Microcontrollers > Microcontrollers/General
Arduino inverter load controller
DamonHD:
On the Arduinos (eg UNO) you can choose to measure analogue inputs against the the fixed (bandgap) reference rather than the supply voltage, which can deal with that issue.
Rgds
Damon
frackers:
--- Quote from: DamonHD on March 02, 2016, 04:37:38 PM ---On the Arduinos (eg UNO) you can choose to measure analogue inputs against the the fixed (bandgap) reference rather than the supply voltage, which can deal with that issue.
--- End quote ---
But be sure that the voltage to be read does not exceed the reference (2.56v in the case of the bandgap). Potentially means a change to the scaling resistors you are using.
OperaHouse:
Since you mentioned water heating, here is my proposal. The C-45 is a basic ON/OFF PWM
controller. When off the panel voltage rises to the open circuit voltage of the panels.
Solar panels are basically current sources. When the PWM turns on, the panens are connected
directly to the solar panels. Current is limited to what the panels can produce and the
panel voltage is that of the battery. Not he most efficient method but it works and is reliable.
The maximum power a panel can produce is the product of volts times amps. The maximum is
called the power point voltage. This voltage actually does not vary much with the amount of sunlight,
but with the temperature of the panel. Panel temperature can be monitored and the voltage tracked.
Actually using a fixed voltage adjusted seasonally does not lower efficiency that much. In the
case of heating water it is often advantageous to be on the high side of the power point voltage
because the fixed resistance of the heating element is generally higher than ideal. In most
situations the PWM of the heater will be at 100% for long periods of time.
My proposal is to take those periods of time when the panels are at open circuit voltage and
charge a small capacitor bank. When the voltage of that capacitor bank exceeds the power point
voltage of the panels, that power is PWMed into the water heater. During normal C-45 on periods
the charge controller sees no difference in operation, but we have harvested energy during the
off periods. The diode is absolutely necessary to provide isolation. Without it peak currents
would exceed what the controller would handle. The independent micro controls everything and
can power down when there isn't any sun. In this case a wall wart power supply for the micro is
ideal.
This is similar to what I do at my camp. With about 1000W of panels I am able to divert over
2.5KWH to water heating after supplying daily household needs. Remember this is power most
systems waste every day because solar systems have to be over designed in order to function
on minimal days. For you this allows water heating without tapping into the reservs of the
super cap storage system.
The control system I've used is fairly simple. I have an up/down counter that is entered into
the PWM output. If voltage is above the setpoint the count is increased, below it is decreased.
Of course there is a little deadband to keep it from banging around. For values weel beyond
the the count is modified by a much higher number to increase the response time from major
changes in light level. Pwm count is not allowed to go above 255 or below 0. I use opto
isolators to drive the FET and shift gate levels to 12V. Optos are slow, in fact turn off
is four times slower than turn on. It doesn't matter. 490Hz PWM is pretty slow and I don't
experience FET heating. Drive values below 5 are just set to zero. Above 250 the drive is
just set to fully on, 255. This prevents narrow pulses at either end which mostly result in
FET heating. Use a high speed driver if you want. You will likely be shooting yourself in
the foot. I won't feel sorry for you at all when you have a pack load of problems. The opto
isolation is nice when dealing with high currents. I have separated the micro from the driver
by 50 feet with no problems. Don't worry about calculating voltage. Just set the A/D count
around 500 and adjust the pot for maximum PWM count. Somewhere I think I have a standalone
version of the software.
richhagen:
When I get an hour or two to look into it, I am going to look up the band gap parameters and change out the capacitor. If I have to keep the voltage divider output under 2.56 V I will have to change resistor values, but that is easy to do. I will probably have to adjust the program a bit to get reasonable resolution with the smaller voltage range, but it is definitely worth giving a try to see if it settles down my voltage measurements. I think I will play with a second Arduino and once I figure it out, make the changes to the load controller, I've just left it set up as it is, it switches on the lights and heater for now, as long as it's cold out I can get some benefit from it even now.
I have three C-40's here, so if I can reliably harvest power currently wasted, that would have value. I will have to review your design further as I am not sure I understand all of it at present, again when I can get some time to digest it I will. Rich
DamonHD:
Note: for the UNO at least the internal reference is 1.2V.
Rgds
Damon
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version