MPPT input voltage comparison

[native pepper]

decided not to bother

[bagmaker]

C'mon guys we can do better than this. We all know MPPT is supposed to be better, we all have our favourites too. In general what we have is good.
But something isnt adding up for me, I need help to explain it and hope the education can better the understanding for everyone. Its not a pissing contest, its a scientific on-line school room. There can be no losses if we all respect what we each put in.

OK, some more tinkering today
Cloudy, rain, humid, 33.

I dis-connected panel input into the parrellel controller to check if it was somehow affecting the series controller. No change to series controller output amperage when charging by itself.
i re-connected it and isolated BOTH units from the battery and re-started them. They both went through a start-up regime taking perhaps 15 seconds before outputting current. The both ran a little high before settling to a comfortable output -the same 2-3 amp better on the parrellel controller came up straight away.
There is evidence of a "pulsing" as has been described -like the controllers are looking for the best output available.

So the upshot is that the series controller is SLIGHTLY more effective when bulk charging and the parrellel controller is WAY more effective during all other charge phases.
Not such a biggy on LiFePo4 batts but pretty critical on LA batts, especially on larger stuff (mine is 1120 A/H)

Can I try anything else to test?

Again, its not about mine being better than yours or anything, just interesting physics that does'nt match the known hype.

[BruceS]

Well I'll have a weeny say too?
There are a lot of differing things that can and do effect which way the user or installer goes.
In my particular case I bought the 'wrong' solar panels to suit the solar controller that I preferred to use.
That choice of panel made it essential that I used the MPPT version of the similar controller. (same brand and capacity)
I wanted to go 48V DC for my install.
My 24V solar panels connected 2 in series were too much for the PWM version but were a nice 'fit' for the MPPT one.
My choice had nothing to do with which one produced X amount more than another choice.
The very small difference can easily be compensated by an extra panel or two. (in my opinion)
I think it's very important for anyone starting out to examine very closely the specs of ALL items prior to buying any of them.

[Barboots]

Can I try anything else to test?

I'd have a play with:
- shading scenarios... both minor and quite significant.
- early/late in day performance.

Cheers, Steve

[Busman]

Bruce, do a Pauline please.

What do you mean the "wrong" panels ? In what way ?

Thanks
William

[BruceS]

I will explain in due course William but it won't be for a little while I'm afraid.
I'm busier than a one armed butcher in a bakery..........

[Mrcoolabah1au]

Wow that new

[Busman]

You must have been in Billinudgel 50 years ago Bruce !

[T1 Terry]

C'mon guys we can do better than this. We all know MPPT is supposed to be better, we all have our favourites too. In general what we have is good.
But something isnt adding up for me, I need help to explain it and hope the education can better the understanding for everyone. Its not a pissing contest, its a scientific on-line school room. There can be no losses if we all respect what we each put in.

OK, some more tinkering today
Cloudy, rain, humid, 33.

I dis-connected panel input into the parrellel controller to check if it was somehow affecting the series controller. No change to series controller output amperage when charging by itself.
i re-connected it and isolated BOTH units from the battery and re-started them. They both went through a start-up regime taking perhaps 15 seconds before outputting current. The both ran a little high before settling to a comfortable output -the same 2-3 amp better on the parrellel controller came up straight away.
There is evidence of a "pulsing" as has been described -like the controllers are looking for the best output available.

So the upshot is that the series controller is SLIGHTLY more effective when bulk charging and the parrellel controller is WAY more effective during all other charge phases.
Not such a biggy on LiFePo4 batts but pretty critical on LA batts, especially on larger stuff (mine is 1120 A/H)

Can I try anything else to test?

Again, its not about mine being better than yours or anything, just interesting physics that does'nt match the known hype.

Not trying to restart the brawl here, but these are graphs from a major player in the MPPT market and they clearly show that the higher the voltage in compared to volage out, the greater the inefficiency and also the 98% efficiency is a "cherry picked" figure, not the average. I love the way they blur the information by changing the measurements used, a 60 amp controller but the bottom line of the graph is in watts, so unless you already know a reasonable amount the graph still doesn't tell you a lot. For clarification, the test claims to be conducted at a controlled input and a controlled battery voltage of 12.8v. So the 300w the 98% efficiency was achieve required a 23 amp input, the full scale of the graph is 750w or 58.5 amps input, so still not at full capacity yet the output drops quite a bit.

