Help needed for Solar direct driving (DC) water heating element. Xantrex C40?

Hi All,

Im currently designing an approximately 1kw (4x230w) Solar PV water heating system. Im aware that directly hooking an element to solar is a very inefficient method due to the solar cells not getting to there max power (VI curve) where the sun is at a low angle, or cloud cover etc (output load is not variable). What I am considering is a way to drive the load of the heating element, to match the production of the solar array. PWM seems ideal for this.
Is the Xantrex C40 set as a diversion mode usable for this purpose? I propose I would set the diversion voltage set points to around 60v (vmp of 2 panels in series) and the controller would drive the element proportionally to the power generated. Im concerned however that there may not be enough capacitance in the input side of the controller? Could be fixed with a large capacitor?
Any help would be greatly appreciated!
Kind regards, Mike.

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    No--PWM is not going to "match" the IV curve of the solar panel against the IV curve of the heating element. It takes a MPPT type controller to do that.

    A standard solar charge controller is (almost always) going to need a battery of some sort for the controller to start up and "keep its brains".

    In general, solar electric panels are going to be very inefficient and relatively expensive when compared to standard thermal solar panels to collect heat. There are some "issues" with solar thermal (plumbing, water leaks, air locks, freezing weather, etc.) that make solar thermal collectors a pain to implement--But in general, solar thermal is still the better solution.

    I am guessing that you are in Auckland NZ area? Any solar power system is highly dependent on where the system is located... No sun, no solar energy collect (marine layer, deep winter for those closer to the poles, etc.).

    Can you tell us a little more about your power needs? Why you want to use solar electric vs solar thermal?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • nz.mike.christiansen
    nz.mike.christiansen Registered Users Posts: 3
    Thanks very much for your reply Bill,

    I am indeed in NZ, just slightly north of Wellington.

    ​The reason we are looking at doing PV hot water is mostly around compliance costs associated with other forms of energy production. Our water heating is currently resistive electric only. A evacuated tube or other form of "wet" heating requires both a building concert and a plummer to install here. Our area is also very frost prone. Solar PV is beginning to look quite cost efficient as panels are available at approx $1/w, and we have pleanty of north facing roof space.

    A grid connected solar system was always what we intended to install. Unfortunately our power company are not obliged to pay for self generation at any fair rate. Currently power costs approx $0.30/kwh to import, and we may be paid $0.09/kwh to export. This comes after additional connection fees and additional line fees. Installation of a grid connected PV system also requires both building consent and electrical inspection adding significant cost. The fact that we are only paid 1/3 for power exported is the real killer in the math.

    Currently approximately 40% of our power is spent on water heating. My theory is that this component of power is the easiest / most financially rewarding part of our energy consumption to self generate. Energy can be stored, displacing units of power saving $0.30/kwh and without compliance costs associated with other systems. Works of this nature can be carried out without consent or inspection if ripple free DC voltage is kept under 120v. Certain conditions must also be met around HW cylinder systems. (an inline thermostat is a requirement on a non-vented cylinder)

    I guess for the costs involved, a small performance hit might be worthwhile taking to avoid costly MPPT or inverters as this investment could be put into additional solar capacity. Matching IV of the solar array to a size appropriate element might be the best / most simple system with the fastest pay back period. I am yet to find a cheap MPPT controller that is not intended to charge batteries or as part of a grid connected inverter suitable for hot water heating.

    Again, thanks Bill for taking the time to help me out. I really appreciate any valuable assistance.

    Regards, Mike
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Here is a thread on connecting solar panels directly to resistance water heating:

    http://forum.solar-electric.com/forum/solar-electric-power-wind-power-balance-of-system/advanced-solar-electric-technical-forum/24323-short-circuit-thermostat

    Unfortunately, one of the posters did not really follow my reasoning and results. In "poor/winter" weather, direct connect solar panels to resistance heater efficiency is quite poor. But--Look at the costs (of solar vs electric bill)--Perhaps it makes sense for you.

    Other options:
    • Simply run solar water heater ~9 months of the year--Will reduce your costs by 75% for not too much money (good do it yourself type project).
    • People do solar water heating even in freezing climates. I bit more work, but can be done. Solar Shed and other Solar Thermal Links
    • Use heat pump type water heater (lots of options out there--Air to Water, Water to Water, retrofit, etc.)--Roughly 2-3x more efficient vs resistance heaters--But does need relatively "warm" air to be efficient (over ~55F or so for many).
    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Ethan Brush
    Ethan Brush Solar Expert Posts: 235 ✭✭
    Hi All,

    Im currently designing an approximately 1kw (4x230w) Solar PV water heating system. Im aware that directly hooking an element to solar is a very inefficient method due to the solar cells not getting to there max power (VI curve) where the sun is at a low angle, or cloud cover etc (output load is not variable). What I am considering is a way to drive the load of the heating element, to match the production of the solar array. PWM seems ideal for this.
    Is the Xantrex C40 set as a diversion mode usable for this purpose? I propose I would set the diversion voltage set points to around 60v (vmp of 2 panels in series) and the controller would drive the element proportionally to the power generated. Im concerned however that there may not be enough capacitance in the input side of the controller? Could be fixed with a large capacitor?
    Any help would be greatly appreciated!
    Kind regards, Mike.

