Forum Replies Created
-
Posts
-
Does your controller have a yellow PTC fuse? It’s known that in some cases when the ambient temperature is high, the resistance on the PTC fuse will rise significantly to block the current flow. The easiest solution is to use solder a wire across it to bypass the fuse. Let me know if this solves the problem.
1.35V is ok for D1. Different multimeters will give different results. D1 is a diode with fairly high(up to 3V) forward voltage. So I think D1 is fine.
Forgot to say that I can also send you a replacement. But I would like to check your returned board first to figure out what’s causing the issue. Thanks.
Agree with mrburns42.
My suspicion is that IC1 is probably damaged and should be replaced. D1 and D2 are unlikely to be damaged, but just to be sure, you can use your multimeter to measure the diode forward voltage on D1 and D2. D1 (R3000) should be somewhere around 0.8V to 1V. and D2 (1N5819) should be around 0.1 to 0.2V. If not, you should replace them as well.
If the controller works with ospi_manual and selftest, then I am pretty sure it’s not a hardware problem, but rather a software problem. Since the google calendar program requires persistent Internet connection, you may want to check if that could be the cause of the problem.
Not sure why this happens, but you are welcome to return the product. Just send me an email at [email protected]
The zener diode is optional (although it’s useful for over-voltage protection). You are right that USB power is sufficient to test the microcontroller and the web interface.
It’s very odd that your initial voltage testing passed but then components got damaged when other ICs and LCDs are plugged in. In any case, you can easily test if these components are still working by powering the controller through USB. The worst case is that the LCD and Ds1307 may be damaged (because they are connected to the +5V VIN line), but I assume the zener diode did protect them and hopefully they are fine.
The ntp server ip is currently hardcoded in the firmware (you can change it by recompiling the firmware and then uploading it to the microcontroller).
To implement what you need, you can use the EtherCard::browseurl function. The EtherCard library has a few examples to demo this, for example, the webclient example:
https://github.com/jcw/ethercard/blob/master/examples/webClient/webClient.ino
shows how to use browseurl to periodically get a webpage and send the content to serial.I see, so the 2.5V you measured is not through the linear regulator. I assume once you have a working MCP1700-33 soldered, it should measure 3.3V.
If you need to purchase any components, you can either follow the part list to buy from Digikey, or I can send you any requested components at a small cost (just email me at [email protected])
I haven’t used Zap Controls so I am not sure if that could be the cause of the issue.
One thing you can try is to unplug the expansion board (detach the extension cable), and then retest the main unit. See if this makes any difference.
2.5V Vcc means the MCP1702-33 linear regulator is probably damaged. You will have to replace it.
Also, was the 4.97V measured when it’s powered by USB or 24VAC? If by USB, there might still be a problem in your 24VAC -> 5V conversion (i.e. the switching regulator section) that caused the damage in the first place.
That is possible but clearly you have to modify the source code. Basically you can modify the code to periodically retrieve information from a server and perform actions accordingly. In fact, the firmware already has a component, namely the NTP sync, which is similar to what you want to do: it periodically retrieves the current time from an NTP server.
Alternatively, you may want to check out the OpenSprinkler Pi, which has a demo program that makes use of the Google Calendar for scheduling sprinkler events. It’s a polling-based method, and the
The likelihood of D2 shorted is pretty small. D3 may be damaged because it’s a zener diode, and if the voltage presented on it is way above 5.1V it will be burned. So try to take off D3 first (just cut the leads from the front side of the PCB, you can always re-solder a new D3 back by soldering it from the front side of the PCB).
The likelihood of PCB trace problem is extremely small: all PCBs have been 100% tested. We use the same manufacturer for assembled OpenSprinkler boards and have not encountered any PCB trace problem so far.
If the LCD never gets past the ‘Connecting to the Network’ message, it means the Ethernet controller fails to initialize. This has nothing to do with a valid Ethernet connection or not: if the Ethernet controller is initialized correctly, it should eventually get past that screen and display the current time (even though the time may be wrong).
There are several reasons that could cause this, for example, the pins of the Ethernet controller were not soldered reliably, the Ethernet controller is not plugged in properly, or a pin might be accidentally shorted to ground or something. Also check if the 25MHz crystal is soldered to the correct position (make sure you did not solder it to the position of X1). Does the orange LED ever blink?
Have you checked this section of the user manual:
http://rayshobby.net/?page_id=730#httpgetYes, that’s correct, and as you can see, those files have extension names .pde, not .ino, right? I suspect that maybe you have opened them in Arduino 1.0, which automatically changed the extension name to .ino.
In any case, I highly recommend you to use the VirtualBox image that I created for compiling the source code. Please follow the instructions:
http://rayshobby.net/?page_id=732#compileHmm, where did you download the source code? The source code published on OpenSprinkler GitHub are .pde files, not .ino.
If you have an existing Arduino, you can use it as an ISP programmer to program the OpenSprinkler. Here are the instructions of how:
http://arduino.cc/en/Tutorial/ArduinoISP
Although it will be quite tricky because you will need to layout the wires yourself. So it’s probably still easier to get a DIY programmer from my website.You cannot plug in the OpenSprinkler’s microcontroller to your Arduino and program it that way. The reason is that Arduino itself does not have an ISP programmer — it only has a FTDI chip on board, hence it requires the chip to have a bootloader which takes too much space for the latest OpenSprinkler firmware.
The difference of an ISP (in-system programming) programmer is that it directly writes to the microcontroller’s flash without any bootloader. Therefore it takes zero bootloader space.
Well, if what you described is true, then the datasheet should list the ‘minimum trigger current’, right? But it only shows maximum trigger current. I think there is a difference between ‘maximum trigger current’ vs. ‘maximum current’, in that the former refers to the maximum current that guarantees to trigger the triac, while the latter refers to the maximum current that can be tolerated. Also, to be honest, I haven’t seen any transistor that would only tolerate 5-7mA of current — that is rather small for any semi-conductor component.
Thanks for sharing. I am traveling this week and have very limited connection. I will mark the post and take a look at your code after I get back. Thanks.
Note that hardware 1.1 does not have built-in USBtinyISP programmer. Are you using an external USBtinyISP programmer, or are you using FTDI cable? As described in the firmware update instructions, you must use an ISP programmer (FTDI cable does not work any more due to the size of firmware 1.8.3):
http://rayshobby.net/?page_id=732Yes you should remove the linear regulator because that’s connected to the VIN line and it’s very likely to have been damaged.
To solder reliably, you need to keep the soldering iron hot. Otherwise it may not reflow the solder well, causing fake solder joints. Normally fake solder joints won’t cause a big problem, but in some cases they will. For example, if the ‘Feedback’ pin of the switching regulator is not soldered well, it will cause the regulator to falsely sense that the output voltage has not yet reached the desired level, hence it will keep increasing the output voltage and that will potentially damage the components connected on the output line.
As described in the user manual, if you change the time zone, you should restart the controller to get the new time. This is because every time you restart the controller, it will perform an NTP sync immediately using the time zone in your settings. Now, if you don’t restart the controller, eventually it will still get the new time because the NTP sync is performed every 24 hours by default.
another thing to keep in mind is that the ‘maximum gate trigger current’ refers to the current that guarantees to trigger the conductance of the triac. For example, the gate current that guarantees conductance is no more than 7mA. You can certainly supply higher gate current to make sure it reliably stays on, but 7mA is sufficient to make the guarantee. The maximum gate current is the I_GM (i.e. without damaging the triac), which is usually a lot higher (up to hundreds of milliamps).
