No Grid Power = No Solar Power

[RegGuheert]

They can regulate all the way to zero current.

Bluntly to the point - I think we all agree that an MPPT charge controller can be designed to slide around the E/I curve - but in the real world they don't unless they searching for the max power point.

Yes, they do. That is how they regulate the voltage of the battery.

 

[QueenBee]

RegGuheert

We seem to be talking past each other. Please assume that I know how MPPT works and what is theoretically possible.

What I'm looking for is what's really available on the market, with documentation.
=Smidge=

I tried seeing how the Sunny Island and Sunny boy do it and all I could find is that the the Sunny Island changes the frequency but I couldn't find how the Sunny Boy responds to this.

 

[RegGuheert]

What I'm looking for is what's really available on the market, with documentation.

And here is the description of operation from page 31 of the Midnite Solar Classic manual:

Battery Charge Stages and Meanings
Bulk MPPT
This stage of the Classic means; that the Classic will be putting out as much current as it can to raise the battery voltage to the absorb voltage set point. This is also known as constant current mode.
Absorb
This stage means that the Classic will maintain the absorb set point voltage until the batteries are charged. This stage is terminated at the end of the Absorb time or the End Amps set point whichever is reached first. At this stage the classic is not putting out maximum current, as that would increase the battery voltage over the Absorb set point. This is also referred to as constant voltage mode.
The absorb time is proportional to the bulk time. (i.e. the time bulk takes to reach the absorb voltage.) The battery it’s considered “full” at the end of the absorb charge cycle.
Float
A Float cycle follows after the Absorb cycle is completed; Float is displayed on the screen. Battery voltage is held at the float voltage set point.

Note that MPPT can ONLY be used in Bulk charging mode, since in all other modes, you cannot use the full output of the PV array and must regulate at a lower power level.

 

[Smidge204]

Note that MPPT can ONLY be used in Bulk charging mode, since in all other modes, you cannot use the full output of the PV array and must regulate at a lower power level.

So to answer my question "Is MPPT be used to throttle the power output of the PV system?" is, apparently, "No."

At least as far as the Midnite Solar Classic is concerned.

I tried seeing how the Sunny Island and Sunny boy do it and all I could find is that the the Sunny Island changes the frequency but I couldn't find how the Sunny Boy responds to this.

I'd expect it to respond by shutting itself off, like any well behaved grid-interactive inverter would when it sees out-of-range power quality.
=Smidge=

 

[RegGuheert]

So to answer my question "Is MPPT be used to throttle the power output of the PV system?" is, apparently, "No."

But you never asked this question. What you stated was this:

What I'm not aware of is MPPT inverters that deliberately deoptimize panel production in order to deliberately decrease output to match load.

Now you are aware of an MPPT charge controller that does precisely this. ALL MPPT charge controllers do this.

At least as far as the Midnite Solar Classic is concerned.

The Midnite Solar Classic controller operates EXACTLY as I have described. In MPPT mode, the controller moves the operating point around until it finds the maximum power point. Once the battery gets full enough, it deliberately decreases the output to match the load.

I'm sorry this is news to you guys. Shipping MPPT charge controllers have been operating this way in customer installations for about fifteen years now.

 

[DaveEV]

Google for "sunny boy frequency shift power control" to see how SMA does it.

 

[Smidge204]

So to answer my question "Is MPPT be used to throttle the power output of the PV system?" is, apparently, "No."

But you never asked this question. What you stated was this:

What I'm not aware of is MPPT inverters that deliberately deoptimize panel production in order to deliberately decrease output to match load.

Now you are aware of an MPPT charge controller that does precisely this.

No, no they don't. They always optimize panel output, not deoptimize. They convert voltage to current and current to voltage as necessary, keeping power as high as possible. Then, when there is nowhere for the power to go, they switch out of MPPT mode and do something else entirely.

Google for "sunny boy frequency shift power control" to see how SMA does it.

Found this. Page 8 (last page) in very useful. It seems the SMA systems do indeed have a mechanism to control inverter output via frequency. Interesting and informative, though it's not clear if MPPT algorithms are employed to reduce the power or if some other mechanism is employed.
=Smidge=

 

[RegGuheert]

No, no they don't. They always optimize panel output, not deoptimize. They convert voltage to current and current to voltage as necessary, keeping power as high as possible. Then, when there is nowhere for the power to go, they switch out of MPPT mode and do something else entirely.

Not so. The power stage does precisely the same thing it does in Bulk mode charging, it just moves to a different operating point on the curve, which deoptimizes the production of the panel. There is no such "something else entirely" to be done.

It's a very simple concept which I have explained in detail multiple ways.

