MPPT Con ?

[Paul Elliott]

One more general observation about MPPT, and why it's not snake oil: Typical solar panels are designed to charge a battery through a lossy converter (or through a direct connection). They have enough cells to provide sufficient charging power under a wide range of illumination, temperature, and battery chemistry. This means that in most cases, available energy is being wasted. MPPT controllers are able to convert this energy, rather than it being dissipated in the panel.

The MPPT controller needs to be well-designed and priced to be able to realize the power gain at a tolerable price. This is what we're trying to evaluate.

[cuthbert]

Originally Posted by noelex 77

That’s not my understanding. They offer the most advantage when there is a large difference between the solar panel voltage and the battery voltage.
Any shade such as a small shadow from a halyard will significantly reduce the solar panel voltage and reduce or eliminate the MPPT. The inherent losses of the voltage conversion in the MPPT regulators remain.
This makes them less sutable for a shade-challenged platform such as a sailboat.

Taking the second point first, not really, a MPPT is possibly advantageous in a partially shaded scenario. What the device is doing is changing its impedance so that the optimal Voltage x Current (= max power) is maintained most of the time as per the graphs on the first page of posts. The algorithm used varies between manufacturers but it is essentially "perturb and observe" the input impedance of the controller (hence the voltage) is increased until the current drops off the "knee" and then it is backed off a bit so that it is on the "knee" of the graph, how often it does this and how well determines how it can cope with transiting clouds, a partial shadow etc. A clear blue sky on a cold day and a low cost controller with an input impedance matched to a panels "knee" will be similar in performance.
As to the first point, yes, sort of, as others have posted earlier if that mismatch of voltages is due to the batteries being low and capable of taking the full unregulated output from the controller. When the batteries are near charged you lose the MPPT advantage as you don't need maximum power of the panels.
But if you mean that generally a big voltage between panels and batteries gives an advantage, not so. Most of these power regulators are buck but it could just as easily be a boost device or even a hybrid buck/boost. Regulators are typically efficient over a range and low voltage dropout should never be a factor.

It really comes down to how many panels you have room for and if you need to squeeze the most out of an existing setup and get an early morning charge (and then of course factoring the extra cost of an MPPT vs a Honda EU1000 or other alternate)

[mbianka]

Well the video with the MPPT vs standard PWM does show the advantage of the MPPT controller power wise under the conditions that it was operating. Personally, I still don't think that it is not worth the extra cost (3x at this time) over a standard solar controller for a boat. I'd also wonder about the MPPT controller that changes the voltage curent ratio to achieve this maximum power point. I wonder if the higher voltage of the MPPT controller will damage sensitive AGM or Gel batteries as opposed to flooded bank when it boosts up the voltage?

[cfarrar]

Capt Mike,

The MMPT regulator only maintains higher voltage on the upstream side, at the solar panel. On the downstream side, the voltage of current delivered to the battery follows a three-stage charging routine, adjustable to battery type, just like any other smart charger.

[mbianka]

Originally Posted by cfarrar

Capt Mike,

The MMPT regulator only maintains higher voltage on the upstream side, at the solar panel. On the downstream side, the voltage of current delivered to the battery follows a three-stage charging routine, adjustable to battery type, just like any other smart charger.

That's good to know thanks for that info.

[goboatingnow]

listen to cuthbert he knows what hes talking about.

couple of clangers in some other posts.

MPPT are PWM devices, only in sofar as they have buck regulators, ( in the same way that moden mains battery chargers are PWM devices). But MPPT only applies to the technique of tracking the power point of the solar panel, it has nothing to do with the charging part of the design, that could be 1 stage, 2 stage 3 stage or some combination. As other have said the only use of MPPT is when you have capacity to draw out the maximum power of the panel. ( at a given sunlight).

MPPT isnt snakeoil, its got nothing to do with charging effiiciency, its only refers to panel outout efficiency

Dave

[Paul Elliott]

Originally Posted by goboatingnow

MPPT are PWM devices, only insofar as they have buck regulators

Dave, to amplify on your excellent points, the majority of MPPTs are indeed buck regulators, but there are a few that are boost designs (buck steps the voltage down, boost steps it up). The designs are related and all use high frequency PWM in the step-up or step-down conversion. This is not the same thing as the PWM used by some of the non-MPPT controllers.

[hellosailor]

Dave-
"But MPPT only applies to the technique of tracking the power point of the solar panel, it has nothing to do with the charging part of the design, that could be "
I'd say you're only batting 500 there.<G>

MPPT does not track the "power point" of the solar panel. It monitors the output voltage of the solar panel, compares it to the present voltage of the battery, considers the capacity of the battery (normally inputted during setup) and then uses a microprocessor to determine the optimum charging voltage and current. As part of that optimization process, there is some program or array that the microprocessor uses to determine what is "right" and by putting different numbers into that logic array (or formula, whatever they are using) you can just as easily use one-step, three-step, or continuous-step charging algorithms.

