PWM versus MPPT

[hellosailor]

socal-
"Why not get the extra 20 mpg, especially if the price of the Prius is only 20% more in initial purchase price? "
Actually a friend bought one of the early Priuses. The exact math worked out that IF the battery lasted ten years and IF he got the $10,000(?) federal/state rebate and IF he drove something like 20,000 miles a year...it would break even with the car I had bought for about 50-60% of the upfront price.
So the devil is very much in the details on how it does or doesn't breal even. If it doesn't break even (and in my case, with lower miles driven every year) then it doesn't matter if it gets 30% better mpg, because the fixed costs swamp that out.

Paul-
Bottom line, from my point of view, is that they are both black boxes. The power output from each one is PWM, it is pulsed DC. The MPPT box makes all its magic, and then charges up the output bank, and goes "zap" every time there's enough power to reach the necessary voltage. Pretty much the same thing a PWM controller will be doing, except the MPPT controller has more options about when and how it makes each pulse.
For that matter, the old Delcotron alternators were PWM also. In the 1970's, they put out a fixed 14.3-14.4 volts, no more no less, and the alternator went into "listening mode" to sense the battery voltage (over the charging wire itself) anywhere from something like 100x to 10,000x per second. Radical technology for the time. No voltage changes, just changed the output frequency.
From the battery's point of view, either it gets pure DC or pulsed DC, and how that happens...Not my circus, not my monkey.

[Paul Elliott]

Originally Posted by hellosailor

Paul-
Bottom line, from my point of view, is that they are both black boxes. The power output from each one is PWM, it is pulsed DC. The MPPT box makes all its magic, and then charges up the output bank, and goes "zap" every time there's enough power to reach the necessary voltage. Pretty much the same thing a PWM controller will be doing, except the MPPT controller has more options about when and how it makes each pulse.

I'm OK with the black box view. But if we're going to compare PWM and MPPT, we have to recognize that there are significant and critical differences in how they operate.

PWM puts out pulsed DC. During this pulse, the solar panel is directly connected to the battery. The panel delivers whatever current it can at the battery voltage. The PWM charger controls the pulse duty-cycle and perhaps frequency as part of the charge regimen. (I realize that you already know this.)

MPPT may put out pulses, or more likely it puts out a fairly smooth DC current. This comes by way of a switching regulator that adjusts itself to find the maximum current times voltage (power) the panel can deliver. The MPPT controller is continuously adjusting to track this maximum power point based on panel illumination, temperature, etc. This typically delivers more charging current to the battery than can a PWM controller. As the battery continues to charge, the MPPT controller will back off on the current delivered to the battery (as does a PWM controller). (I'm pretty sure that you already know all this, too.)

As I see it, these differences are pretty profound. An MPPT controller probably does *not* deliver current pulses to the battery, and if it does, this is quite different than how a PWM controller does it. MPPT, in theory, provides benefits that a PWM controller can not deliver, by virtue of how these controllers operate.

An MPPT controller is *not* a fancier PWM controller, and saying so only confuses the discussion. IMNSHO (In My Not So Humble Opinion).

[hellosailor]

Good point, but it begs for someone to get out an oscilloscope and query some MPPT makers. Since PWM apparently charges batteries (especially wet lead) faster than pure DC, wouldn't it be counterproductive for MPPT makers to be trying to smooth out their power, as opposed to just pulsing it out more frequently? Or do you actually already know which makers are or aren't doing it pulsed versus heavily filtered?

It would seem that all an MPPT controller has to do, to outperform a PWM controller, is for the MPPT controller to read the battery voltage, read the array, and set the "zap now" voltage at the right level. Rashly assuming they are like Blue Sky, where the controller leads the battery voltage by a fixed amount (offhand maybe something like .3-.4 volts) while concentrating on putting out as much amperage as the battery can handle, during the bulk phase.

[Maine Sail]

A PWM controller is essentially a direct coupled path between the PV and the battery when it is below the absorption voltage limit. This is the bulk stage where the PV is continually increasing battery voltage. Once the battery has attained the controllers voltage limit it then begins the PWM absorption/float etc. stage. PWM is nothing more than controlling the ON and OFF duration in order to maintain the voltage set point, in a smooths ready fashion, and not over-shoot it. PWM control happens so fast the voltage will look rock steady on any DVM.

An MPPT controller is not a direct path between PV and battery when in bulk. It is decoupled so the panel can attain the maximum power point and the MPPT controller can then convert any excess voltage to additional current. An MPPT controller can only provide boost when the controller is in bulk. Bulk just means the PV system has not yet brought the batteries to the voltage limit point. Bulk is the only stage with an MPPT where you can get a boost..

With most marine solar systems bulk lasts a lot longer than most assume it does because the PV current is so low compared to a charger, inverter/charger or alternator. With solar on boats, unless the array is huge, we attain the absorption voltage limit higher in the SOC curve so the MPPT boost stage lasts longer than it does in most land based systems where real estate is more of a non-issue....

For MPPT or PWM controllers, when they attain the voltage limit they are both controlling the ON/OFF duty cycle or in PWM mode...

[Pete7]

Originally Posted by conachair

Just to add another angle, I'm just about to order a victron 75/15 to replace a useless PWM controller for about €50 more than the PWM cost. Not so much for the MPPT side but as the quality looks much better and the programmability much better, not locked into going to float far too early and looks like excellent data recording facilities as well.
So hard to see why to bother with PWM when there are high quality MPPT controllers available at a reasonable price.

