Efficiency differences in down conversion.

ILFE
ILFE Solar Expert Posts: 364 ✭✭
vtmaps wrote: »
The reason is that MPPT controllers are more efficient when they perform less down-conversion of voltage. If the distance between your panels and controller is great you will want to configure your panels in series to have a high string voltage, but that high string voltage will be easier to down convert to 24 volts than to 12 volts.

I have seen the above statement, more or less, a number of times on this site.

Can anyone tell me how much of a difference we are talking, for a controller to down convert to 24vdc as apposed to 12vdc? I mean, are we talking 1% to 3% loss due to inefficiency? Or, are we looking closer at 10% to 15%?

Opinions, please?
Paul

Comments

  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Efficiency differences in down conversion.

    https://docs.google.com/spreadsheets/d/1N1-4G5zmTuS-ZP77I4LhZjG2OND41RCpa9-zgkUtsyo/edit?usp=sharing

    This is for 12V system, Panels VOC is ~58-60V
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Efficiency differences in down conversion.
    ILFE wrote: »
    I have seen the above statement, more or less, a number of times on this site.

    Can anyone tell me how much of a difference we are talking, for a controller to down convert to 24vdc as apposed to 12vdc? I mean, are we talking 1% to 3% loss due to inefficiency? Or, are we looking closer at 10% to 15%?

    Opinions, please?

    In my opinion, it's not about efficiency per se, it's about heat production. If your charge controller is handling 1000 watts, for example, one percent is 10 watts. Try putting a 10 watt lightbulb in a metal box the size of your controller, wait an hour, and feel the heat. And remember, that's EXTRA heat on top of what the controller is already producing. The extra heat will shorten the life of the controller.

    If you're only using your controller at a small percentage of its capacity, the efficiency matters less.

    I just looked up the power charts on a Midnite Classic 150 controller. On a 24 volt system it is rated to output 94 amps with an input voltage of 90 volts. Bump up the voltage to 120 volts and it can only output 82 amps. The reason is that the higher voltage causes extra heat production. The limiting factor is heat. btw, it's not a good idea to continuously run a controller at its heat-handling limits.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Efficiency differences in down conversion.
    ILFE wrote: »
    I have seen the above statement, more or less, a number of times on this site.

    Can anyone tell me how much of a difference we are talking, for a controller to down convert to 24vdc as apposed to 12vdc? I mean, are we talking 1% to 3% loss due to inefficiency? Or, are we looking closer at 10% to 15%?

    Opinions, please?

    It's a question of down converting from what input Voltage (Vmp, not Voc), current level, and which controller (yes; some are better than others).
    From 35 Vmp to 12 VDC system vs. 35 Vmp to 24 VDC the difference is minimal on a good controller. As array Vmp rises in relation to system Voltage the controller efficiency decreases. The specifications for this are not generally available for all controllers, but worst-case 150 Vmp to 12 Volt system will still be less than 10% efficiency drop on a good controller.

    As a rule-of-thumb you want to keep array Vmp to no more than 2X what would be 'ideal' for the system Voltage (id est 35 Vmp on a 12 Volt system, 70 on 24, 140 on 48). With a 48 Volt system there usually is not any need to run higher than 'normal' array Vmp and running above 1.5X can easily lead to trouble with the Voc on cold mornings.
  • nsaspook
    nsaspook Solar Expert Posts: 396 ✭✭✭
    Re: Efficiency differences in down conversion.

    I lose on average 10% due to 12vdc system line losses from the panels at full power to the charge controller. It's a lot but the cost of upgrading the system to a higher voltage and using down conversion with a MPPT controller is too expensive. The best solution is to move the DC side of the system to the remote monitor shed side near the panels and send the power as utility voltage to the rest of the system in the house but that would require a major reconfiguration and rewiring that's just not worth the cost for 10% of 500+ watts. Sometimes you just have to let it be.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Efficiency differences in down conversion.

    yes, heat is the biggest problem, but the smaller the spread between input and output it does give a tad more efficiency to it too. i'm thinking it is maybe a percent tops going to a 24v bank from a 12v bank. one would gain from reduced losses in the wires as well so it isn't just the cc. costs for some do prevent going the more efficient routes at times too as i can relate.
  • ILFE
    ILFE Solar Expert Posts: 364 ✭✭
    Re: Efficiency differences in down conversion.