 

[hgoudey]

It has been correctly pointed out that maximum power tracking mode is not compatible with load matching, by virtue of having to track the load it can not try to maximize available solar output at the same time. The float mode of a DC solar charge controller is closer to what Smidge has been asking for. When a battery reaches float voltage the charge controller will regulate solar output down to the a trickle that holds the float voltage, but if a substantial new DC load is applied the output will rise to match it. It may be possible to operate in this mode without a battery, (using a large capacitor like the ones they drive big car stereos with?). I don't see how this is very beneficial, though. It would be more robust to involve a small battery.

The major inconvenience of switching between a grid-tied PV configuration and a DC charge controller is that the string voltages are almost always quite different requiring different series/parallel wiring, generally not something you can swiftly change in a backup scenario. Even if you work that out, you end up with a DC power source and have to add an inverter (not you grid-tied PV inverter) to get AC, assuming that is what you want.

If you don't want a generator, want to stick with a grid-tied only solar inverter most of the time and don't mind spending $2000-4000+, you can install a battery based inverter that charges/floats off the AC and automatically transfers to powering a sub panel off the batteries when the grid goes down (without connecting any solar to this DC bus through a DC charge controller). If you use the Sunny Island product a typical grid tied Sunny Boy inverter will get load regulation commands to add its solar power to the battery backed circuits in a stable, on-demand fashion. Or, if you use something like the Outback/Radian system you can use it to spoof a local grid and turn on an agreeable grid-tie inverter, but you will need a dump load that can handle excess solar above demand levels. This can be on the DC battery bus side of the battery inverter (Outback inverters are bidirectional and will push DC onto the battery in a fairly uncontrolled charging process in an attempt to keep the AC side stable). The various good old DC solar charge controllers discussed earlier often have a mode to act as a excess power dump (usually triggered by a voltage rise on the battery bus). You can also have an variable AC load (the aforementioned water heater dump), but you need a mechanism to monitor and regulate this load based on the balance of AC generation and other AC loads. I am not aware of off the shelf solutions for that, but there could be something out there.

Ultimately a cost effective EV-based back-up power solution including solar is a matter of standards and integration. All the hardware is pretty much there for anyone who owns and EV and a grid-tied inverter, it just needs to be designed to play together. Using the vehicle inverter/charger in a bi-directional fashion to maintain a household grid island behind a transfer switch and coordinate output with a solar inverter, matching generation with loads would be a nice solution that shouldn't really require much more hardware, just some thoughtful integration.

Howdy

 

[Smidge204]

No, no they don't. They always optimize panel output, not deoptimize. They convert voltage to current and current to voltage as necessary, keeping power as high as possible. Then, when there is nowhere for the power to go, they switch out of MPPT mode and do something else entirely.

Not so. The power stage does precisely the same thing it does in Bulk mode charging, it just moves to a different operating point on the curve, which deoptimizes the production of the panel. There is no such "something else entirely" to be done.

It's a very simple concept which I have explained in detail multiple ways.

So when you said "Note that MPPT can ONLY be used in Bulk charging mode" what did you really mean?

We're talking *specifically* about MPPT algorithms. There is no contention whatsoever that inverters/controllers can control their output, we are talking about specifically using MPPT and only MPPT control to accomplish this. So far, everything you have said and shown indicates that MPPT is only used to maximize power output, not reduce it. You've demonstrated that MPPT is used to reduce voltage at te top of battery charging cycles, but at the same time current is increased - that is not reducing power. If power is reduced it's a happy accident, because MPPT is trying to keep power as high as possible. I'm looking for an example where MPPT control is deliberately used to operate a panel outside it's optimal parameters to decrease their power output.

It should be possible, but the question is if it's done or not. :roll: There may be technical reasons why it's not, which could be fun to explore.

The float mode of a DC solar charge controller is closer to what Smidge has been asking for.

Not *exactly* what I'm asking for... there are absolutely ways to limit the power from a PV system, I'm looking for examples of MPPT control methods being used for that purpose. I agree that MPPT is apparently not compatible with load tracking since the "M" stands for "Maximize" - and you wouldn't want maximum power when trying to load match! What you need is LPPT (Load power point tracking) or OPPT (Optimal power point tracking)! :lol:

Thank you for your other inputs as well.
=Smidge=

 

[QueenBee]

I tried seeing how the Sunny Island and Sunny boy do it and all I could find is that the the Sunny Island changes the frequency but I couldn't find how the Sunny Boy responds to this.

I'd expect it to respond by shutting itself off, like any well behaved grid-interactive inverter would when it sees out-of-range power quality.
=Smidge=

That seems inefficient though. That would mean it have to have a cycle like this. Full power from PV charging batteries and running household load -> batteries full kill PV and run house from batteries -> batteries need a charge turn back on PV, repeat until the sun goes down.