Morningstar 60 amp MPPT.jpg

This is obviously a lab test as it would be impossible to maintain each input voltage or current (amps) or control the battery voltage to a flat 12.8v or maintain the temp at 25*C. Real world figures may not even be as good as these figures, they would have used the very best results they could get when publishing a chart.

T1 Terry

[Firegirl54]

Hi Guys,
I hope the following helps rather than hinders understanding! And my apologies for this ridiculously long post.....

I think the title of this thread is a wee bit misleading. We have not been comparing MPPT and PWM, we have simply been comparing MPPTs at differing voltage inputs.

Bagmaker has been comparing two different voltage inputs from 2 arrays, each with exactly the same power(watts) feeding into 2 identical MPPT controllers. Let’s assume each panel can produce 10 amps in the given conditions. One MPPT then is taking 54V (3 x 18V from series connected panels), and 10 amps, and converting that to say 13 volts at battery. The other is taking 18V and 30 amps (3 x 10 amps from each of the paralleled panels) and also converting to 13 volts at battery.

The power input (V x A = 540W in this example) to the MPPT controllers from each array is identical. The output from the controllers, however, will vary roughly in line with the graph the Terry has posted above. MPPTs do have an efficiency “penalty” which is bigger for a larger voltage drop. From the graph, the MPPT with the 18V input might be around 97% efficient, giving a power output (ie input to battery) of 0.97 x 540 = 524W, or 40.3A at 13V. But the 54V one will be less efficient, maybe 94%, giving a power output of 0.94 x 540 = 508W, or only 39A. Pretty much as Bagmaker has demonstrated for us in his experiments, and exactly as the science predicts. Of course, as Terry has pointed out these efficiency values are derived under lab conditions, and may well be quite different in the real world.

Now, coming to MPPT vs PWM.....hmmm.

Lets start in Fairyland where the standard conditions for solar panel output apply....panels at 25deg, max power output at 18V, max current at close to short-circuit current, lets say 10 amps for this theoretical 180W panel, battery at 12V in “bulk” charge and will accept whatever we can supply. Our MPPT controller (assume it is 95% efficient) will output 18 x 10 x .95 W = 171W, or 171/12 = 14.25A. Our PWM controller, although it is closer to 100% efficient, can only hold the panel voltage at the battery voltage, 12V. The panel can still only output 10A. Thus, even with the efficiency penalty, the MPPT comes out 2.25A ahead. But that was fairyland!

Lets go back to the real world. The sun is not at 90deg, there might be some cloud about, its bloody hot though, no wind and panel temps around 70 deg. Now the short circuit current for our panel reads say 5A, its max power point is now at a voltage of just 13.5V. The battery is at a more respectable 12.8V. The efficiency of our MPPT has dropped to say 90%. Do the same maths as above. The PWM controller now wins by about 0.3 A. Really minimal difference. ...so take your pick, depending on other features (eg adjustable outputs, price, robustness etc etc,.

Back to Fairyland again (well not really) for anyone who may be interested. We do use a Victron 100/30 MPPT, chosen mainly for its adjustability for our Lithiums. Its not ideal but that is another story....... The MPPT function is particularly useful to us in one area though...We can feed the output of a cheap 20 amp 12 to 19V DC converter into it to charge when driving. I have measured 19.1V and 15A going into the Victron, and 20A at 13.9V coming out. (I have set max current output conservatively at 20A so as not to overload the DC converter or the car wiring...see below) The measured efficiency of the Victron controller was then (20 x 13.9)/(15 x 19.1) x 100 = 97%. Interestingly the el-cheapo DC converter was even more efficient. Measured 22.8A at 12.7 V going in, so (15 x 19.1)/(22.8 x 12.7) x 100 = 98.9%. Voltage at car battery was 13.8V and there was another 0.26v drop between the Victron controller and lithium battery, so overall efficiency between the car alternator/battery and the lithium house battery was (20 x 13.64)/(24 x 13.8) x 100 = 82%, with by far the largest component of loss in the Toyota-supplied car wiring to the Andersen plug (only 8mmsq).

Cheers, Leslie