    It's a Kid controller. I was considering going as high as a 48v array, down to a 12v bank. That's why I was asking.

    Thanks, gentlemen, for the replies. I truly appreciate the input.
    Paul
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Efficiency differences in down conversion.
    ILFE wrote: »
    It's a Kid controller. I was considering going as high as a 48v array, down to a 12v bank. That's why I was asking.

    That will work very well.
    The problems arise when you try to down convert much higher voltage to 12 volts. The kid can handle triple digit voltage on the input, but that should be done only for 48 volt batteries.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Efficiency differences in down conversion.
    As a rule-of-thumb you want to keep array Vmp to no more than 2X what would be 'ideal' for the system Voltage (id est 35 Vmp on a 12 Volt system, 70 on 24, 140 on 48 ). With a 48 Volt system there usually is not any need to run higher than 'normal' array Vmp and running above 1.5X can easily lead to trouble with the Voc on cold mornings.
    On a 12V system 35V is ideal Vmp ?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Efficiency differences in down conversion.
    drraptor wrote: »
    On a 12V system 35V is ideal Vmp ?

    No it would be 2X the 'ideal' Vmp. In other words twice the Vmp you would have if using a PWM type controller:

    12 Volt system: 17.5 Vmp
    24 Volt system: 35 Vmp
    48 Volt system: 70 Vmp

    When you go above 2X these numbers on an MPPT controller is where the drop in conversion efficiency can be a problem. In the case of a 48 Volt system the rise in Voc is more likely to present a problem first.
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Efficiency differences in down conversion.
    No it would be 2X the 'ideal' Vmp. In other words twice the Vmp you would have if using a PWM type controller:

    12 Volt system: 17.5 Vmp
    24 Volt system: 35 Vmp
    48 Volt system: 70 Vmp

    Twice the Vmp works OK for 12 and 24 volt systems. There's no need to go 140 volts Vmp on a 48 volt system... using GT panels your optimum string is 3 panels... about 90 Vmp. Four typical GT panels puts you at 120 Vmp where you start getting in trouble with Voc.

    The published power curves for the Classic 150 show that for a 48 volt battery the output limit is 86 amps with an input of 70 volts, and 76 amps with an input of 110 volts.

    The Classic 200 for a 48 volt battery has an output limit of 78 amps with an input of 70 volts, and 65 amps with an input of 140 volts.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Efficiency differences in down conversion.
    vtmaps wrote: »
    Twice the Vmp works OK for 12 and 24 volt systems. There's no need to go 140 volts Vmp on a 48 volt system... using GT panels your optimum string is 3 panels... about 90 Vmp. Four typical GT panels puts you at 120 Vmp where you start getting in trouble with Voc.

    --vtMaps

    This was in fact mentioned in my post. For some reason you chose to remove it when you quoted me:
    When you go above 2X these numbers on an MPPT controller is where the drop in conversion efficiency can be a problem. In the case of a 48 Volt system the rise in Voc is more likely to present a problem first.
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Efficiency differences in down conversion.
    This was in fact mentioned in my post. For some reason you chose to remove it when you quoted me:

    Sorry about that... I thought you were trying to say that on a 48 volt battery, 70 volts was the ideal Vmp for PWM (true) and that the ideal voltage for MPPT was twice that (not true).

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Efficiency differences in down conversion.
    vtmaps wrote: »
    Sorry about that... I thought you were trying to say that on a 48 volt battery, 70 volts was the ideal Vmp for PWM (true) and that the ideal voltage for MPPT was twice that (not true).

    --vtMaps

    No, it's more that you don't want to go 2X the "PWM Vmp" on any system if you can avoid it. Earlier (or perhaps in another thread as this question does arise often) I mentioned 1.5X as a practical limit for 48 Volt systems due to the Voc issues.

    If anyone is wondering the difference is like this:
    Two 35 Vmp panels in series for a 48 Volt system gives an array Vmp of 70 and an array Voc of 84 (estimated Voltages; not specific or absolute).
    If you were to double that the array Vmp would then be 140, and most charge controllers have a 150 Volt input limit. Add Voltage increase due to cold temperatures and the Vmp would be capable of exceeding the maximum. Meanwhile the Voc would jump to 168 which is what would shut down most and possibly damage some charge controllers.