 

[AndyH]

They can regulate all the way to zero current.

Bluntly to the point - I think we all agree that an MPPT charge controller can be designed to slide around the E/I curve - but in the real world they don't unless they searching for the max power point.

Yes, they do. That is how they regulate the voltage of the battery.

I don't think so, Reg, because the MPPT portion of the puzzle is on the input side of a DC-DC converter, and the battery is on the output side. The maximum power point of a panel or array is constantly changing, as is the voltage of the battery. A charge controller must independently adjust the input and output simultaneously. A battery's voltage changes as it charges and discharges, and as loads are applied and lifted - the MPPT charge controller doesn't stop scanning for the MPPT point when the charge process moves from constant current to constant voltage, or from CV to float. In a battery-backed system, the charge controller doesn't have to react to inverter loads anyway - it's job is only to feed the battery. The battery reacts to the inverter.

From an earlier link:

Since the goal of the PPT to generate the maximum power from the solar panels, operating the solar panels at roughly this voltage is optimal. However, when a solar panel is used to charge a 12v battery directly, the battery pulls the operating voltage of the panel down to its own voltage of 12v.

The basis for a Peak Power Tracker is that the DC/DC converter changes the higher voltage/ lower current solar panel input to the lower voltage/ higher current battery charging power. The microprocessor controls the conversion ratio of the DC/DC converter, keeping the solar panel operating at its MPP. There are obviously a lot more details that go into this design. For a clearer understanding of the process, look at the schematic and software listings for the PPT on my website http://www.timnolan.com" onclick="window.open(this.href);return false;.

 

[RegGuheert]

You've demonstrated that MPPT is used to reduce voltage at te top of battery charging cycles, but at the same time current is increased - that is not reducing power.

You still misunderstand how it works. When the battery gets fully charged, the voltage across the PV array is *increased* so that the current from the array decreases. Since you were at the maximum power point during bulk charging, any increase in voltage reduces the power from the array. By moving off the MPPT point, you are deoptimizing the production of the panel. The panel will now heat up more than if you had taken maximum power from it. As you can easily see from the curves, just moving the voltage up 10% or so will allow you to reduce the current (and therefore the power) all the way to zero. With this small range of voltage movement you can regulate at any power level from zero to the full power of the array.

Let's put it yet another way: Assuming you have an array that is 20% efficient, you will achieve this only at the maximum power point. As you increase the voltage gradually toward the open circuit voltage, you gradually decrease the efficiency of the panel toward zero. The array can be operated at any efficiency between 0% and 20% to meet the needs of the load. That is precisely how a MPPT controller operates as it moves from bulk charging to absorption.

 

[AndyH]

You've demonstrated that MPPT is used to reduce voltage at te top of battery charging cycles, but at the same time current is increased - that is not reducing power.

You still misunderstand how it works. When the battery gets fully charged, the voltage across the PV array is *increased* so that the current from the array decreases. Since you were at the maximum power point during bulk charging, any increase in voltage reduces the power from the array. By moving off the MPPT point, you are deoptimizing the production of the panel. The panel will now heat up more than if you had taken maximum power from it. As you can easily see from the curves, just moving the voltage up 10% or so will allow you to reduce the current (and therefore the power) all the way to zero. With this small range of voltage movement you can regulate at any power level from zero to the full power of the array.

Let's put it yet another way: Assuming you have an array that is 20% efficient, you will achieve this only at the maximum power point. As you increase the voltage gradually toward the open circuit voltage, you gradually decrease the efficiency of the panel toward zero. The array can be operated at any efficiency between 0% and 20% to meet the needs of the load. That is precisely how a MPPT controller operates as it moves from bulk charging to absorption.

Nope. Smidge and engineers that design these things get my vote.

The microprocessor controls the conversion ratio of the DC/DC converter, keeping the solar panel operating at its MPP.

The maximum power point tracker is one 'black box', the charge controller is another. Each does their jobs independently.

 

[RegGuheert]

A battery's voltage changes as it charges and discharges, and as loads are applied and lifted - the MPPT charge controller doesn't stop scanning for the MPPT point when the charge process moves from constant current to constant voltage, or from CV to float.

This is where you are incorrect. It DOES stop scanning for the maximum power point as it moves from bulk to constant voltage mode and instead it regulates to provide the exact current to the battery which maintains the specified battery voltage. If the load increases or decreases, it will adjust the operating point on the array to give the exact power that maintains the battery voltage at the CV set point. If a huge load drops on the battery and the controller is no longer able to hold the battery in constant voltage mode, it will drop back to bulk and go back to scanning for the maximum power point.

