Forum Discussion

BFL13's avatar
BFL13
Explorer II
May 07, 2015

12v MPPT and 24v MPPT vs 24v PWM Test

Got a chance to try out comparing amps to the battery using MPPT vs PWM on a 24v battery bank and comparing those with the usual MPPT on 12v with my 230w panel.

The bank is four 6s, so it was easy to change from 12v to 24v and back.

(Was able to swap them all around with insolation etc as near the same over the time it took as you can do in these tests, so the results are useful but not lab perfect of course. )

(The actual amps are not necessarily the max amps for the day with the sun higher later on, but the amps can be compared with each other within the time frame of the tests.)

MPPT controller is the 20a Eco-Worthy and the PWM is the Solar30. Battery bank "resting" voltage before the tests was 12.4 volts, low enough to accept all the amps.

Voltage and amps results as shown on the controller's display, confirmed with multi-meter.

12v MPPT- 12.7v/14.97 amps (190w)
24v MPPT- 25.4v/7.95 amps (203w) so that would be 15.9a at 12v
24v PWM- 25.5v/7.99 amps (204w) (15.98a at 12v)

The MPPT buck converter is more efficient doing 24-24 than 24-12 as we know, and this test seems to show that.

I expected the PWM on 24v to do about the same as it would with 12v for getting amps, which it did here.

The ambient and panel temps are moderate today, so the MPPT was not suffering from high panel temperature so much like it does in July/August here. I wanted to see this summer if I could beat that problem by using the Solar30 with the battery bank in 24v mode.

I have a spare 12v battery to leave the rig on during the day while the main bank is disconnected at 24v. However the mechanics of swapping things around may make it all too annoying.

The lower amps of the MPPT in July are still plenty for our needs, so there is not much reason to do all that work. I just hate not getting "all my amps" so we'll see if it looks worth it this summer when I do another set of tests with the panel being hot.

With the hot panel I expect the 24v MPPT to not do as many amps as the PWM at 24v, which just uses panel amps and they are not reduced by the panel getting hot. We'll see in July!
  • CA_Traveler's avatar
    CA_Traveler
    Explorer III
    May 12, 2015
    It does rise from 12.1 to 12.4 but the scale doesn't make that very visible. And later that morning it rose to 13.4V and that data isn't shown. With 440AH batteries it took a reasonable house load to pull the battery voltage down with the fully charged batteries. So most of the amps are supplying the house load. The data represents controller output but not much battery charging.

    There were shadows from very distance trees until about 8:00 which was caused by the very low sun altitude. With 9 bypass diodes you can see the effect of the distant leafy shadows with largly 10x volt changes in Vpanel.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    May 12, 2015
    o.k., help me out with this graph.

    I don't understand how Abat can rise while Vbat holds steady?

    (I'm assuming Abat is amps to battery)
  • BFL13's avatar
    BFL13
    Explorer II
    May 12, 2015
    CA Traveler wrote:
    Agree but some caution is in order. BFL13 used 40' of various larger cables and some were aluminum/copper and he measured 29V on a prior test. The PWM results are appreciated and certainly adding to our knowledge but drawing generalized conclusions may/may not be appropriate.


    My panel to controller wiring zoo has a line loss of 1% when amps are in the 7 to 8 amp range. 29.3v at panel and 29.0v at controller is typical. That's with the 24v panel.

    With my PWM (which has to be to a 24v battery with the 24v panel), the amps on the path are the same at near the panel's Isc of 8.3

    But if you have 12v panels to the same wattage and use PWM, now you need fatter wire on the panel to controller path where now you will have about twice the amps and the voltage will be about half.

    I tried it on a voltage drop calculator. With #8 at 40ft distance (wire pair) 24v, 8 amps drop is 1.67% but change that to 12v, 16 amps and drop is 6.67%

    Note the higher percentage is also due to the drop being a bigger proportion of the lower voltage.

    With #4 instead of #8 and the 12v,16 amps the drop is 2.67%

    So a key to the whole business is that with the higher 24v you can have the same actual voltage drop but it is a smaller percentage.

    Sometimes it is the absolute value that matters not the percentage. On the controller to battery path you want the controller to "see" the actual battery voltage for hitting the set point. If you want it to start controlling when battery hits 14.6 and your drop is 0.4, then it will start controlling when the battery is only at 14.2.

    You don't care how much that 0.4 is in percentage, you want 14.6 instead of 14.2

    So you have to be careful with percentages--always ask, eg, "10% of what?"
  • CA_Traveler's avatar
    CA_Traveler
    Explorer III
    May 12, 2015
    Agree but some caution is in order. BFL13 used 40' of various larger cables and some were aluminum/copper and he measured 29V on a prior test. The PWM results are appreciated and certainly adding to our knowledge but drawing generalized conclusions may/may not be appropriate.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    May 12, 2015
    awesome, thanxx!

    I'm just looking at the feasibility of those 'low voltage' 24v panels some have. Yes, at 12.7v, that would not get you far, but that's just a jumping off point. As insolation increases with sun rise, the process would continue. My point is, it seems likely 30+v are not required to charge a 24v bank, using pwm... 30v should suffice... and that is only when Vbatt has risen to Vabs. And with solar generally being at a low charge rate, that should not occur until well into the charge cycle.
  • CA_Traveler's avatar
    CA_Traveler
    Explorer III
    May 12, 2015
    jrnymn7 wrote:
    CA Traveler wrote:
    jrnymn7 wrote:
    CA T, thanks. If Vbatt was say 12.2, at what panel voltage would charging begin?
    The short answer is Vp=12.7, Vbat=12.1 and Abat=1.2


    So, your 24v panel only had to be at 12.7v to initiate charging? Or am I misinterpreting that?
    The Abat represents output from the controller and actually started at less that the 12.7V panel voltage. The amps would be supplying house loads since the battery voltage is to low for any real charging plus amps are also low. I kept the battery voltage low with house loads since I wanted to see how the panels and controller would work. I'm not a big fan of discharging my house batteries say overnight just to run a test and hence use house loads.

    I'll see about posting some graphs.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    May 12, 2015
    CA Traveler wrote:
    jrnymn7 wrote:
    CA T, thanks. If Vbatt was say 12.2, at what panel voltage would charging begin?
    The short answer is Vp=12.7, Vbat=12.1 and Abat=1.2

    The long answer is I don't really know. As I recall the bats were fully charged but I drug down the voltage with house loads as I was interested in panel and controller outputs and what to expect. Only a few days before was the solar install completed so I had my Christmas toy to play with. :B


    So, your 24v panel only had to be at 12.7v to initiate charging? Or am I misinterpreting that?
  • BFL13's avatar
    BFL13
    Explorer II
    May 12, 2015
    full_mosey wrote:
    pianotuna wrote:
    Hi John,

    That would be great.

    What is the voltage on your panel?


    I will be able to start the test next week, around May 20.

    Here are the panel specs:

    85W, 18.3Vmp, 4.69Imp, 22.7Voc, 5.12Isc.
    Two in series, 25ft round trip 12Ga THNN.

    I will start a new topic when I begin.

    Today at 9AM I did read 42.9V at the controller and 66F for a panel.

    HTH;
    John


    Looking forward to your tests. The ones you did on inverter draw/voltage drop were good info too.

About Technical Issues

Having RV issues? Connect with others who have been in your shoes.24,201 PostsLatest Activity: Feb 19, 2025