI don’t think I understand what you mean. To clarify, if you look at the picture you attached, OSPi occupies these pins:
(first 8 pins on the 3.3V column): 1, 3, 5, 7, 9 (Ground), 11, 13, 15 (these are pin numbers, not GPIO names)
(first 5 pins on the 5V column): 2, 4, 6, 8, 10
So these pins should not be used for connecting to your temperature sensor. You can use any other pin.Search ‘rain sensor’ on Amazon and you will see plenty of options. The one I am using is this:
http://www.amazon.com/Orbit-57069-Sprinkler-System-Freeze/dp/B000A7SPPUMost likely it’s the Auto Connect that’s causing the trouble. It basically tries to ping the router and see if it’s still connected. This feature is implemented in the first place because in case there is a power break, and the controller is powered on before the router is, it should be able to re-establish IP from the router.
A quick work-around is to set a static ip on the controller (note, this only works if you set a static ip on the controller, not by using IP reservation on the router), you can turn off auto-connect, and since the ip is fixed, it doesn’t need to be re-established. This is probably the best option if you are still seeing freezing problem.
You posted this message under OSPi, so I assume you have a RPi-based opensprinkler. First, open the enclosure and check if there is any visible damage to components. Check the fuse (lower-left corner) to see if it is broken. If so, it may be as simple as replacing the fuse. Next, you should unplug the RPi and use a USB cable to check if it can power on. Let me know your findings.
1. rain sensor is already supported.
2. land sensor: i’ve never seen a land sensor, but it’s certainly possible to add support for it. first you need to figure out if the land sensor is a digital (i.e. on or off) or analog sensor (i.e. continuous level of voltage output). if digital, it works just like a rain sensor, and you can use any of the spare GPIO pins to interface with the sensor. if analog, you can use one of the analog (ADC) pins on OSPi to talk to the sensor. The hardware connection should be pretty easy once you decide which pin you want to use. then you also need to modify the source code to add support. if it’s a digital sensor, you can pretty much copy what’s already there for the rain sensor.For Mac, you need to temporarily disable the security check: go to System Preference -> Security and Privacy -> click the Lock to Unlock, then choose ‘Allow Apps Download from: Anywhere’. You only need to do this once, I think, for each program. Next time you shouldn’t need to do this again.
Cool, can almost claim that this is a sprinkler controller with an integrated touch interface 🙂 Thanks for sharing!
Dan is probably the best person to ask about this, but I assume the interval_program limits the number of stations to 32. It should be easy to increase this number to anything, I am just not sure if it’s a matter of changing one macro define, or there are multiple places that need to be changed.
Regarding your wiring experiment, I think it’s important to keep the resistance of the wires low — it helps to use thicker wires, or combine multiple wires as you did.
Corvus13’s post says a solution has been found, by changing the 1-wire pin to another pin (other than GPIO 4 which is already used by OSPi).
Sure, you can send it back to the address you typed, and we will take a look at the board to see if it’s still functioning.
One of the vulnerabilities of Pi or BBB is that shorting GPIO pins can easily cause damage to the board. Arduino is much more robust in this aspect. Perhaps in the next revision I should consider adding protection circuitry to the GPIO pins to avoid shorting. Sometimes a simple resistor connected in series is sufficient to provide the protection.