You don't have to believe me since it's printed in the Classic manual, as I quoted. If you don't believe that, then simply ask Midnite. You are their customer, they will gladly answer your questions on the forum. If you don't believe them, you can test it yourself since you have the Classic controller. It reports the exact array voltage and current at each moment. If you test around noontime on a clear day, you can even trace out part of the I-V characteristic of your PV array by changing the load while in constant voltage mode and observing what that does to the PV voltage and current.

 

[RegGuheert]

The maximum power point tracker is one 'black box', the charge controller is another. Each does their jobs independently.

Wrong again. There is a single DC/DC converter. In bulk charging mode, it hunts for the peak power point and it puts every bit of available PV power into the battery. As it transitions into constant voltage mode, the same DC/DC converter's duty cycle gradually adjusts so that the battery voltage stops rising. As the battery gets more filled, the microprocessor adjusts the duty cycle to lower the power from the array and into the battery to keep the voltage constant. One power stage. One microprocessor. The only difference is where on the PV curve the controller operates.

 

[AndyH]

The maximum power point tracker is one 'black box', the charge controller is another. Each does their jobs independently.

Wrong again. There is a single DC/DC converter. In bulk charging mode, it hunts for the peak power point and it puts every bit of available PV power into the battery. As it transitions into constant voltage mode, the same DC/DC converter's duty cycle gradually adjusts so that the battery voltage stops rising. As the battery gets more filled, the microprocessor adjusts the duty cycle to lower the power from the array and into the battery to keep the voltage constant. One power stage. One microprocessor. The only difference is where on the PV curve the controller operates.

Nope. Look at the information I've already provided. I've familiar with the way two devices work and they're very similar. There is an MPPT block that keeps the panel loaded at it's max power point, there's a battery charger that provides the three stages (CC, CV, and float) for lead acid battery charging, and there's a DC-DC in between that bridges the gap between the MPPT and charger.

Bulk (constant current, CC), absorb (constant voltage, CV), float, and equalize are battery charging functions. These have zero to do with PV or MPPT. The battery charger is the consumer of the energy harvested by the panels and MPPT controller. The text in the Midnite manual is about how energy is shuttled to the battery and does not describe how the DC-DC or MPPT function might be manipulated to provide power to the charger.

We DO have hardware and software info for these two projects as I've already linked. There's a third done by the engineers behind the Circuit Cellar which also breaks out these functions in their solar harvesting series, but since I no longer have the magazines, I cannot 'prove' that. The documents I have for the two different charge controllers I use do not provide 'behind the scenes' information - only 'user' and 'installer' details - and are therefore not useful in this conversation.

If you think this is incorrect, please link to the Midnight document or a document from another charge controller and provide a page reference. Thanks!

Here's the link to the Midnite Classic charge controller manual:
http://www.midnitesolar.com/pdfs/classicManual.pdf

Here's the code for Tim Nolan's MPPT charge controller:
http://www.timnolan.com/uploads/Arduino Solar/ppt.pde

 

[TonyWilliams]

I'm almost afraid to jump in with my specific request, but...

My current 6.7kW AC system with 35 SunPower 228w / 35 Enphase M190 needs to be off grid capable. I specifically did not read the links earlier on how to do that "cheap" yet.

My thoughts are that I could add 2kW - 3kW to my current system, and now I'm wondering if it would be better to use a DC system with the Midnight Solar equipment. That would take a smallish battery that will get the whole system operating during the day that I could pull 8kW -10kW at peak solar output (and the batteries would make up the difference, if any) and at night, I would struggle with whatever the batteries can give me.

Your thoughts on my hybrid idea.

 

[QueenBee]

I'm almost afraid to jump in with my specific request, but...

My current 6.7kW AC system with 35 SunPower 228w / 35 Enphase M190 needs to be off grid capable. I specifically did not read the links earlier on how to do that "cheap" yet.

I don't think there is any official Enphase support for using them off grid but the Sunny Island would probably be a good fit for this scenario? Would the additional panels provide enough power to meet demand and charge the batteries? Seems like if they would your idea would probably be a good cost effective solution.

 

[Smidge204]

I'm almost afraid to jump in with my specific request, but...

My current 6.7kW AC system with 35 SunPower 228w / 35 Enphase M190 needs to be off grid capable. I specifically did not read the links earlier on how to do that "cheap" yet.

The Enphase inverters want to see "Grid Quality" power before they'll activate. What you need is an inverter that can take battery power and produce AC power good enough to trick the Enphase units into turning on. Wholesale Solar sells systems that get installed between your incoming utility feed and your home/sub-panel that do exactly this. It uses the "on/off" method of controlling any grid-interactive solar inverters on the load side, via frequency shifting.
=Smidge=