Whether the circuit uses lookup tables, or does real-time calculation, or any other means to make the determination, it is still analyzing the battery parameters and then making "decisions" as they change, and the panel output changes. Way more complex than "tracking" the panel by itself, whatever that would mean.

[mbianka]

I'm pretty content with with my standard Morningstar PWM controllers and the job they do on board. It would take a long time to make up for the cost of replacing them with a MPPT controllers. But, then the price of this unit would be more like it:
MPPT 15A 12/24V SOLAR CHARGE CONTROLLER-30% POWER BOOST - eBay (item 260672306455 end time Dec-21-10 20:00:36 PST)
How it preforms and holds up may be another issue. But, if it works near what the $200+ units it might be worth a try. Though I don't see a setting for charging AGM or GELS so it may be useful for only those with flooded batteries.

[cal40john]

Originally Posted by hellosailor

Dave-
"But MPPT only applies to the technique of tracking the power point of the solar panel, it has nothing to do with the charging part of the design, that could be "
I'd say you're only batting 500 there.<G>

MPPT does not track the "power point" of the solar panel. It monitors the output voltage of the solar panel, compares it to the present voltage of the battery, considers the capacity of the battery (normally inputted during setup) and then uses a microprocessor to determine the optimum charging voltage and current. As part of that optimization process, there is some program or array that the microprocessor uses to determine what is "right" and by putting different numbers into that logic array (or formula, whatever they are using) you can just as easily use one-step, three-step, or continuous-step charging algorithms.

Whether the circuit uses lookup tables, or does real-time calculation, or any other means to make the determination, it is still analyzing the battery parameters and then making "decisions" as they change, and the panel output changes. Way more complex than "tracking" the panel by itself, whatever that would mean.

Maybe I'm reading this wrong but what I get out of this is that you're saying the MPPT controller is always trying to find the max output of the solar panel, then if it determines that the battery should be in a float stage it now does what with that excess energy?

That doesn't make sense. Whatever algorithm is used to charge the battery, only when the controller is in the mode of dumping as much energy into the batteries as possible does finding the max power transfer point make sense. As soon as the controller starts reducing current to the batteries it is throwing away energy from the solar panels. It is opening or shunting the panels, this is deliberately moving away from the max power transfer point.

John

[noelex 77]

Originally Posted by hellosailor

MPPT does not track the "power point" of the solar panel. It monitors the output voltage of the solar panel, compares it to the present voltage of the battery, considers the capacity of the battery (normally inputted during setup) and then uses a microprocessor to determine the optimum charging voltage and current. As part of that optimization process, there is some program or array that the microprocessor uses to determine what is "right" and by putting different numbers into that logic array (or formula, whatever they are using) you can just as easily use one-step, three-step, or continuous-step charging algorithms.

Whether the circuit uses lookup tables, or does real-time calculation, or any other means to make the determination, it is still analyzing the battery parameters and then making "decisions" as they change, and the panel output changes. Way more complex than "tracking" the panel by itself, whatever that would mean.

If we are not careful we are going to confuse a lot of people.
MPPT regulators do track the power point by continually ramping the input voltage up and down. Monitoring the change in current and trying to keep the solar panel voltage at the point where the maximum is power is delivered.
After this point its works like any other regulator charging the batteries, usually in a 3 stage process. (bulk absorption and float)
If the batteries are becoming charged and the regulator has switched to absorption or float the solar panels are producing more power than the batteries can take. If this point is reached then the extra efficiency gained by the MPPT is not needed. Most units continue to MPPT the panels and simply waste the excess power with PWM just like a convention regulator.

[noelex 77]

Originally Posted by mbianka

I'm pretty content with with my standard Morningstar PWM controllers and the job they do on board. It would take a long time to make up for the cost of replacing them with a MPPT controllers. But, then the price of this unit would be more like it:
MPPT 15A 12/24V SOLAR CHARGE CONTROLLER-30% POWER BOOST - eBay (item 260672306455 end time Dec-21-10 20:00:36 PST)
How it preforms and holds up may be another issue. But, if it works near what the $200+ units it might be worth a try. Though I don't see a setting for charging AGM or GELS so it may be useful for only those with flooded batteries.

Be careful there are regulators advertising MPPT when they do not do MPPT at all.
Even if it is a MPPT regulator poor MPPT tracking and power conversion has the potential to deliver less power than a simple regulator
If you want a MPPT regulator get a good one.

[hellosailor]

Mike, that ebay controller is absolutely unique. It has
Thunder protection
listed among the feautres. I guess it is the only MPPT controller on the market that doesn't run and hide under the settee when the thunder starts.

Hopefully they paid a little more attention to the electrical part of their design.<G>

[hellosailor]

John-
"It is opening or shunting the panels," Nope, not at all. It is not doing anything to the panels. All it does wrt the panels, is read the voltage coming from the panels.

