Wall-Wart meets DIY MPPT

Xringer · 09-01-11, 08:19 PM

Here's the Ebay Wall-Wart DC 12V 1A 1000mA Power Supply Adapter for CCTV Camera | eBay
Input: :100-240V AC 50/60 Hz
Output: DC 12V 1000mA

I've not tested it yet, but I think it will run on 100-200vdc..

The PV input is four 200w panels in series, with a MPP voltage of 115.6 VDC.
The output load is a 10 ohm resistance.

When properly adjusted (and maybe tweaked), the comparator will turn
on the FET, when the input PV voltage is above 115.6 Volts.
The power stored in C1 will be discharged into the 10 ohm load.

115.6v / 10 ohms = 11.56 Amp (11.56A * 115.6V = 1.336 KW).
Of course, the short pulse when C1 discharges won't really produce 1.336 KW.
(1 watt is 1 amp at 1 volt, for 1 second).
But, when the voltage climbs back up (if the sun is still bright), I expect the
FET will be switching off and on thousands of times a second..
(Depending on the size of C1).

On a good day, I'm hoping the total wattage delivered to the 10 Ohm load, will be close to 800w..

This project is an attempt to get the maximum power out of my 800w array.
Currently, it's delivering only about 500w (to heat my boiler).
The low power is due to the low resistance of the load (10 Ohms).

The correct load would be about 16.7 Ohms..
115.6v / 16.7 Ohms = 6.922 amps.
And 115.6v * 6.922a = 800w.

By using a lower resistance, (10 Ohms) I can still get pretty good power,
even on partly cloudy days. With the FET running at a lower frequency.
(C1 would charge up a little slower).

This hack is not really a power point tracker, since it will be
pre-adjusted to the PV array's max power point voltage (115.6v), and will
only take power from the panels when that voltage is reached.

When it starts getting dark and the PV output voltage is below 115.6v,
the FET will stay off.

My plan is to connect a 120vac 1-watt lamp across the 10 ohm load.
This will be a visual indication of how well the gizmo is working.
I should be able to tweak R3 for max lamp brightness, for best results.

I do not think my clamp-on current meter is going to work very well
with this rig.

I think the lamp will be better to use than leaving my Oscilloscope
connected to the output..

Your comments are welcome.

Cheers,
Rich

pinhead · 09-02-11, 08:12 AM

Well all looks good, and you are right that it will likely work with DC in place of AC as the front end of that device is a bridge rectifier. But then I noticed that the input and output grounds of your wall wort are connected. I think this may cause this kind of circuit some trouble.

Xringer · 09-02-11, 10:46 AM

I'm pretty sure those grounds have to be connected..

There is a 12Vdc 'connection' back to the input, but it's possible that
it's on the + side, and it won't make any difference.

It's also possible mine is using an Opto-isolator to couple the PWM back..
That would really be the best way to design these things..
You do not want one side of the AC line to be felt out on the DC end
of your cell phone charger..

One thing I've been doing with these Chinese Switcher wall warts,
is checking them for leakage with my Ohm meter. They appear to be Open..

09-01-11, 08:19 PM	#1
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	Wall-Wart meets DIY MPPT Here's the Ebay Wall-Wart DC 12V 1A 1000mA Power Supply Adapter for CCTV Camera \| eBay Input: :100-240V AC 50/60 Hz Output: DC 12V 1000mA I've not tested it yet, but I think it will run on 100-200vdc.. The PV input is four 200w panels in series, with a MPP voltage of 115.6 VDC. The output load is a 10 ohm resistance. When properly adjusted (and maybe tweaked), the comparator will turn on the FET, when the input PV voltage is above 115.6 Volts. The power stored in C1 will be discharged into the 10 ohm load. 115.6v / 10 ohms = 11.56 Amp (11.56A * 115.6V = 1.336 KW). Of course, the short pulse when C1 discharges won't really produce 1.336 KW. (1 watt is 1 amp at 1 volt, for 1 second). But, when the voltage climbs back up (if the sun is still bright), I expect the FET will be switching off and on thousands of times a second.. (Depending on the size of C1). On a good day, I'm hoping the total wattage delivered to the 10 Ohm load, will be close to 800w.. This project is an attempt to get the maximum power out of my 800w array. Currently, it's delivering only about 500w (to heat my boiler). The low power is due to the low resistance of the load (10 Ohms). The correct load would be about 16.7 Ohms.. 115.6v / 16.7 Ohms = 6.922 amps. And 115.6v * 6.922a = 800w. By using a lower resistance, (10 Ohms) I can still get pretty good power, even on partly cloudy days. With the FET running at a lower frequency. (C1 would charge up a little slower). This hack is not really a power point tracker, since it will be pre-adjusted to the PV array's max power point voltage (115.6v), and will only take power from the panels when that voltage is reached. When it starts getting dark and the PV output voltage is below 115.6v, the FET will stay off. My plan is to connect a 120vac 1-watt lamp across the 10 ohm load. This will be a visual indication of how well the gizmo is working. I should be able to tweak R3 for max lamp brightness, for best results. I do not think my clamp-on current meter is going to work very well with this rig. I think the lamp will be better to use than leaving my Oscilloscope connected to the output.. Your comments are welcome. Cheers, Rich

09-02-11, 08:12 AM	#2
pinhead Helper EcoRenovator Join Date: Jul 2011 Location: Ontario, Canada! Posts: 41 Thanks: 0 Thanked 5 Times in 5 Posts	Well all looks good, and you are right that it will likely work with DC in place of AC as the front end of that device is a bridge rectifier. But then I noticed that the input and output grounds of your wall wort are connected. I think this may cause this kind of circuit some trouble.

09-02-11, 10:46 AM	#3
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	I'm pretty sure those grounds have to be connected.. There is a 12Vdc 'connection' back to the input, but it's possible that it's on the + side, and it won't make any difference. It's also possible mine is using an Opto-isolator to couple the PWM back.. That would really be the best way to design these things.. You do not want one side of the AC line to be felt out on the DC end of your cell phone charger.. One thing I've been doing with these Chinese Switcher wall warts, is checking them for leakage with my Ohm meter. They appear to be Open..