mppt vs pwm

[pianotuna]

Hi,

A nice little video comparing mppt vs pwm with the batteries in bulk mode.

mppt vs pwm

[KJINTF]

Mark

Good to hear from you - it's been a while
What new stuff for 2015 has Rogue got to offer us?
Always in the market for new toys

[lorelec]

Salvo wrote:
Right, Don needs to review that thread.

I think Rogue dropped his gains after the discussion from 20% down to 8%; but it's been a while.

No, it's entirely possible to get 20% gain with MPPT. Much more than that, actually, if you're using an array with a Vmp far above the battery voltage (which only MPPT controllers will allow you to do with any measure of efficiency). If you've been stuck in the boondocks for a long time, you might not realize that the popular thing to do these days is to use panels meant for grid-tie installations -- which do indeed have a higher Vmp than traditional panels meant for off-grid installs.

For CONSISTENT gain, with an array that has a Vmp closer to the battery voltage (17v Vmp, for example, with a 12v battery), around 10% is typical when you factor in the effects of temperature on Vmp. This is 1990's thinking, though. Time to move on.

[Salvo]

Bingo!

Add in the decline of output power with temperature, mppt gains over pwm should be only about 8 to 5%, not 20%

This is nothing new! The 'experts' should have known this from the get-go. The Rogue guy used the same erroneous calculation to show how much better mppt was. We set him straight.

[jrnymn7]

Salvo,

If you're referring to the 4 amps used in the comparison, he does say it's roughly equivalent to a 60w panel. However, if I understand the point you're making; and if I understand how mppt works; the Imp used by mppt would be lower than the Isc used by pwm, and therefore, the power supply should have been adjusted accordingly, thus resulting in lower output wattage from the power supply, and therefore lower amperage out of the mppt.

[Salvo]

We've been over and over this issue many time. It doesn't hurt to have you think about it and try to figure out why this test doesn't represent the real world.

I'll give you a hint, mppt and pwm operate on different points of the V-I curve. Why use Imp for pwm?

[jrnymn7]

red31 wrote:
Small shunt controllers do not do abs charging, only bulk.

Yes, and that is how my modded pm4b behaves. There is, of course, a point at which amps begin to taper, but it has nothing to do with the setpoint being reached. Voltage simply continues to rise, as amps taper, until near full charge.

It was the line on page 13 that said the battery only gets what it needs, instead of having a higher voltage forced into it, that caught my curiosity.

[Ductape]

Salvo wrote:

BFL can tell you the reason.

I'm sure you believe that your method of disagreeing with others while providing no substance other than to delegate is effective. I see it differently, but I doubt that will change unless you think about it and come to the same conclusion for yourself.

I say this in the sincere hope that your posts will be more technically detailed, as I'm sure you have a lot to share.

Now, back to our regularly scheduled programming I hope. 😉

[BFL13]

From the ASC manual: (It has a variation of the on/off as described below)

"CHARGING METHOD: Low Frequency Pulse Width Modulation : - The ASC
provides pulses of charging current at varying durations to maintain the batteries at
a full state of charge and insure the long-term health on the batteries. The “on” and
“off” cycle time is wide enough to eliminate electronic noise and short enough to
insure the batteries are always topped off. This method allows the batteries to
reach a higher voltage with the “on” cycle and protect the batteries from gassing in
the “off” cycle. This will provide the benefits of the higher voltage (reduced
sulfation and stirred up electrolyte) and prevent excessive gassing and excessive
water loss.
A TYPICAL DAY: - A typical daily cycle will be as follows. As the charging starts for
the day and battery capacity is low, charging will be continuous. As the battery
charges up, current will pass into the battery for a while and eventually stop. Later,
charging will resume and the system will continue this cycle throughout the day.
During the course of the day, the duration of the charging period of each cycle will
get shorter (cycling on for shorter periods and staying off longer). When the battery
is close to full charge, it will pulse current into the battery to achieve and maintain
full charge. This pulse charging is indicated by the "CHARGING" light occasionally
turning on and off."

Note that the battery reaches its high set-point before it is fully charged, so each charging event gets it higher in SOC and that makes the durations change each event. These charging events are rapid so it makes you dizzy as voltage and amps go up and down with each on/off.

The Series method is different, where Vbs is maintained steady instead of being turned on and off.

I had an ASC that did the above and then a LandStar Series type that held Vabs while amps tapered. I don't know if there is any quality of recharge difference in the two methods but the Series method is smoother to observe than giving the battery shots of voltage and current. Can't say whether the battery likes one of them better.

[red31]

Small shunt controllers do not do abs charging, only bulk.

[jrnymn7]

The argument for 'on/off', on page 13 of the pv7 manual, reminded me of Vector's charging profile. And when I modded my pm4b45, it started using a similar profile.

Instead of holding Vabs at the setpoint, Vabs slowly rises as battery soc rises, and reaches the setpoint at near full charge. IIRC, this does not increase charging time. It seems 'on/off' would have a very similar profile?

[red31]

direct connect, PWM and shunt will all get the battery to 14.2v in the same period of time (let's ignore efficiency differences)! the shunt turns off and waits while the battery discharges, direct connect keeps charging till the operator disconnects, PWM takes the batt up to some set point (generally above the shunt 14.2v set point) and then we all know what it does from there!

What if the video had used a 12.000v batt, 48watts vs 70, that's 45% gain.

[jrnymn7]

A few things I noticed in the video:

1) The 24% gain was at peak conditions.

2) The positive lead from the fluke meter is attached to the neg battery post. Is that just to insert the fluke into the circuit for metering?

3) The pv7 is not pwm, in fact the manual says this:

"Charges batteries at full power, below the plate saturation point. This charges batteries quickly while reducing the electrolyte depletion (water loss) by up to 90% over conventional constant voltage methods such as “PWM” & “High Frequency” charge regulators. "


------------------------------------------------------------------------------

I found page 13 of the pv7's manual rather interesting. The question arises, is it true 'on/off' is a better charging algorithm than pwm; or any fixed voltage algorithm for that matter? And the guy in the video says it takes much longer to reach a full charge. Is this in fact true, or does fixed voltage simply waste a lot of power in the form of heat?

http://www.flexcharge.com/custom-1/PV7-PV7D-PV14%20full%202012.pdf

[Salvo]

If you believe that video than you got the wool pulled over your eyes. It's bogus in regards to pwm vs. mppt gains.

BFL can tell you the reason.

Ductape wrote:
After 12 pages, I just want to thank PT for posting a good example of how DC to DC conversion can deliver higher charge currents and wattage than simply connecting the panel to the battery. Lots of good discussion also regarding the language of science and engineering vs. marketing claims.

Almost as lively as the gas vs. diesel debates.

[BFL13]

My ASC solar controller had a reverse blocking diode in it. My Sharp 130w panel had place in the junction box where you could attach one to wire terminals. It also had two by-pass diodes already installed.

The advice in the manuals was not to use two reverse blockers at once, so if the panel had one you should remove it and let the controller's do it (since you can't remove the controller's)

[MrWizard]

jrnymn7 wrote:
Apparently, panels use to have blocking diodes installed, but not anymore?

i think only low power trickle panels ever had reverse diodes

solar makes use of "bypass diodes" for shaded cells or panels
and "blocking diodes" in special applications

there is no reason for blocking diodes on a panel that is to high voltage to be directly connected to a battery bank and has to use a controller
the controller performs that function