When that eBay product says "This 15A MPPT type solar charge controller intelligently regulates the working voltage of solar panels,"
that's a flat-out lie. Their YnGlitch is as good as my Cantonese. The "working voltage" of a solar panel is determined by chemistry and physics and you can only regulate the voltage once it gets out of the panels and into a circuit. OK, to be picky, you can always make shadows on the panel or reduce the amount of light that hits it, in order to reduce the voltage coming out of the panel. That's still not "voltage regulation" in the sense that charging systems do it. We don't regulate the panel voltage, we regulate the charging system voltage. A difference easily lost in translation.

And as noelexe said there's going to be confusion happening here, part of which is probably because the panel makers would rather have everyone confused and unable to make comparisons.

The "Maximum Power Point" in MPPT controllers does not refer to a mysterious power point in the solar panels, but as far as I know, to the MPP required to optimally charge the batteries. Which is confusing because "power" means wattage, not voltage or amperage. And for an optimum battery charge, you need to monitor both voltage and amperage. What an MPPT controller does, is take all the power (wattage) from a solar panel and convert it into the proper combination of volts and amps that a battery needs. This is as opposed to a conventional solar controller, which is either a "dump regulator" or a conventional "down only" regulator that simply dumps (totally wastes) excess voltage.

The MPPT controllers don't "dump" any power from the panels. They don't throttle the panels, they don't adjust the panels, they don't turn them on and off. They just soak up all the power (wattage) and then transform it, literally with a transformer circuit or a DC-to-DC equivalent of that (sometimes both) using capacitor charging and pulsing, hence the "pulse width modulation" part of their output. (Once the batteries are fully charged, they will "ignore" the panels, disconnecting from them or doing something else with the solar power. But that's after the charging is done.)

Transformers, or capacitor charging, can be used to transform voltage into amperage, or vice versa. So instead of dumping excess power, the MPPT controller now has to determine what the power needs of the battery are, and how to adjust the power to best feed them. "Adjust" in this case meaning optimize the voltage and amperage as separate factors. Batteries can pretty much suck in all the amperage or voltage you can supply to them--but if you feed it to them at the wrong levels, they'll go BOOM. For each specific voltage level, there is a limit to the amperage you can feed them, and vice versa. You could go nuts trying to adjust voltage and amperage and compensate the amounts of both every time one or the other changed by 1%, even without trying to remember those numbers also change every time the temperature changes.
And before microprocessors, using just analog circuits, it would have been a nightmare trying to constantly compensate both volts and amps. The charging curves for the batteries are no secret, the problem is that the charger needs to keep checking the battery condition, keep fetching the right information, keep resetting the charging parameters (voltage and amperage based on battery charge state, temperature, etc.) and do it all reliably, to prevent that BOOM.
The two-stage and three-stage chargers are all "simplified" way to do that. They take some broad assumptions about "this can't make it go BOOM" and they stop there. The microprocessor controllers (i.e. MPPT) literally are looking up the right parameters, maybe one per minute, maybe once per second, and they are continuously giving the battery the maximum charging AMPERAGE that is available, at the optimum charging VOLTAGE. A higher voltage just boils the electrolyte and is wasted. Or makes the battery go BOOM.

If you watch the output display on an MPPT controller that has one, you'll see the output voltage keeps creeping up, usually by 0.1 volt increments, as the battery charges. The amperage will also be going up, initially, as the panel output goes up as the sun rises toward noon. Once the batteries are charged, the amperage will creep down again, to a trickle, unless something is drawing power.

But you don't see the "14.4 bulk, Snap!, 13.8 float" etc. changes in the charging voltage that a 3-stage makes. MPPT is a continuous stepping up/down, as it reads actual battery condition (and temperature, if equipped) and makes fine adjustments to match. It is one of those tasks that microprocessors really do very very well, constantly looking at things, comparing them, adjusting them. Exactly how each one does it, and which one is better (versus overkill or puffery) is harder to tell, but they all follow the same principals. And if they weren't all practically hand made (small market volume) they would be pretty cheap, too.

Like the thunder-protected one on eBay.<G>

[cuthbert]

Originally Posted by mbianka

MPPT 15A 12/24V SOLAR CHARGE CONTROLLER-30% POWER BOOST - eBay (item 260672306455 end time Dec-21-10 20:00:36 PST)
How it preforms and holds up may be another issue. But, if it works near what the $200+ units it might be worth a try. Though I don't see a setting for charging AGM or GELS so it may be useful for only those with flooded batteries.

It would be interesting to ask the vendor to remove the cover and state which IC is being used as the design basis. It could be an analog solution such as the Texas Instruments bq2031 (app note SLVA378–December 2009)

I have done a fair bit of work with the Cypress PSOC chipsets and they have an app note on a PSOC microprocessor based solution that is worth a look as a MPPT primer (Application note AN54121).

Either way I would still hold on for something that has a known big IC manufacturer such as NXP and their 32 bit integrated solution at $5.70. It can only be a matter of months for some more low cost MPPT controllers.