MPPT Con ?

[noelex 77]

I am considering upgrading my conventional old solar regulator for a MPPT one. I have 330w of solar.
Manufacturers of MPPT regulators talk about a 30% improvement and hoping for 15-20% in the real world.
However some research has suggested the improvements are much less than this.
The only objective measured data I could find showed a conventional PWM regulator outperformed 2 MPPT regulators !!
It has also seem sensible that with the shadows etc present on a boat a single MPPT regulator would have a lot of compromise tracking the multiple panels.
Large domestic solar installations talk about most of the gain from MPPT is from the reduced wire losses, which don’t apply in the parallel configuration used on many yachts.

Has anyone done some objective tests? In a yacht with a parallel solar array it would be easy to bypass the regulator and connect the panels directly to the batteries. I would be interested what real world gains MPPT produce on a boat. under various conditions.(If anyone is prepared to do this a multimeter (don’t go over the 10 or 20a limit of most meters) or clamp on multimeter would be best. There have been reports of many regulators overestimating the current delivery.

[witzgall]

Noelex;

Could you post the study you found, so we could all look at it? I have never heard anything like this before.

Chris

[noelex 77]

Originally Posted by witzgall

Noelex;

Could you post the study you found, so we could all look at it? I have never heard anything like this before.

Chris

The web page was:
Another Newby : General Talk
The PL20 is conventional (none MPPT) regulator. As you can see in this test it performed better. I have no idea if there was a problem with the test, or if a different MPPT would have performed better.
I have always assumed the gains from MPPT would significantly out way the losses from the voltage conversion and powering the regulator, but assumptions can be misleading.
I am beginning to have some doubts especially a single regulator with multiple panels with varying shadowing , which is typical of many boat situations.
It would certainly be nice to quantify the gains (or otherwise) experienced when using a MPPT regulator in the unusual solar installation that is present on most boats.

[cal40john]

I didn't see where he talks about the state of charge of the batteries. As stated by another poster on that thread, MPPT only gives you increased performance when the regulator is not limiting output. It only makes sense to maximise power transfer when you're trying to use the maximum capabilities of the panel. Once the batteries reach the state of charge that the regulator is reducing the current you're no longer trying to maximize output and the MPPT function isn't being used. Three 80 watt panels should put out 20 amps, you only see 10 on any of the tests. It appears to me that this was a test of regulation, not maximizing output. Redo the test with heavily discharged batteries.

As one of the people on that thread posted, if your batteries are usually close enough to charged up that the regulator is regulating the output, the extra cost of MPPT is a waste of money.

John

[noelex 77]

Originally Posted by cal40john

I didn't see where he talks about the state of charge of the batteries. As stated by another poster on that thread, MPPT only gives you increased performance when the regulator is not limiting output. It only makes sense to maximise power transfer when you're trying to use the maximum capabilities of the panel. Once the batteries reach the state of charge that the regulator is reducing the current you're no longer trying to maximize output and the MPPT function isn't being used. Three 80 watt panels should put out 20 amps, you only see 10 on any of the tests. It appears to me that this was a test of regulation, not maximizing output. Redo the test with heavily discharged batteries.

As one of the people on that thread posted, if your batteries are usually close enough to charged up that the regulator is regulating the output, the extra cost of MPPT is a waste of money.

John

I am certainly not convinced by the results of the test I posted, hence my call for some more data.. As you point out there are many possible flaws.
However, I don’t think however the regulators were close to limiting the current. The test was stopped at 10am given that the batteries were accepting a lower charge before this its seems unlikely that there was any limitation of the charge current. Regulation would be seen by a reduction of charge acceptance, despite higher luminous flux as the sun rose higher.

[cal40john]

Originally Posted by noelex 77

I am certainly not convinced by the results of the test I posted, hence my call for some more data.. As you point out there are many possible flaws.
However, I don’t think however the regulators were close to limiting the current. The test was stopped at 10am given that the batteries were accepting a lower charge before this its seems unlikely that there was any limitation of the charge current. Regulation would be seen by a reduction of charge acceptance, despite higher luminous flux as the sun rose higher.

Actually he provides the answer to this, I didn't catch it the first pass. His last column shows an 80 watt panel short circuited. Multiply that number by 3 to give you close to what the 3 panels together should be putting out for the light given.


From: Electrical Characteristics of Solar Panels or PV Modules - AltE



John

[hellosailor]

The PL-20 numbers (provided by PL20?) only show amperage. MPPT controllers do not necessarily use a 3-stage charge profile, so simply amperage with an unknown battery discharge depth may not be telling the whole story.

