diy solar

diy solar

Solar dump load to heat water (off grid system) Victron

ApproximatelySean

New Member
Joined
Jan 21, 2021
Messages
30
So I have a common problem but im hoping things have inproved since none of the solution i've found so far are optimal, I'd like to dump excess solar PV 'Not battery power!' into a water heater, preferably id like the dump whatever is available be it 500w 1kw or 3kw so PWM seems like a good idea for the dump load?
Originally i was looking at power diverters like the Eddi, I cant remember if they pwm but they dont seem to work off gird anyway,
'Or could i fool the eddi into thinking it on grid by attaching a heating element dump somewhere else?'
In any case id appreciate any recommendations, i'm running victron, smart shunt, smart solar, multiplus, Lithium 5kwh server rack, im willing to go the arduino route if necessary, however since the victron smart shunt is 'Smart' surly there is a way to program that after the battery has reached 100% soc that a pwm dump load will start but not draw any current from the battery? seems like a perfect use for the smart shunt..
 
If I understand how PWM works it would also be a dump load without the benefit of being able to store heat because of PWMs inefficiency.
 
I don't think victron is that smart, just better than nothing. It can tell you when there is excess, just not how much. You need a proportional control which looks at array voltage and the victron could just verify it is OK with that diversion. Proportional control is fairly easy with arduino dumping directly from the array efficiently. PWM is not equated with inefficiency when coupled with capacitor storage. It can be less than a 5% loss. I've done arduino hot water systems.
 
It's possible to "trick" on-grid diverters like Eddi and Immersun. You basically need to put the "grid" CT around 2 opposite circuit branches (PV and load) in opposite direction that will measure the excess PV power to be diverted to the water heater. Note, the water heater circuit needs to be excluded from the CT. However, the problem with on-grid diverters is that they have pretty slow response so your Victron could potentially see very high backfeed power for a while (seconds range) when large loads turn off. That maybe fine if your Victron and battery bank can handle it.

Faster high power off-grid DC diverters are available but not cheap and they typically work off battery bank voltage and don't use PWM.
 
I don't think victron is that smart, just better than nothing. It can tell you when there is excess, just not how much. You need a proportional control which looks at array voltage and the victron could just verify it is OK with that diversion. Proportional control is fairly easy with arduino dumping directly from the array efficiently. PWM is not equated with inefficiency when coupled with capacitor storage. It can be less than a 5% loss. I've done arduino hot water systems.
So when your DC dumping are you cheking for high panel voltage? and did you ever successfully work in pwm? I still dont see why an arduino could not ramp up the pwm duty untill the smart shunt reports a negitive draw on the battery or even an in curent that is marginaly lower then float..
 
On the topic of PWM, can a high enough feqency be used to avoid flicker? and what would the long term effects of pwm powering say a 2.5kw load on a 3kwh victorn multiplus? is that kind of iregular load a problem ?
 
Would you be better off changing the MPPT for a morningstar Tristars PWM , using it on load division , almost like a wind turbine dump load


Just an idea
 
Victron 150/60 MPPT has relay output that can be set up with "battery voltage high" rule to turn on at certain voltage and off at another with a time delay to avoid relay chatter. You can set it just below absorption voltage to have it come on before controller starts backing off charging current.
 
Victron 150/60 MPPT has relay output that can be set up with "battery voltage high" rule to turn on at certain voltage and off at another with a time delay to avoid relay chatter. You can set it just below absorption voltage to have it come on before controller starts backing off charging current.
I have been looking at that, here's the full options with my 250 60, I'll go over it again when back at base as I can't make this demo do anything, still without pwm I'll either have to choose a very underated heater (and lose most of my opertune load) or risk bouncing my battery every time a cloud passes over
 

Attachments

  • Screenshot_20230517_200425.jpg
    Screenshot_20230517_200425.jpg
    44 KB · Views: 10
without pwm I'll either have to choose a very underated heater (and lose most of my opertune load) or risk bouncing my battery every time a cloud passes over
With standard 4500w 240v heating element used in 50 gal water heaters you can run it on 120V from your inverter which would make it consume only 1100w or 22A from your battery. It's not detrimental at all to the battery to cycle this load on/off every 5min.
 
