IV Curve with MPPT Charge Controller

[jflorey2]

This really does agree nicely with the data shown in post 75. The controller is in absorb, and ran a sweep showing that 318 W was the maximum power point, at 82.8 V and 3.84 A. The power conversion circuit drove the operating voltage up to 95.9 V, producing 1.91 A and 183 W. There is no way Input current = Output current in that condition.

Agreed. With MPPT you get a lot of benefit when bulk charging. You also get benefit when you switch to absorb, with the benefit decreasing as charge current decreases. Only once your battery charge current hits the Impp of the array do you lose the benefit of MPPT over PWM - because the battery cannot accept any more current than the panel can generate with either controller. (And it should be pointed out that these benefits are separate from the benefits you get from being able to run smaller gauge wire using a higher voltage array.)

[Sunking]

You are right, it isn't operating at Vmp, but the controller still ran enough of an IV curve sweep to show what Vmp would have been. The 95.9 V operating point is between Vmp and Voc, as we'd expect, and is efficiently converted to the constant 14.4 (or whatever) battery voltage that is being held for absorb. I don't understand why you are fighting this.

Not fighting it, stating facts. Do you know what chipset is in the Controller?

I do know and it is a Ti BQ series. like the BQ24650. Read the description and application notes. You will see this:

The bq24650 device is a highly integrated switchmode battery charge controller. It provides input voltage regulation, which reduces charge current when input voltage falls below a programmed level. When the input is powered by a solar panel, the input regulation loop lowers the charge current so that the solar panel can provide maximum power output. The bq24650 offers a constant-frequency synchronous PWM controller with high accuracy current and voltage regulation, charge preconditioning, charge termination, and charge status monitoring.

Read any of the chipsets, all of them switch to PWM mode when not in Buck Mode.

[Sunking]

The buck converter is still running, just at a lower duty cycle, sufficient to maintain the voltage at the output at a lower current. That's how they work. Power electronics 101.

A Buck Converter is PWM. Go read Power Electronics 101

[bcroe]

A Buck Converter is PWM. Go read Power Electronics 101

To clarify, yes a buck converter uses pulse width modulation to convert higher voltage at lower
current to lower voltage at higher current. But there is an inductor in there, allowing an efficiency
approaching 100%.

Here in the PV solar corner of the world, PWM refers to a switcher without the inductor, which
has the same input and output current; efficiency not so great. Badly named. Bruce Roe

[Mike90250]

........
........
Read any of the chipsets, all of them switch to PWM mode when not in Buck Mode.

Whoa buddy. I can speak to the high end MPPT controllers, they will MPPT at any stage of charge, in order to maintain the battery voltage. They will vary the MPPT power as needed, either for Thermal, Max Amps, or Battery Voltage control. They don't always run at "max smoke" MPPT.

When MPPT is no longer needed, and there are enough native panel amps to maintain Vbatt, that's when they drop down to simple $5 PWM, but when more amps are needed, they start up the MPPT again - even in Float or Absorb

[Sunking]

When MPPT is no longer needed, and there are enough native panel amps to maintain Vbatt, that's when they drop down to simple $5 PWM, but when more amps are needed, they start up the MPPT again - even in Float or Absorb

Mike that is what I am saying, perhaps poorly explained by myself, but that is what I saying. By your words "When MPPT is no longer needed", is the point where Panel Input Current = Output Current = PWM Mode.

There are only two true statement you can say about Output Current of a MPPT Controller and that is Output Current will be either Equal to Input Current, or Greater than Input Current.

• When Output Current is greater than Input Current, the controller is Buck Mode and using the Charge Pump of the Inductor/Capacitor at the output of a PWM regulator.
• When Output Current = Input Current, the converter is in PWM Mode, and there is no Buck from the Charge Pump. There is no place for it to go.

I agree with you if the load or battery demands more current than PWM mode can supply, will go back into Buck Mode assuming there is enough excess panel power available.

A Buck Converter is a PWM Regulator with an Inductor and Capacity on its output sometimes called a Charge Pump. So anytime the Controller is in Buck Mode is using the Charge Pump to buck voltage down, and boost current up. When the duty cycle falls to a certain point, and that point is when Output Current = Input Current when MPPT is no longer needed, The Charge pumps are not being used, thus is now pure PWM. Just the Nature of the Beast

We are on the same page.

[BackwoodsEE]

This morning brings an example of the charge controller finding a maximum power point well below Voc. It is dead-on, within a volt.

One panel has all three of its substrings shaded. You can see the slightly different levels of shading on each of those substrings by the slightly different current levels, stairstep fashion, between 68 V and 107 V. Bulk charging with about 80 W of inverter load.

The previous IV sweep (dashed pink line) shows that there was even more shading a little while earlier, as you would expect in the morning with trees nearby.

Capture.PNG

I went out and took a picture of the panels. (Maybe it's time to wash them off, but I've been busy building a real system this summer.)

IMG_2304.JPG

[J.P.M.]

This morning brings an example of the charge controller finding a maximum power point well below Voc. It is dead-on, within a volt.

One panel has all three of its substrings shaded. You can see the slightly different levels of shading on each of those substrings by the slightly different current levels, stairstep fashion, between 68 V and 107 V. Bulk charging with about 80 W of inverter load.

The previous IV sweep (dashed pink line) shows that there was even more shading a little while earlier, as you would expect in the morning with trees nearby.

Capture.PNG

I went out and took a picture of the panels. (Maybe it's time to wash them off, but I've been busy building a real system this summer.)]

