IV Curve with MPPT Charge Controller

[BackwoodsEE]

Thought you'd enjoy seeing this plot produced just now from the PV monitoring device I designed and built. The sun is rising over some trees near my 3-panel experimental setup, and the Morningstar TS-MPPT-60 is doing a fine job of maximizing the power draw from the array while letting the bypass diodes do their work in the shaded panel strings.

Capture.PNG

[sensij]

Cool! What is the relationship between the top and bottom charts (different scales on x axis?)?

[BackwoodsEE]

The upper plot shows voltage and current vs time, with the vertical dashed line showing when a change in voltage triggered the device to acquire a new set of samples. The lower plot shows current (and power) vs voltage.

[bcroe]

Thought you'd enjoy seeing this plot produced just now from the PV monitoring device I designed and built. The sun is rising over some trees near my 3-panel experimental setup, and the Morningstar TS-MPPT-60 is doing a fine job of maximizing the power draw from the array while letting the bypass diodes do their work in the shaded panel strings.

Thats cool. They really work. Bruce Roe

[Sunking]

What are you using for a LOAD? FWIW it cannot be fixed. It must be dynamic.

[jflorey2]

What are the X scales for both graphs? Seconds? Hours?

[BackwoodsEE]

What are you using for a LOAD? FWIW it cannot be fixed. It must be dynamic.

Nothing. My box just sits between panels and charge controller and waits for the CC to do an IV sweep. When it detects a voltage change beyond a given threshold, it stores samples into a buffer and then sends them to the PC for display.

I have another box that actively does an IV sweep, but it's not hooked up right now. It switches in some big electrolytic capacitors with a MOSFET, sending the IV samples to the PC, and then discharges the caps with another MOSFET while normal current flow to the CC is resumed.

[BackwoodsEE]

What are the X scales for both graphs? Seconds? Hours?

X-axis for the upper plot is in seconds, and for the bottom plot is in PV volts.

[Sunking]

Nothing. My box just sits between panels and charge controller and waits for the CC to do an IV sweep. When it detects a voltage change beyond a given threshold, it stores samples into a buffer and then sends them to the PC for display.

Huh? Then your plots have no meaning. There is no power. What am I missing?

[BackwoodsEE]

Huh? Then your plots have no meaning. There is no power. What am I missing?

When the charge controller obtains an update of the current vs. voltage curve of the PV array, it adjusts its buck converter many times per second for a second or two (sometimes less) to present a rapidly changing load to the PV array. My device detects when it starts that process by monitoring the PV voltage. When a sufficiently large voltage deviation is detected, it triggers acquisition of 4096 voltage/current samples and then sends them to the PC. What you see in the bottom plot is the measured current values during that acquisition plotted against the measured voltage values, for each of the samples. I also calculate power by simply multiplying voltage times current.

The upper plot just shows the sample acquisition as it happens vs. time. It's useful for seeing what kind of MPPT sweep the charge controller did. Sometimes it just does a small sweep to optimize its output, and other times it does a full sweep from battery voltage to Voc.

Here is a pic of the device showing the analog and digital boards and 20A shunt. All analog circuitry is isolated with a floating flyback-type DC-DC converter power supply, separate ground plane, and optoisolators for the digital communication to the A/D converter.

PV-monitor.jpg

[BackwoodsEE]

Here's the latest plot. The TS-MPPT-60 is bulk charging the battery in full sun with some modest loads. The panels are not quite optimally oriented. Since there is no shading going on, the charge controller is happy to just do a small sweep to ensure it's operating at the MPPT. The dashed line in the bottom plot shows the previous IV curve, which is so close to the current IV curve that it covers up the current IV curve's green plot line. You can see the charge controller brought the PV voltage all the way down to (charging) battery voltage during that previous sweep, and that Voc for the string of 3 panels is around 104V.

Capture.PNG

[Sunking]

Still confused, What is the LOAD? Power has to have some place to go. If not then there is no power.

[jflorey2]

Still confused, What is the LOAD? Power has to have some place to go. If not then there is no power.

From his post - "The TS-MPPT-60 is bulk charging the battery in full sun." That's the load.

[Sunking]

From his post - "The TS-MPPT-60 is bulk charging the battery in full sun." That's the load.

That is not a load you can test with. You should know that.

[jflorey2]

That is not a load you can test with. You should know that.

Agreed. You wanted to know what the load was; it's a battery. It's not a great load for testing, but it does provide a load, which is why power is a valid axis in his plots.