Why is mppt better than pwm?

[Bossrox]

I was just pondering what makes an mppt so much better than a pwm? They both do the same thing but a pwm is a better value. I've sat & watched my 2 victron mppt's in full sun doing some kind of internal gymnastics with the output waffling all over the place quite frequently like it's trying to figure out what to do which is wasting lost energy. If you got a 1000 watts going into either type, you get 1000 watts out minus any efficiency loss so I just don't get the point of what makes an mppt better.

 

[Primal1]

When we made the switch from pwm to mppt we noticed 2 major differences

We made float several hours earlier on normal days, usually we made float around 4pm with the pwm, 2pm float ( or earlier) is now the norm.

We make float more often on cloudy, rainy and winter days than with pwm.

Nothing else was changed in the system and we saw massive improvement in performance of the overall system. Renogy wander pwm was replaced with an Epever MPPT if tbe brand performance could make a difference.

 

[Diysolar123]

old will prowse video

 

[snoobler]

I was just pondering what makes an mppt so much better than a pwm? They both do the same thing but a pwm is a better value. I've sat & watched my 2 victron mppt's in full sun doing some kind of internal gymnastics with the output waffling all over the place quite frequently like it's trying to figure out what to do which is wasting lost energy. If you got a 1000 watts going into either type, you get 1000 watts out minus any efficiency loss so I just don't get the point of what makes an mppt better.

MPPT will (almost) always gather more power.

PWM operates by shorting the panels to the battery, forcing them down to battery voltage. If a panel's Vmp is at 17, and you force it to 13.5, it's going to take a 3.5/17 = 20.5% hit to performance. In general, you'll NEVER get more than 80% of rated out of an array connected to a PWM controller in perfect solar conditions.

MPPT allows the panels to operate at peak power.

PWM can't have high PV input voltage, they need Vmp about 1.5X battery nominal voltage, so current coming into PWM controllers is higher and array voltage drop is worse. Your 3S12P array would likely need to be 2S18P and have greater wiring losses between the array and PWM.

MPPT is tolerant of being over-paneled

PWM can't be over-paneled.

Concerning value... If you're talking $/kWh, then you have to look at it this way... How much more does the MPPT cost than the PWM when you consider that you have to buy 10-20% more of an array to match the output of the MPPT?

 

[Bossrox]

Snoobler, you always got a sensible answer. Smart fella! Thx!

 

[MichaelK]

Besides Snoobler's excellent outline, there's one additional advantage I can mention. With PWM you must carefully match the panels voltage to the battery bank. This means you more or less HAVE to use 12V panels that put out only 15-18VDC. You pay a price premium for 12V panels.

MPPT allows you to swap in far cheaper surplus grid-tie panels that normally operate at much higher voltages. The MPPT controller transforms the raw high voltage grid-tie voltage to exactly what the battery wants. So, the W/$ is significantly higher with MPPT coupled with grid-tie panels.

 

[pollenface]

As a rule of thumb, you always want to pick a PWM controller that is 120% the Isc of your PV array.

I think a 60 cell panel (29-31vmp) with PWM charging a 24v bank would be fairly good value.

 

[snoobler]

As a rule of thumb, you always want to pick a PWM controller that is 120% the Isc of your PV array.

Yep. They can't regulate current... they just pass along whatever the panels are giving, and they have to be rated for it. Definitely wise to have some headroom.

I think a 60 cell panel (29-31vmp) with PWM charging a 24v bank would be fairly good value.

Yep. 60 cell with PWM is the only option that doesn't take a big performance hit, but they are not great if you have FLA and need to do equalization on them.

 

[NwCali]

I suspect the OP has either a mismatched system or a setting wrong then?

 

[snoobler]

I suspect the OP has either a mismatched system or a setting wrong then?

Good point... nothing here has addressed his concern:

@Bossrox

As I understand it:

3S12P 10kW, 48V system, 2X Victron 150/100.

You don't care for their erratic behavior/power fluctuations in full sun.

1) if they aren't synchronized in a VE.Smart bluetooth network, they should be. That will allow them to coordinate.
2) Does this happen all the time, or more commonly in absorption/float+load phases? When in absorption/float+load, they are voltage limited, and since you have two on there, they will react to each other if not synchronized. Even if synchronized, they may still get wonky from time to time.

 

[rin67630]

MPPT will always gather more power.
...

My god!
Repeating so much wrong statements... from a moderator here!
I did already explain that that such general statements just lack veracity. One should consider things with a bit more differentiation.