After measuring clean/fouled array performance for going on 4 years now, FWIW, I've found the eye to be a poor indicator of panel fouling. Yours look about as dirty as mine, and at this time mine are fouled to about a 5% degradation below clean performance. Hose'm off and you'll restore ~ 2/3-3/4 of the lost performance. Hose'm, wipe'm with a soft cloth and hose'm again and you'll get them as clean as new. I don't worry about water spots after being unable to measure a difference in performance between water spotted and very careful removal of the water spots after clening/rinsing.

[max2k]

After measuring clean/fouled array performance for going on 4 years now, FWIW, I've found the eye to be a poor indicator of panel fouling. Yours look about as dirty as mine, and at this time mine are fouled to about a 5% degradation below clean performance. Hose'm off and you'll restore ~ 2/3-3/4 of the lost performance. Hose'm, wipe'm with a soft cloth and hose'm again and you'll get them as clean as new. I don't worry about water spots after being unable to measure a difference in performance between water spotted and very careful removal of the water spots after clening/rinsing.

do you realize you just killed young and fragile panel cleaning industry?

[J.P.M.]

do you realize you just killed young and fragile panel cleaning industry?

I appreciate the humor.

Long story, but the Reader's Digest version: As part of having too much time on my hands and more time than money, since retirement I've been able to pursue queries about many things that have been on my mind for a long time. One such matter is the nature and rates at which solar arrays foul (get dirty) and what rain or other cleaning methods may do to control the fouling. To that end, I take daily measurements, as clear skies permit, and I'm around at the time of daily minimum array incidence angle, quasi instantaneous and quasi steady state data collection to see what I can see. Sort of confirming prior similar activities, it looks like, for my array only, and subject to revision based on what I may find in the future:

1.) Without rain, things foul up at a rate something like ~ 0.6% to 0.8% per week linear performance degradation for ~ 8 weeks or so after a cleaning.
2.) Again, without rain, after ~ 7- 8 weeks or so, it appears that rate of fouling tends to become asymptotic at ~ 5-6 % total fouling, and increasing at a much slower rate beyond ~ 8 weeks or so.
3.) Morning dew which might leave my rain gage thinking we got ~ 0.01" precip. tends to cake up the dust on the array a bit and seems to increase the fouling rate for a few days. That's hard to quantify however.
4.) A decent rain, maybe 0.2"-.04" precip. at a "decent" rate, tends to restore ~ 2/3 to 3/4 of the fouling induced performance loss. So, if the array is, say, 5% fouled, a decent rain will reduce the fouling to ~ maybe 1% to 2% or so.
5.) Water spots from hard water do not seem to inhibit the performance of a clean array in any way I can notice or quantify from that of a cleaned and hand rubbed array.

I believe, based on what I've measured, that hitting an array with H2O from a hose at the rate of ~ 3/4 to 1 gal. per panel is about equivalent to the cleaning provided by a decent rain.

Based on the above. I believe paying to have an array cleaned is a waste of money. Either hit it with a hose ~ 1X/month if it doesn't rain or put 5% extra into the size and call it the price of laziness.

[max2k]

I appreciate the humor.

Long story, but the Reader's Digest version: As part of having too much time on my hands and more time than money, since retirement I've been able to pursue queries about many things that have been on my mind for a long time. One such matter is the nature and rates at which solar arrays foul (get dirty) and what rain or other cleaning methods may do to control the fouling. To that end, I take daily measurements, as clear skies permit, and I'm around at the time of daily minimum array incidence angle, quasi instantaneous and quasi steady state data collection to see what I can see. Sort of confirming prior similar activities, it looks like, for my array only, and subject to revision based on what I may find in the future:

1.) Without rain, things foul up at a rate something like ~ 0.6% to 0.8% per week linear performance degradation for ~ 8 weeks or so after a cleaning.
2.) Again, without rain, after ~ 7- 8 weeks or so, it appears that rate of fouling tends to become asymptotic at ~ 5-6 % total fouling, and increasing at a much slower rate beyond ~ 8 weeks or so.
3.) Morning dew which might leave my rain gage thinking we got ~ 0.01" precip. tends to cake up the dust on the array a bit and seems to increase the fouling rate for a few days. That's hard to quantify however.
4.) A decent rain, maybe 0.2"-.04" precip. at a "decent" rate, tends to restore ~ 2/3 to 3/4 of the fouling induced performance loss. So, if the array is, say, 5% fouled, a decent rain will reduce the fouling to ~ maybe 1% to 2% or so.
5.) Water spots from hard water do not seem to inhibit the performance of a clean array in any way I can notice or quantify from that of a cleaned and hand rubbed array.

I believe, based on what I've measured, that hitting an array with H2O from a hose at the rate of ~ 3/4 to 1 gal. per panel is about equivalent to the cleaning provided by a decent rain.

Based on the above. I believe paying to have an array cleaned is a waste of money. Either hit it with a hose ~ 1X/month if it doesn't rain or put 5% extra into the size and call it the price of laziness.

thanks for detailed info, I'm saving this for my own future reference- hard numbers are rare find even with disclaimers which IMO makes them 'harder'.

[J.P.M.]

thanks for detailed info, I'm saving this for my own future reference- hard numbers are rare find even with disclaimers which IMO makes them 'harder'.

Don't treat them hard. They're only for my system and the weather makes things notoriously variable. I'll get out of this thread now w/apologies to
OP and others. I'll report on results in another, new thread @ some point.