First of all, MPPT controllers ARE PWM. The difference is MPPT adds brains on top of the PWM, continuously adjusting the voltage and amperage to provide maximum effective charging power--as defined by each manufacturer after rigorous application of trade secrets, black magic, and proprietary information. (No, really, that's corporate espionage territory.)

The "rules" for the most effective charging amperage and voltage apparently are more complex than the simple 3-stage charging conditions most of us have been taught. The MPPT controllers typically claim a 2-4% power loss for their overhead. PWM, no matter how it is controlled, is generally considered to give about 10% more efficiency than straight DC charging.

But how is a plain PWM charger dealing with overvoltage from the solar panels? Amperage claims are nice, but is that "amperage" averaged? Averaged how? Are the pulses 100 milliseconds on, 100 off? 10 off? 1 on 1 off? And where's the power going to, or not going, when the pulse isn't being applied? In an MPPT system is is being stored and transformed, into either voltage or amperage, as the charger determines appropriate.

Taking a straight DC charging source (solar) and chopping the output into pulses, without saying what happens in between them, or what happens when the solar panels over/under volt compared to the desired "14" volts for charging, sounds like oversimplifying the real situation.

All I know if that from watching the output of an MPPT system (voltage and amperage both measured continuously) I was surprised to see it keep "tickling" the batteries with "just enough" voltage as it cranked in amps. Not a 3-stage profile, but an optimized charging profile, revised every minute(?) or two, based on battery charge state (programmed in capacity) as well as available power.

How much brains the MPPT system (or the PWM system) really has, and how it is really controlling things...spy versus spy, no one wants to get too explicit about that.

[Paul Elliott]

Originally Posted by cal40john

Actually he provides the answer to this, I didn't catch it the first pass. His last column shows an 80 watt panel short circuited. Multiply that number by 3 to give you close to what the 3 panels together should be putting out for the light given.

And you will notice that in no case do either the PL20 (conventional) or the MPPT-250 regulators put out anything close to the stated short-circuit current of the three paralleled panels. This means that the regulators are in charge-limiting mode and the differences are probably due to the specific setpoints and charging modes for the regulators. This is not a valid test of MTTP efficiency.

It is of course important how the regulators manage the Absorption and Float stages (where MPPT is irrelevant), but to test the claimed gains from MPPT you need a well-discharged battery.

For what it's worth, I've used Blue Sky MPPT controllers, and they seem to give me slightly better charging performance than a standard controller. Unfortunately, where I had the MPPT controllers mounted they would get the occasional drop of salt water, and this was enough to kill two of them. When this happened the second time (on a passage to Hawaii), I just bypassed the controller and connected my 3 100W panels directly to the battery (through circuit the breakers). I was using enough power while at sea so that overcharging was not an issue. Once in Hawaii I bought an inexpensive weatherproof non-MPPT controller and have been using that since then. It works well enough for now.

[cuthbert]

Originally Posted by hellosailor

First of all, MPPT controllers ARE PWM. The difference is MPPT adds brains on top of the PWM, continuously adjusting the voltage and amperage to provide maximum effective charging power--as defined by each manufacturer after rigorous application of trade secrets, black magic, and proprietary information. (No, really, that's corporate espionage territory.)

The "rules" for the most effective charging amperage and voltage apparently are more complex than the simple 3-stage charging conditions most of us have been taught. The MPPT controllers typically claim a 2-4% power loss for their overhead. PWM, no matter how it is controlled, is generally considered to give about 10% more efficiency than straight DC charging.

But how is a plain PWM charger dealing with overvoltage from the solar panels? Amperage claims are nice, but is that "amperage" averaged? Averaged how? Are the pulses 100 milliseconds on, 100 off? 10 off? 1 on 1 off? And where's the power going to, or not going, when the pulse isn't being applied? In an MPPT system is is being stored and transformed, into either voltage or amperage, as the charger determines appropriate.

How much brains the MPPT system (or the PWM system) really has, and how it is really controlling things...spy versus spy, no one wants to get too explicit about that.

Some good questions/points here.

First off, is an MPPT PWM? Yes and no. In the truest sense any solar regulator is PWM as it will stop passing current when the voltage across the batteries denotes full charge and restart again when it has dropped. Albeit big pulses of long duration, but you could (and some do) sell that as PWM, most would consider a true PWM to be a charging algorithm that typically was regulated by both the battery volatge and temperature and the pulses of such short duration that the overall power output could be controlled by regulating the pulse width (typically using a microprocessor).