No, MPPT will not ALWAYS gather more power!
(I am speaking about off grid situations only. On grid, there is no discussion MPPT is a must.)

Off-grid, there here are two situations, where MPPT just gathers less power than a PWM controller.
Using a panel of the same nominal voltage than the battery,
(e.g. a 12V lead acid battery with typical 18Vmp panel (<100W) or a 24V acid battery with typical 36Vmp panel)
and
a) the panel gets very hot (under tropical conditions)
b) the solar irradiance it is very low (under 50W/m²)
under these conditions the panel's Vmp (which is not a constant) comes very close to the battery voltage and a MPPT controller will not be able to handle efficiently the tiny voltage difference -> a PWM with just harvest more.

Finally, there is a reason, why panels are designed that way (Vmp18 or 36V) :
They were historically optimised for PWM usage with standard batteries.

MPPT needs a good voltage difference, else they may operate, yes. But they will operate miserably!

The rule of thumb is to use MPPT with 36Vmp panels for a 12 V battery and 72V panels (or panel combination) for a 24V battery.
Do not use MPPT with matching nominal voltage panels! (18Vmp with a 12V battery or 36Vmp with a 24V battery).

Then there is also one common situation where MPPT gathers more power than a PWM controller, but it just doesn't matter.
If you are in a sunny situation and your battery is full at 11:00 anyway.
After that you will just be throwing energy away, so it just does not matter if you got 100% at 11:00 or at 12:00.

Last but not least, PWM controllers have less own consumption than MPPT controllers, on designing low-power solar powered systems, it may matter.

PWM can't be over-paneled.

...

Of course you can! it is even recommended!
The only caveat is to match the nominal voltages.

 

[rin67630]

As a rule of thumb, you always want to pick a PWM controller that is 120% the Isc of your PV array.

I think a 60 cell panel (29-31vmp) with PWM charging a 24v bank would be fairly good value.

I don't think so.
Having a PWM controller that can handle more amps is not a disadvantage.
Under hot conditions or over-cloudy weather 36Vmp is better. (The 29-31vmp tag value will become a real 25Vmp and not be enough any more to full charge your battery to the 28.8V cycle value).

 

[Tecnodave]

Snoobler is very correct when pointing out that PWM with 60 cell panels will not equilize a FLA battery on a hot day. My main system is Rolls-Surette S-530 Flooded lead acid and need 32 volts at equilization . 72 cell panels at a minimum for a proper eq. on large format FLA banks. I do use 2 MPPT main controller , MidNite Classic 150 but have 2 auxiliary PWM controllers that serve as “wake up” early AM and ”sunset“ arrays that all controllers using same type of panel...Sharp NE-170,180 & 190’s , all 72 cell polycrystalline, 2S6P for the Classics and all parallel for the Trace C-40 PWM controllers that just wont die and go away, not that that is a bad thing...

 

[JoeHam]

Then there is also one common situation where MPPT gathers more power than a PWM controller, but it just doesn't matter.
If you are in a sunny situation and your battery is full at 11:00 anyway.
After that you will just be throwing energy away, so it just does not matter if you got 100% at 11:00 or at 12:00.

Unless the clouds roll in at 11:00.

Then it matters a lot!

 

[snoobler]

My god!
Repeating so much wrong statements... from a moderator here!
I did already explain that that such general statements just lack veracity. One should consider things with a bit more differentiation.

Sorry. Next time I'll say "almost always"

Off-grid, there here are two situations, where MPPT just gathers less power than a PWM controller.
Using a panel of the same nominal voltage than the battery,
(e.g. a 12V lead acid battery with typical 18Vmp panel (<100W) or a 24V acid battery with typical 36Vmp panel)
and
a) the panel gets very hot (under tropical conditions)
b) the solar irradiance it is very low (under 50W/m²)
under these conditions the panel's Vmp (which is not a constant) comes very close to the battery voltage and a MPPT controller will not be able to handle efficiently the tiny voltage difference -> a PWM with just harvest more.

Right. So I'll go with "almost always".

Finally, there is a reason, why panels are designed that way (Vmp18 or 36V) :
They were historically optimised for PWM usage with standard batteries.

MPPT needs a good voltage difference, else they may operate, yes. But they will operate miserably!

The rule of thumb is to use MPPT with 36Vmp panels for a 12 V battery and 72V panels (or panel combination) for a 24V battery.
Do not use MPPT with matching nominal voltage panels! (18Vmp with a 12V battery or 36Vmp with a 24V battery).