With standard 4500w 240v heating element used in 50 gal water heaters you can run it on 120V from your inverter which would make it consume only 1100w or 22A from your battery. It's not detrimental at all to the battery to cycle this load on/off every 5min.
Now that's the thing, I got watching videos on "micro cycling" and what this guy was saying is a discharge of (5%) x20 times a day = one full battery cycle?? Idk if this is accurate but if true would totally change the "free power dump load" if it means I need to replace my batt 2 years early
 
I got watching videos on "micro cycling" and what this guy was saying is a discharge of (5%) x20 times a day = one full battery cycle??
No that's not true. Besides, 5min at 22A would only be 1.8% of 100Ah capacity. In addition to that non issue, some power will come from solar panels themselves.
 
Would you be better off changing the MPPT for a morningstar Tristars PWM , using it on load division , almost like a wind turbine dump load


Just an idea
I didn't think anyone used pwm for charging batteries anymore? Perhaps if the Victron when entering float mode triggers a relay to divert DC to that pwm charger and run
No that's not true. Besides, 5min at 22A would only be 1.8% of 100Ah capacity. In addition to that non issue, some power will come from solar panels themselves.
Right, well my inverter is 220v and so is my heater, it has two modes sink 1100w and bath 2800w so while it would be nice to utilise the full 2800 as a Dump when the sun is full blast, with the system your describing it would probably be better to stick with 1100w?
 
Depends on how much solar you have and inverter max rating.
On a good day I have 4kw PV, inverter rated for 3kw 5kw peak, and dyness 48v 100ah lithium, I done some tests and while I have 4kw pv that can drop to 500w when I cloud passes over so the battery will have to carry the full2.8kw load untill the charge drops enough to cutoff the dump load or the sun comes back, on an average day I'd expect It to cycle like that at least 20 times
 
Hmm, not enough spare inverter capacity. Your case is getting a bit complicated. Consider switching 2800w heating element to a 2kw 48V DC element if water heater is located close by to avoid too much wire voltage drop. DC heater can be switched on/off via DC rated solid state relay.
 
Hmm, not enough spare inverter capacity. Your case is getting a bit complicated. Consider switching 2800w heating element to a 2kw 48V DC element if water heater is located close by to avoid too much wire voltage drop. DC heater can be switched on/off via DC rated solid state relay.
Why 48v? Could I not take the pv direct from panels? Can't remember exactly but it runs at about 190v
 
You can but it won't be MPPT, will not be logged as solar generation in victron app and may confuse your charge controller.
 
I do PWM water heating directly off array. It is the most efficient method as far as conversion loss from wires, battery, electronics and inverter. The battery full output of the controller can be used to switch the array voltage in a direct connect method. Studies have shown that 2X ideal resistance produces more daily power. The early two module ACTii appears to have a common voltage shunt option. I know of no others. Many have built their own as these are fairly simple circuits I have a very minimal battery and heating water from panel voltage doesn't screw with charging or battery level and saves having to have an extra KW of battery. Here is what I use, the wall picture is my first arduino system under bubble wrap IP67 protection. One of these was made with a $9 inverter board from China. While many of you have tons of excess power, this picturehotwater_1.jpg is a southern view of my camp. See any panels? A challenging site, yet I have plenty of hot water due to how efficient the system is.
 

Attachments

  • wallheat.JPG
    wallheat.JPG
    77.8 KB · Views: 29
  • InkedCamp View_LI.jpg
    InkedCamp View_LI.jpg
    1 MB · Views: 29
  • NHW01.jpg
    NHW01.jpg
    277.9 KB · Views: 28
  • HOT_WATER_SHED.JPG
    HOT_WATER_SHED.JPG
    105.9 KB · Views: 28
Back
Top