    I have contemplated this a bit and have some thought and ideas (mostly not too different from Bill's)

    1. One idea is just to direct connect the elements to the array and "fine tune" the array and heating element(s) for a good full power match under ~average~ full sun conditions. Note that you power harvest will drop off very significantly when the current drop from low sun, clouds, etc. As an example (and this is quick and crude - of course 12V panels put out about 17V and that is what the wattage is based on, etc but the point is to illustrate how bad things get when the match is lost), say you take a 12V, 60A element. That is .2 Ohms. Say you connect 6 panels in parallel that are 12V, 120 W. That is 60 amps. So in decent full sun, you could harvest the full 720 watts. Now say irradiance is such that you have 1/4 the current or 15 amps total now. That should be 180 watts, but you cant use it since you are stuck with the .2 ohm resistance. So your elements drag the voltage down to 15A X .2 Ohm = 3 volts ! That is 45 watts ouch! OTOH though, perhaps full sun is where you get most of your results anyway, so who cares about the cloudy periods? I am not sure what the percent loss would be.

    2. What about getting a MPPT controller and hooking it up to a super small 12 V battery, and diverting off of that with the diversion controller?
  • nz.mike.christiansen
    nz.mike.christiansen Registered Users Posts: 3
    Thanks everyone for your consideration.

    I do like the idea of Ethan, charging a small battery, altho I was thinking of possibly using a large capacitor bank for an accumulator as they may be lower maintenance. I guess one would need to ensure the voltage from the bank was not to drop too low over night as the componentry may need a power supply for logic etc.

    Im planning on proceeding with the heating as simply direct connect initially and monitoring the system to estimate what benefits may be achieved with additional controllers etc. This may be quite effective over the summer months.

    Thanks again Bill and Ethan.
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Rapid PWM plus some capacitance on the source side can act as a current controller and with the right controls, it can be used to track MPPT.  Ie, the design of a MPPT controller that will only drive a resistance load is somewhat simpler than one that requires  a specific and continuous output voltage.

    I am available for custom hardware/firmware development

  • RTM
    RTM Registered Users Posts: 1
    Hi Mike, did you resolve this ?   I am grappling with similar, have a solarHart hot water heating which is on its last legs. Wondering whether to replace this (reviews on replacement not great, current unit has been great) or use solar panels more or less exactly as you were trying to do. Don't really want to go to expense of being able to feed the power back into the grid, as they don't want our power anyway. You could email me direct if you are still connected to this forum....or I will keep checking back here for a week or two. All the best
    RTM  (jchristrtm@gmail.com)
  • NANOcontrol
    NANOcontrol Registered Users Posts: 260 ✭✭✭
    The PV heating and control is in its infancy.  If you are on the grid, it is easiest to grid tie.  There are systems that sense when you are feeding back to the grid and that forces to  energy to go proportionally to the water heater.  Built in adaptive timers prevent/delay the heating of water till the peak times where water is stored at extreme temperatures.  This is all expensive and complicated for people who never figured out how to make their VCR stop flashing 12:00.

    I heat water off grid and it is cost effective compared to direct solar that is a horrible pain to use.  I sense that Diversion controllers built into charge controllers are not that effective.  Why drag extra current through a controller when you really want the highest voltage going to the heater element to reduce wire losses. As electric water heaters have two elements, it is easy to just use the lower for isolated PV heating.  I do this with a capacitor bank and PWM the heaters and it is quite efficient.  I will be going into production of a board that will do this next year as this market is not well served.  This can be added on to any existing charge controller.
  • zeppole
    zeppole Registered Users Posts: 4
    I just completed connecting solar panels directly to my water heater and the results are encouraging.  I am using 8 Renesola 300 watt 72 cell panels, in 2 strings four panels in each string.  I got the panels very cheaply (about .28 watt including shipping) so my total cost is less than $800 including control circuit, etc.  Just using the original 4500 watt element (12.7 Ohms reistance) which is not optimal -- 9 to 10 ohm reistance element should yield better results.  I am using a separate thermostat and a Omron 100 amp relay as a control circuit. 

    In this month of December I have routinely hit 1.6KW delivery to the water heater, with daily production of 5 to 7 KWh.  Sunny days have raised my well water from 52 degrees to 135 degrees.  Given the short days and the low sun aspect I'm guessing spring and summer production should be considerably higher.  Cloudy days are another story altogether -- very little production.  I am using a Takagi propane tankless water heater downstream of the electric water heater which works very well.

    There is very little in the marketplace for MPPT controllers for direct connection -- one in Eastern Europe called Kerberos:  http://www.solar-kerberos.cz/en/what-is-kerberos.

    I'm not really convinced that an MPPT controller in this application is warranted given the expense and added system complexity -- particularly when solar panels keep coming down in cost.  However, I am intrigued by the post of NANOcontrol above regarding the use of capacitors and PWM.

    Hopefully, a cost effective solution can be found to optimize direct pv connection to water heaters (as well as other pure resistive loads).
  • Spongology
    Spongology Registered Users, Users Awaiting Email Confirmation Posts: 3
    Running a Solar Water Heater can reduce your problem.