Not according to Outback as you were made aware in the other thread. They recommend 12-24V over battery nominal.

Then there is also one common situation where MPPT gathers more power than a PWM controller, but it just doesn't matter.
If you are in a sunny situation and your battery is full at 11:00 anyway.
After that you will just be throwing energy away, so it just does not matter if you got 100% at 11:00 or at 12:00.

Right. And nobody uses loads at 11-12. When I'm running my RV A/C and other loads that are close to my array output, I'm glad I have an MPPT.

Last but not least, PWM controllers have less own consumption than MPPT controllers, on designing low-power solar powered systems, it may matter.

And we know this is the basis of your anti-MPPT crusade.

Of course you can! it is even recommended!
The only caveat is to match the nominal voltages.

Completely untrue. A PWM charge controller can't regulate current. If you put an array that will output 70A on a 60A PWM controller, you'll cook it.

 

[Wibla]

Speaking as an electrical engineer, it boggles my mind that people are still trying to claim that PWM is better than MPPT

 

[rin67630]

Unless the clouds roll in at 11:00. Then it matters a lot!

Probably not. The rest of the day might be sufficient to get full charge anyway.
When you design an off-grid solar system (without gas generator-back-up) the battery must last for at least 4 days without noticeable solar contribution.
Here, with our "english" weather, it is not uncommon to have 3 weeks and more without direct sun with 50W/m² irradiance only.
So the only way is to massively over panel to be able to charge at least to 50% also under these conditions.
Then every hour of sun rays will bring you a full charge.

Speaking as an electrical engineer, it boggles my mind that people are still trying to claim that PWM is better than MPPT

You are probably addressing me, aren't you?
Where did i claim that "PWM is better than MPPT" without explaining when and only when?

I just corrected the wrong statement that MPTT is ALWAYS better than PWM.

I just mentioned the situations where this is not true.
And at a rule of thumb is that you should not use MPTT with a panel that has not enough voltage (at least 12 to 20V) over the battery maximum voltage. A lot of people do, just because they are told "that MPPT" is superior.
3V over Vmaxbat e.g. Vmp = 18V with a 12V lead-acid battery is not enough, your MPPT controller will deliver a miserable performance!

And now I'll stop preaching.
If some are fine with their simplistic paradigms, they should remain with them.

 

[rin67630]

Completely untrue. A PWM charge controller can't regulate current. If you put an array that will output 70A on a 60A PWM controller, you'll cook it.

Practically not. The battery's dynamic internal resistance will prevent that. The SCC is not working against an iron wall.

 

[Wibla]

You are probably addressing me, aren't you?

What would make you think that?

I just corrected the wrong statement that MPTT is ALWAYS better than PWM.

By construing a situation that will almost never occur for most users of these systems.

I just mentioned the situations where this is not true.
And at a rule of thumb is that you should not use MPTT with a panel that has not enough voltage (at least 12 to 20V) over the battery maximum voltage. A lot of people do, just because they are told "that MPPT" is superior.
3V over Vmaxbat e.g. Vmp = 18V with a 12V lead-acid battery is not enough, your MPPT controller will deliver a miserable performance!

This has already been refuted by snoobler, but I'll add on to the pile because I can, using a real-word example:
My boat setup is (currently) a very modest 2x30W panel setup with 21Voc/18Vmp and a Victron 75/15, wired to a lead acid battery bank. I've tried to connect the panels up in both series and parallell, and I'm consistently able to recharge the battery bank to float with the panels in parallell even during hot days. The Victron only need 5V delta over Vbat to start up, and will keep going until the delta is 2V. On an unseasonably hot day, I'll get 17Vmp, that's still 15V (potentially) into the battery, thus more than enough to get the job done.

Practically not. The battery's dynamic internal resistance will prevent that. The SCC is not working against an iron wall.

Meet my friend, the LiFePo4 battery!

 

[snoobler]

Practically not. The battery's dynamic internal resistance will prevent that. The SCC is not working against an iron wall.

Please elaborate.

If you're still dealing with tiny SLA, then I might agree as they can't accept much current, but if you're dealing with a typical large bank for off-grid storage, you can push much larger currents, and the battery will just take it.

Example: I have a 936Ah 12V battery sitting next to me. It should be charged at around 100A. I routinely push 300A into it starting at about 60-70% SoC, and it stays below absorption voltage. If I had 60A of PV potential on a 40A controller, my battery would readily accept the full 60A and burn out the controller.