Over-dimensioning: figuring the optimum point

I'm searching for the most bang for the buck, and would like to over-dimension to fully utilize each controller.  40a controllers are considerably less expensive than 60a or 80a, and the premise is that it would be cheaper to fully utilize a number of 40a controllers.  So let's look at the case of a single 40a controller and the question of how many watts of pv to optimally send it's way.

For specific numbers, I'm looking at the EPever Tracer 4210a that is rated 520w@12v and 1040w@24v and can handle 300% of rated watts input or max 92Voc before short circuit (only 150% of rated power if reversed polarity).  This means it will charge the battery at mpp when the pv is outputting under 520w actual, and will charge the battery at exactly 520w if the pv is actually producing more than this, clipping the extra production.  The advantage of over-dimensioning is to reach this max output sooner and stay there longer, with the disadvantage being the power lost to clipping.

Most people will try to stay under the 520w rating and install 500w STC of pv, which only actually produces 400w in actual conditions, underutilizing the controller they bought.  It seems at the very least I could run 690w STC to put an actual 520w NOCT through the controller without clipping.   Perhaps I would be better off at 520x1.5=780w STC which is about 585w actual and will be clipped at least in summer and/or near noon.  Does that seem right?  Theoretically, I could use 1560w pv but the clipping would be huge and I'm sure the value point is somewhere lower.  But how do I find that optimum point theoretically?  Anybody know of some data on this or good resources for more information?

Comments

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    If the "EPever Tracer 4210a" is built beefy and has enough cooling in the room, sure try it out.    But the 92v max seems really odd. Good controllers are more often near 150Voc
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • SupraLance
    SupraLance Registered Users Posts: 21 ✭✭
    I will go with the EPever 4215bn instead of the 4210a if the pv I'm using will benefit enough from the higher max volts to be worth the extra $50.  My big question, though, is with how many watts pv should I overpanel a 40a 520w controller?  I know it's hard to tell without somebody trying it and recording results, but if you were trying to get the most watts per dollar spent using a 40a 520w mppt controller, how much pv would you pair it with?  I'm new to all this and hoping to draw from the experience of others, although I'm coming at it from a different direction.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    There are a lot of factors to this, and many are location specific. For example, running a controller flat out will shorten its life, but more so in a hot, poorly ventilated enclosure. Panel output can be higher than STC in a cold, high altitude location, or much lower in a hazy, sea-level location.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • SupraLance
    SupraLance Registered Users Posts: 21 ✭✭
    Good points.  I'm in Southern Missouri, 1200' above sea level, 37 degrees north latitude.   Would you agree that I'm unlikely to ever see more than 80% of STC, or will I actually hit STC for 15 minutes near noon with the panel tilted right at the sun on a clear day?  If I understand this correctly, When they draw the curves I've seen for over-dimensioning with the top of the curve clipped, the peak of the curve actually will be about 75-80% STC...?  Is that right?

    I was reluctant to run a fan on the controller because I'm trying to limit loads on the system, but if doing so allows me to generate more power than the fan uses, then I will do so.  So your saying that the more pv I over-dimension the controller with, the more fan I need to have on it?  That makes good sense.  So with a standard inexpensive fan like most people would use on a controller, how much pv would you use?  How about if there was no cooling?  Would that make over-dimensioning at all a bad idea?
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    The amount to over panel should be calculated using, location, panel orientation, tilt, minimum and maximum temperature, all of which factor into the equation, often seasonal adjustments in tilt can offset losses which a fixed array would suffer. The only way to approach STC figures would be in extreme cold where the sun is unable to heat the panels above ambient fridge temperature ,unlikely in your location, least of all at noon properly aimed.
    Pushing equipment to run a it's maximum thresholds is, in my opinion, a mistake which will rear it ugly head sooner rather than later, for this reason higher end manufacturers have string calculators to assist in design, following their recommendations ensures operating within safe parameters according to it limitations.
    Operating at 12V restricts the controller, is there a reason why 24V is not an option?

    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • littleharbor2
    littleharbor2 Solar Expert Posts: 2,039 ✭✭✭✭✭
    As to the external fan, You can find 12 volt computer fans online for as little as a couple bucks. Small quiet and very efficient.

