Why is mppt better than pwm?

[fat_old_sun]

So what configuration of the 5 panels you used when testing the PWM SCC? all 5 in parallel?
What is the Is and Imp of the panels?

all panels in series for MPPT and all in parallel for PWM. Panel Isc 6.85A & Imp 6.25A

 

[Bud Martin]

all panels in series for MPPT and all in parallel for PWM. Panel Isc 6.85A & Imp 6.25A

For the PWM SCC with series MOSFET's which are in series with batteries and with Gate PWM driven, the max current will be what the panels current are and not more during bulk charging, so 30A you are seeing is correct.
BTW, I'm on North bay and it is nice and sunny so my system is doing great, but does drop on hot day due to panel temp rise, I am ready for PG&E rolling blackout.

 

[fat_old_sun]

Are you saying that with MPPT your 5 panels in parallel will under perform the PWM?

Or, are you simply showing a pathological case where MPPT can be mis-configured to perform worse than PWM?

Under the ambient conditions of the test today, yes, a MPPT performed slightly worse than the PWM. As I explained earlier, the panels were at 55C. This alone causes a 15% drop in their Vmp.

The Epever MPPT was not mis-configured. There are some factors to consider in the MPPT vs PWM debate. In my tests today, I believe the panels' temperature played a big role to cause the MPPT perform not as well as the PWM.

There's a paper by Victron that compares the performance PWM & MPPT as a function of panel temperature. It clearly shows the efficiency hit an MPPT takes at higher panel temperatures. What the paper doesn't appear to consider is the power consumption by the MPPT circuit itself; the PWM I used probably consumed 4-5 Watts, while the MPPT consumed close to 30 Watts - there's a reason why it has a heat sink 4X that of the pwm controller.

 

[Warpspeed]

A confidentiality agreement with my employer means I cannot post the actual official circuit on the internet, but I can sketch something from memory if that would help.

They are all a very conventional buck converter stage that consists of a electrolytic across the dc input from the solar panels, a high side mosfet switch, a fast recovery catch diode, a series inductor, then a second electrolytic after the inductor across the dc output to the battery.
Five basic components.

That, when driven with a suitable duty cycle will decrease the dc solar panel voltage, and increase the dc charging current, as well as control the loading on the solar panels.

Its this duty cycle control that determines what the whole thing does. It can be given a whole range of functional features, including optionally power point tracking.

 

[Bud Martin]

A confidentiality agreement with my employer means I cannot post the actual official circuit on the internet, but I can sketch something from memory if that would help.

They are all a very conventional buck converter stage that consists of a electrolytic across the dc input from the solar panels, a high side mosfet switch, a fast recovery catch diode, a series inductor, then a second electrolytic after the inductor across the dc output to the battery.
Five basic components.

That, when driven with a suitable duty cycle will decrease the dc solar panel voltage, and increase the dc charging current, as well as control the loading on the solar panels.

Its this duty cycle control that determines what the whole thing does. It can be given a whole range of functional features, including optionally power point tracking.

That is the buck converter topology that I have seen in the MPPT SCC, but how about the PWM SCC that uses the series MOSFET topology which does not convert high PV Voltage low current to low Voltage and high current to charge the batteries? So your company does not make the so called 'PWM SCC' similar to the videos.
BTW< I also use lots of Buck (instead of linear regulator) and Boost converter topology in the products I design.
Here is what inside the typical cheap PWM SCC, just a bunch of power MOSFETS, no big power inductor since there is no buck converter (only small buck for the USB PS).

 

[Warpspeed]

Just hard switching the solar panels without a choke, its difficult to see how something like that could be EMC compliant, or even work very well.

But then the Chinese can claim compliance or anything else, and as there is no functional legal system in China (as we know it in the West) why should they even care.

The Chinese can make all kinds of claims on the box and in advertising, none of which need to be true, and there is absolutely nothing you can do about that.
Caveat emptor.

There is just so much trickery and deception these days, it makes it very difficult for the less knowledgeable to research anything properly or make rational decisions on what to purchase. Worse still, people read a lot of stuff on the internet and believe they are learning something worthwhile, where in fact, they are unknowingly just repeating and arguing total B.S.

I am beginning to realize its a total waste of time attempting to discuss technical matters on internet Forums these days.
All very discouraging, but that is how it is.
Forums are for entertainment purposes only, not for serious factual discussion.

 

[Bud Martin]

We are not talking about the compliance, we are talking about topology.
So company like Morningstar is also one those company with trickery PWM SCC?

World's Leading Solar Charge Controllers | Over 4 Million Sold

Optimize battery health with the best solar charge controllers. Choose from a wide selection of MPPT and PWM options with advanced features and certifications.

www.morningstarcorp.com

 

[gelmjw]

@

fat_old_sun​

I liked your post. Your measurements and stats are fine.

What did you expect before you set up the test?

Was 5S the most advantageous configuration for MPPT? Would 5P be worse because the VMP could be reduced to 14.25V(16.77 - 15%) as 14.25V would not likely allow any MPPT? I'm not sure 14.25V is enough to charge in any case.

I also believe MPPT is at a disadvantage at these marginal Watts. What would you do with 1000W?

 

[fat_old_sun]

The panels were in series for the MPPT and so, there was sufficient headroom. I didn't measure each panel's output voltage, but the string's voltage a little over 75V. so the individual panel voltages must have been around 15V.

I used 5 panels simply to max out each controller. With four (or six) I can try 2S2p or 3S2P, but with five panels 5S is the only option.

As to the advantages, with an MPPT a single pair of wires from the RV's rooftop will do; however, this wire must be 8 or 10 AWG if you want to minimize voltage drop. With a PWM, you can use thinner wires (since you don't care about the voltage loss, up to a point) but have to run several pairs.

In my view, the choice of panels determine if one should use MPPT or PWM. If you're using residential style 300+ Watt panels, go with an MPPT. If you're using smaller 12V 100w panels, a PWM would be better.

 

[12VoltInstalls]

Forums are for entertainment purposes only, not for serious factual discussion.

I think that is neither empirically nor subjectively supportable.

In this “forum” (sic) alone I have learned much that supports and strengthens past understanding(s) as well as learned- or relearned- many things. All of which encompasses factual discussion.

Further, by your own colloquial utterance you- if your statement was in fact: true- have dismissed and negated your contribution and comments as well as stated they have no value beyond a casual titillation of the audience.

Thankfully at least that claim is subjective drivel.

 

[rin67630]

In my view, the choice of panels determine if one should use MPPT or PWM. If you're using residential style 300+ Watt panels, go with an MPPT. If you're using smaller 12V 100w panels, a PWM would be better.

ABSOLUTELY !
Residential style 300+ Watt panels are usually 36V MPP. fine to use with a 24V battery and PWM or a 12V battery with MPPT, but not with a 12V battery and PWM.
100W Panels are mainly 18V MPP, fine to use with a 12V battery and PWM, but not to use with a 12V battery and MPPT, a very frequent mistake.

 

[rin67630]

Here is what inside the typical cheap PWM SCC, just a bunch of power MOSFETS, no big power inductor since there is no buck converter (only small buck for the USB PS).

By the way, a quick test, if the sold "MPPT" is a real or a fake one, is to measure if the Panel (-) and the Battery(-) are directly connected (0Ω) or not.
MPPT controllers usually switch high side and Panel (+) and the Battery(+) are separated by a FET transistor and the Buck's coil.

PWM controllers usually switch low side and Panel (-) and the Battery(-) are separated by a FET transistor.
Their Panel (+) and the Battery(+) are directly connected (0Ω).

There might be exceptions, but to 99% this rule is valid.

 

[sunshine]

100W Panels are mainly 18V MPP, fine to use with a 12V battery and PWM, but not to use with a 12V battery and MPPT, a very frequent mistake.

The biggest mistake is comparing a method of voltage regulation, PWM, with a totally different subject, MPPT, that is used to control that regulation by whatever method is used.
Fair enough to follow the jargon ...PWM CC and MPP CC...for a while because that is the common generic label but by page 10 the correct distinction should have been made.

 

[rin67630]

The biggest mistake is comparing a method of voltage regulation, PWM, with a totally different subject, MPPT, that is used to control that regulation by whatever method is used.
Fair enough to follow the jargon ...PWM CC and MPP CC...for a while because that is the common generic label but by page 10 the correct distinction should have been made.

In fact one should have named PWM controllers "dumb battery voltage" (DBV) controllers, most of them do not even have any measure of current.

 

[sunshine]

In fact one should have named PWM controllers "dumb battery voltage" (DBV) controllers, most of them do not even have any measure of current.

Yes. This should be the main emphasis when comparing the differences between various charge controllers....explaining what they can do and what they can not so anyone can make a decision based on their needs and knowledge level.

"most of them do not even have any measure of current.".....fair enough since they are only voltage controllers so they do not need to know!

 

[Warpspeed]

Its really all a simple matter of matching impedances.

If the rated peak power voltage of the solar array is very close to matching the required charging voltage of the battery, then just connecting the two directly together will pretty much guarantee the most efficient power transfer possible without needing anything else.

Connecting a dumb on/off controller between the two will not alter that, it should still work surprisingly well and serve the single purpose of preventing overcharging of the battery.

If you go to the other extreme, where the rated peak power panel voltage is MUCH higher, say twice (or more) the required battery charging voltage, then a properly designed high frequency buck regulator would be quite capable of matching the two voltages at a very high conversion efficiency, and provide optimum power transfer by driving the buck regulator at the appropriate duty cycle.

So there are really two types of PWM under discussion. The simple very crude (fake pwm) on off switch type, which can work well only if the solar and battery charging voltages are pretty much the same. And the more sophisticated high frequency PWM buck regulator which can efficiently transfer power when the solar voltage is a fair bit higher than the battery voltage.

Unfortunately many people just don't understand any of the electrical theory behind all of this, and when the Chinese resort to out right lying and deception in their published disinformation, that does not help the situation either.

 

[rin67630]

"most of them do not even have any measure of current.".....fair enough since they are only voltage controllers so they do not need to know!

The better ones do measure the battery current as well and can limit the precharge and the bulk charge average current by PWM to protect the battery, but they are quite rare.
A PWM controller should also include a blocking diode (better an ideal one) to prevent current flowing back from the battery into the panels. Many do not. If you make a short on the panel side with the battery connected, you will be greeted by Aladdin getting out of the SCC !...

 

[Don B. Cilly]

I have a cheap-as-it-gets one that I use to charge things "on the fly" (I have a few old small panels lying around) and it does measure current, because it shows a reasonably accurate value of it on the display.
And it's not even blue... well it has a blue stripe of sorts and two blue buttons ;·)



Also MOSFETs will not arrow reverse current.
-

 

[efficientPV]

By the way, a quick test, if the sold "MPPT" is a real or a fake one, is to measure if the Panel (-) and the Battery(-) are directly connected (0Ω) or not.
MPPT controllers usually switch high side and Panel (+) and the Battery(+) are separated by a FET transistor and the Buck's coil.

PWM controllers usually switch low side and Panel (-) and the Battery(-) are separated by a FET transistor.
Their Panel (+) and the Battery(+) are directly connected (0Ω).

There might be exceptions, but to 99% this rule is valid.

Not a good test. Many MPPT if not most use a FET on the low side panel - as the blocking diode.

 

[rin67630]

Not a good test. Many MPPT if not most use a FET on the low side panel - as the blocking diode.

A regular ideal diode for the blocking function is high side.