Hardware build complete pics https://drive.google.com/folderview?id=0B6tlp1ncZgYkekMwbjA5cll5a0U&usp=sharing. Will still need to sort out the software side. This is my first adruino project, and i am quite proud how it came out. i learned a lot, especially that there is a huge difference between getting a project working on a breadboard, and actually "production-izing" it for the tough environments a dew controller may encounter in the dark, cold, humid nights. I think that the single most expensive component of the build is the box itself.
This build is a bit different from the schematics on this project.
1) No colorful leds (to keep the night vision)
2) No fan (my dew straps together draw less than 1A)
3) The toggle switches actually not connected to the arduino, they switch the dew straps from the PWM outputs to the "raw" power supplied to the box, with the reasoning is that if the electronics is dead i still can use the same box to run the dew heaters. Also, bypassing the electronics is more efficient if you just want to put the heat strap on max output
4) No fuse on the box - my fuse is upstream of the power connector, and especially since the nano is not connected to the 12v+ i think this will be enough for my setup
5) The reference temperature probe is a dht22 and is outside the box, this way it can be a much more reliable source of truth, and no need to set offsets to account for the heat inside the project box
6) A lot of the wiring is done directly between the outputs/connectors on the project box
7) All of the wires between the connectors on the box either go to the terminal connectors, or to a homebrew pin headers on the board - with the goal of making it easy to take the board out and making it easy to make changes to it
8) One of the lessons learned - do not buy the cheap stereo connectors, they are a mess, better get any old cable with a stereo connector (the stuff that you normally get with any point and shoot camera with a stereo -> 2 rca works great, just chop off the rca connectors and solder to the temperature probe)
9) It was very hard to cut out the hole for the lcd, this is why i taped the edges of the hole with electrical tape, to hide the ugliness. i think i will try that again, maybe just 3d print that. come to think of it, i have access to a 3d printer, it may be the best solution for the future projects.
10) The knob on the right of the box is a 10k resistor which controls the brightness of the lcd, to preserve the night vision. It is connected directly to the pins on the i2c module
Once fully done, i plan to post the full schematics of the build. on the software side i also want to make this support the temperhum protocol, so that my backyardEOS could read the temperature from the main environmental sensor
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Alex
Looks great, congratulations. Yes the case is probably the most expensive part.
In the revised build, the connections are now made with XH2 2P/3P headers/sockets on the board, makes connection/disconnection very simple.
You have made a few customizations which is really great. Thanks for sharing
Regards
Robert
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Wow, did you come up with the revised build since my post or you were working on that from before ? that looks very simple and clean, with the xh2 connectors. Not sure if putting the resistors on the stereo sockets is not going too far though, in my layout it would be still easier to put them on the board
as for the schottky diodes, i assume these are to resolve some issues coming from pwm + long cables ? does that create some kind of radio interference or capacitance issue on the cable discharging back into the main circuit (i don't really know what i am talking about, yes ?). Because my cables are about 1.8 meters and anything shorter would create a problem in my setup, a thought came to me and it may be completely off base, but why not bypass pwm all together and turn pins 7/8 on/off within the main program loop ? if we are looking at 20/40/60/80/100% "dials" then we can just have a counter which will do manual "very low frequency" pwm within the main program loop (for 20% turn on for one cycle and turn off during the other 4), its not like we are running a motor off these outputs, we are just controlling a heater
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I had been working on it for about 3 weeks and ran out of TIP120 transistors - as you can see they are missing on the photo - waiting for some to arrive from ebay. I have also used the XH2 headers/dockets on myFocuser design, I think I have already posted some pics there of my new easydriver build with a rotary encoder,
The schottky diodes - I sometimes think overkill - in reality probably not needed, voltages and currents are low, and my own cables are about 1.5m, so I don't think they are needed (yes, for back emf caused by switching on long impedance cables - in reality the dew controller cables are probably not be high impedance). So they act as protection for the TIP120 (even though the TIP does have protection diodes built in - an extra form of safety net).
Turning pins 7/8 on would be the same as using PWM - its the switching off that causes the issue, not necessarily how often the switch off occurs. Whenever you have current flowing in a loop, and then interrupt that loop to stop the current, there is going to be a reverse effect. The question is how big is it, how long is its duration, and what damage might it cause.
The schottkys are overkill because the TIP120's already have reverse protection built in. So they are only optional, not mandatory.