Indeed this was my concern too, so many cheap controllers on the market that you have no idea if they are actually PWM / MPPT or just an on off switch. The solution was to buy a known brand MPPT helped because the price has dropped significantly in the last few years.

Pete

[sailorchic34]

Originally Posted by sailorboy1

"Cheaper" isn't really the goal of a solar installation is it?

For some of us cheaper or rather less expensive, is always a factor. No question that MPPT is more efficient. But if my batteries are at (OK NEAR) 100 percent by 1:30 PM Compared to 12:30 for a MPPT it's no big deal. Either works fine. I put the extra money into a third panel which gives me more bang for the buck. (This was several years ago)

Myself I have three panels as it allows two of them to be in full sun most of the time. With a sailboat, there would be times during the day where a single 300 watt panel could be shaded (I'm talking at anchor as the tides change).

For that reason, for me, having three panels and a cheap PWM controller gives me overall better system performance then a single panel with a MPPT controller would.

But I am also driven by budget limitations, at least for another 7 months when SS kicks in. That will be huge compaired to the $600/mo I've been dealing with with that work thingy. Of course I would rather make $600/mo from a boat at anchor then $7500/mo in an office. Been there done that. I might be weird that way.

Of course 13 years ago on a previous boat, I purchased a 100 watt panel for $700. My how the prices have dropped where a similar panel is $65.

I should know that I picked up a slightly used Reology 30 amp PWM controller for $25 including shipping. Good enough for me.

[sailorchic34]

Originally Posted by Maine Sail

A PWM controller is essentially a direct coupled path between the PV and the battery when it is below the absorption voltage limit. This is the bulk stage where the PV is continually increasing battery voltage. Once the battery has attained the controllers voltage limit it then begins the PWM absorption/float etc. stage. PWM is nothing more than controlling the ON and OFF duration in order to maintain the voltage set point, in a smooths ready fashion, and not over-shoot it. PWM control happens so fast the voltage will look rock steady on any DVM.

An MPPT controller is not a direct path between PV and battery when in bulk. It is decoupled so the panel can attain the maximum power point and the MPPT controller can then convert any excess voltage to additional current. An MPPT controller can only provide boost when the controller is in bulk. Bulk just means the PV system has not yet brought the batteries to the voltage limit point. Bulk is the only stage with an MPPT where you can get a boost..
.

Hey Maine Sail, I was wondering if you every compared MPPT output at various times of the year, Say a cool clear spring time day compared to a hot summer day. Just wondering as the panel thermal coefficient of the panel voltage should see a fall off of output in hot summer. Just wondered if there was real world data on that.

[conachair]

Originally Posted by sailorchic34

Hey Maine Sail, I was wondering if you every compared MPPT output at various times of the year, Say a cool clear spring time day compared to a hot summer day. Just wondering as the panel thermal coefficient of the panel voltage should see a fall off of output in hot summer. Just wondered if there was real world data on that.

Goes into it a bit in here...


https://www.victronenergy.com/upload...WM-or-MPPT.pdf

[hellosailor]

"Just wondered if there was real world data on that."
Coupla years back when I was trying to get some reality checks on solar, I recall a lot of those questions were answered on the Sandia National Labs web site. They've done a lot of work on solar and documented most of the variables. If not them...the numbers are out there, posted by credible sources.

[sailorchic34]

Originally Posted by conachair

Goes into it a bit in here...


https://www.victronenergy.com/upload...WM-or-MPPT.pdf

Not exactly what I was looking for. I am oddly well aware of how MPPT works. But there is a thermal coefficient that as the cell temperatures increase, V output decreases. So in the heat of summer there might only be 15v output from the panel mid day. This is why a 12V solar panel outputs 17 volts at STP. It allows for the voltage drop as cell temperature rises

I have not seen real world testing of that but figured Maine Sail probably has done that

[RandomTravels]

Originally Posted by conachair

Just to add another angle, I'm just about to order a victron 75/15 to replace a useless PWM controller for about €50 more than the PWM cost. Not so much for the MPPT side but as the quality looks much better and the programmability much better, not locked into going to float far too early and looks like excellent data recording facilities as well.
So hard to see why to bother with PWM when there are high quality MPPT controllers available at a reasonable price.
The PWM controller actually won't let about half the amps through even below 14v.

I have separate Victron 75/15 MPPT controllers for each of my 4 panels. I'm really pleased with how much energy I get out of this set-up under many different conditions. I also use the blue tooth dongle with their Viltron Connect app to monitor output - very convenient. I have excess energy on a typical day, so considering using the Load output from one of the controllers to dump excess energy into the water heater. Has anyone else tried this?

[conachair]

Originally Posted by sailorchic34

Not exactly what I was looking for. I am oddly well aware of how MPPT works. But there is a thermal coefficient that as the cell temperatures increase, V output decreases. So in the heat of summer there might only be 15v output from the panel mid day. This is why a 12V solar panel outputs 17 volts at STP. It allows for the voltage drop as cell temperature rises

I have not seen real world testing of that but figured Maine Sail probably has done that

??

Did you read the whole document? Much more than a discussion of the concepts behind mppt, There's half a dozen pages discussing temperature and how mppt isn't much better than pwm as the panel temperature rises.

7. Relative performance graphs
7.1 Relative performance as a function of temperature