The potential power of the panel does not have to be used in the same way as a wind generator must be and most are left open circuit when the available power is not needed.

But this is true of any solar regulator, the potential advantage of a MPPT is that it has "intelligence" in that it "looks" back at the panels and changes its impedance to maintain the voltage/current curves on the "knee" to keep the panels at maximum power. However, a well matched panel to charger impedance on the most cheap and cheerful system will also give near optimum results, hence you could see no advantage with an MPPT in some cases.

Its this variance of potential vs actual that gives the price advantage in designing and manufacturing MPPT chargers and selling at what seems to me ( I'm a EE with 20 years of experience) early technology adaptor prices. I myself will not buy one YET, because it is still a market at premium price and homebrew algorithms. However, the solar market is booming and the IC manufacturers are moving fast into this area. NXP announced a new IC late summer, an MPPT IC incorporating a 32 bit processor and algorithm, now the first samples are available at $5.70.

http://www.nxp.com/news/content/file_1752.html

So in response to the OP's original comment about looking to upgrade, I would say hold on another six months or so and we should see nearly every solar controller moving towards being MPPT and at a price similar to the most basic products on the market.

[SV Demeter]

So cuthbert I would be a fool to pay $550 for that sleek looking Outback Flexmax 60 MPPT controller cause in 6 months its gonna cost how much?

Very interesting thread here for sure. Thanks all.

[cuthbert]

Originally Posted by SV Escape Plan

So cuthbert I would be a fool to pay $550 for that sleek looking Outback Flexmax 60 MPPT controller cause in 6 months its gonna cost how much?

Very interesting thread here for sure. Thanks all.

I have 28" samsung LCD tv that cost me $2000 in best buy 5 years ago, but I wanted one then. I don't think the prices will plummet but as more ic's become available with more features and low costs more manufacturers will step into the market or the existing market players such as outback will be able to shrink their product size and cost overhead and still have a good margin.

[cfarrar]

Paul,

How many amps did your 3 100W panels provide to the batteries in direct sun, without a controller? Let's assume batteries at about a 50-75% charge state - or around about 12.2-12.6 volts.

[Our Blue Sky MMPT controllers would provide about 17.5-17.7 amps in that situation, I figure].

[Boracay]

I don't know if it is relevant but would multiple small, cheap regulators each wired (in my case) to one or two 50w panels with the output going to the batteries (T105 bank) via some fairly heavy duty leads (12m long!) make sense?

Doing a balancing act while installing 100+w panels on top of my cockpit cover gets less appealing as I age...

[Paul Elliott]

Originally Posted by cfarrar

Paul,

How many amps did your 3 100W panels provide to the batteries in direct sun, without a controller? Let's assume batteries at about a 50-75% charge state - or around about 12.2-12.6 volts.

[Our Blue Sky MMPT controllers would provide about 17.5-17.7 amps in that situation, I figure].

Unfortunately, I never made accurate comparisons, and since MPPT #2 is now toast, I won't have the chance to do the test unless I get another MPPT. It would be simple enough, I would use my clamp-on ammeter and wire up a switch to do an A/B comparison. I would want to be able to switch back and forth quickly as the sun angle and battery state are always changing.

I did test one of the panels several years ago when I first got them, and it put out about what the specs said it should: Open circuit voltage, short circuit current, and current at a few load points. Remember that the panel output depends on the sunlight level (and angle) and temperature, so it's pretty complicated to make comparisons at different latitudes, seasons, mounting configuration, etc. It did seem that the MPPT was putting out about 20% more charging current than a direct connection when the batteries were low, but it may have been closer to 10%. I just don't have the data.

[noelex 77]

Originally Posted by Paul Elliott

And you will notice that in no case do either the PL20 (conventional) or the MPPT-250 regulators put out anything close to the stated short-circuit current of the three paralleled panels. This means that the regulators are in charge-limiting mode and the differences are probably due to the specific setpoints and charging modes for the regulators. This is not a valid test of MTTP efficiency.

I think it has more to do with time of day. The test was stopped at 10am before the panels were receiving enough light to put out their rated current.
I don’t see how the regulator could be charge limiting at say 6:30 am at 1.1A, but an hour later the batteries will accept 4A an hour later again they will accept 8A.
This profile is characteristic of the regulators in boost phase when the batteries will accept all the power the solar panels can put out.