Cheers
Robert
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hardware build complete pics https://drive.google.com/folderview?id=0B6tlp1ncZgYkekMwbjA5cll5a0U&usp=sharing. Will still need to sort out the software side. This is my first adruino project, and i am quite proud how it came out. i learned a lot, especially that there is a huge difference between getting a project working on a breadboard, and actually "production-izing" it for the tough environments a dew controller may encounter in the dark, cold, humid nights. I think that the single most expensive component of the build is the box itself.
This build is a bit different from the schematics on this project.
1) No colorful leds (to keep the night vision)
2) No fan (my dew straps together draw less than 1A)
3) The toggle switches actually not connected to the arduino, they switch the dew straps from the PWM outputs to the "raw" power supplied to the box, with the reasoning is that if the electronics is dead i still can use the same box to run the dew heaters. Also, bypassing the electronics is more efficient if you just want to put the heat strap on max output
4) No fuse on the box - my fuse is upstream of the power connector, and especially since the nano is not connected to the 12v+ i think this will be enough for my setup
5) The reference temperature probe is a dht22 and is outside the box, this way it can be a much more reliable source of truth, and no need to set offsets to account for the heat inside the project box
6) A lot of the wiring is done directly between the outputs/connectors on the project box
7) All of the wires between the connectors on the box either go to the terminal connectors, or to a homebrew pin headers on the board - with the goal of making it easy to take the board out and making it easy to make changes to it
8) One of the lessons learned - do not buy the cheap stereo connectors, they are a mess, better get any old cable with a stereo connector (the stuff that you normally get with any point and shoot camera with a stereo -> 2 rca works great, just chop off the rca connectors and solder to the temperature probe)
9) It was very hard to cut out the hole for the lcd, this is why i taped the edges of the hole with electrical tape, to hide the ugliness. i think i will try that again, maybe just 3d print that. come to think of it, i have access to a 3d printer, it may be the best solution for the future projects.
10) The knob on the right of the box is a 10k resistor which controls the brightness of the lcd, to preserve the night vision. It is connected directly to the pins on the i2c module
Once fully done, i plan to post the full schematics of the build. on the software side i also want to make this support the temperhum protocol, so that my backyardEOS could read the temperature from the main environmental sensor
Hi Alex
Looks great, congratulations. Yes the case is probably the most expensive part.
In the revised build, the connections are now made with XH2 2P/3P headers/sockets on the board, makes connection/disconnection very simple.
You have made a few customizations which is really great. Thanks for sharing
Regards
Robert
Wow, did you come up with the revised build since my post or you were working on that from before ? that looks very simple and clean, with the xh2 connectors. Not sure if putting the resistors on the stereo sockets is not going too far though, in my layout it would be still easier to put them on the board
as for the schottky diodes, i assume these are to resolve some issues coming from pwm + long cables ? does that create some kind of radio interference or capacitance issue on the cable discharging back into the main circuit (i don't really know what i am talking about, yes ?). Because my cables are about 1.8 meters and anything shorter would create a problem in my setup, a thought came to me and it may be completely off base, but why not bypass pwm all together and turn pins 7/8 on/off within the main program loop ? if we are looking at 20/40/60/80/100% "dials" then we can just have a counter which will do manual "very low frequency" pwm within the main program loop (for 20% turn on for one cycle and turn off during the other 4), its not like we are running a motor off these outputs, we are just controlling a heater
I had been working on it for about 3 weeks and ran out of TIP120 transistors - as you can see they are missing on the photo - waiting for some to arrive from ebay. I have also used the XH2 headers/dockets on myFocuser design, I think I have already posted some pics there of my new easydriver build with a rotary encoder,
The schottky diodes - I sometimes think overkill - in reality probably not needed, voltages and currents are low, and my own cables are about 1.5m, so I don't think they are needed (yes, for back emf caused by switching on long impedance cables - in reality the dew controller cables are probably not be high impedance). So they act as protection for the TIP120 (even though the TIP does have protection diodes built in - an extra form of safety net).
Turning pins 7/8 on would be the same as using PWM - its the switching off that causes the issue, not necessarily how often the switch off occurs. Whenever you have current flowing in a loop, and then interrupt that loop to stop the current, there is going to be a reverse effect. The question is how big is it, how long is its duration, and what damage might it cause.
The schottkys are overkill because the TIP120's already have reverse protection built in. So they are only optional, not mandatory.
Cheers
Robert