    2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old  but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric,  460 Ah. 24 volt LiFePo4 battery bank. Plenty of Baja Sea of Cortez sunshine.

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    In your location you may see ~STC values for a short time, for example, on a cool/cold spring day when the sun comes out from a cloud. The panels would soon warm though, and power would drop. I would expect ~75% to be more typical. The "curve" of output would be clipped to the extent potential output exceeds controller capacity. This may or may not be the 40a rating, as the controller may throttle output based on internal/FET/PCB temperature.

    I like the idea of overpanelling (within reason) relative to battery capacity, but not so much relative to controller capacity. Adding ventilation, effective heat sinking, and installation in a cool spot can help, but I suspect the loss of lifespan is still likely to offset the savings gained by not adding another controller.

    Assuming a reasonable, typical installation (ie not in a hot enclosure) I would stay under whatever manufacturer suggests as max wattage. Midnite is 120% of STC, for example. With extra cooling measures, you might be able to push it another 20% or so.

    As Mcgivor notes, you would get a better watts per controller dollar by increasing battery voltage. 40a@12v=480w. 40a@24v=960w. 40a@48v=1920w.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Without regard to the amount of over paneling, if you controller is designed to handle over paneling, weather your controller is maxed out will have more to do with the system uses. Since they need to be wasteful, and batteries will limit their input when nearing full, controllers are rarely running at full capacity.

    A well designed system will stay in the upper 20% of battery capacity on most days, only after several cloudy days followed by full sun, will you see a potential for high capacity charging. Even then the system will ramp up as the sun moves toward solar noon and once it reaches about 80% of the battery capacity, the battery will except less current.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • SupraLance
    SupraLance Registered Users Posts: 21 ✭✭
    edited February 2018 #10

    I'm was looking at 12v initially because I have two 230ah 6v batteries in use at 12v, and know that it's bad to mix old and new batteries. But the eventual plan is 24v, so you're right that it would be a good idea to design for the planned 230ah@24v battery now.

    So the question is, “With how many panels should I over-dimension a 40a controller that clips at 1040w@24v, to generate the highest watts per dollar spent, assuming enough electrical demand to use it all?”

    Comments have led me to switch the plan to the EPever 4215bn, with higher allowed Voc and better cooling, and to plan to add a 12v computer fan. EPever suggests max wattage for this controller as 3120w, which would result in an enormous amount of clipping to stay under the rated 1040w output, and would undoubtedly produce a lot of heat and shorten the controller life. I obviously don't want to over-dimension by that much.

    So Estragon mentioned Midnight suggests 120%, which would be 1250w STC. Am I right to think this typically means actual production under real world conditions of around 900w (75%), and am I right to figure this would produce around 900x5=4500whr/day in a location with 5hrs/day solar insolation?

    If I can bump up to 140% with the fan, this would be 1450w STC, around 1100w actual production, clipped at 1040w. So does it sound right to say this would produce 1040x5=5200whr/day?

    I think theoretically (although it may not be smart to do), if I over-dimensioned further, I never increase above 1040w but I do increase the number of hours/day I sit there, right? So if I over-dimension to 1600w, do you think that I could get 1040x6=6240whr/day actual production, if the 40a controller is kept cool enough?

    Does anybody know how I can go about figuring out how many hours/day the controller will see it's max 1040w at different amounts of over-panelling? I suspect this would require graphing production points throughout the day and then clipping the graph at 1040w and calculating the area under…


  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    IMHO, there are too many variables that can't be controlled for to answer your question for a smallish scale application. At utility scale, you could test for MTBF under various conditions etc, with usable sample sizes, have reasonably predictable loads with statistically meaningful standard daily and seasonal variances, etc.

    You could try asking the engineers who designed the controller if it would stand up to full current for X hours per day, or if it would safely clip Y current, but I doubt you'll get an answer you can bank on. In theory, you could massively overpanel, and run 60a (minus auto-derating, if any) through the controller for as long as there's enough light and pv to produce it, and a load to demand it.

    In our little systems, we just have to give it our best shot. In theory, theory and practice are the same. In practice, they rarely are.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter