Solar panels for battery backup with trickle charge

lbenson

Senior Member
#1
I've been looking at a backup water system for my house to provide water from my shallow well in case the electricity goes out. I had been thinking about a 12V pumping system, but then after mulling over how to get the water where I wanted it, it seemed to me easier to just provide backup to the 120V mains by putting what I want to keep running (pump, fridge, freezer) on a separate branch with a transfer switch to automatically switch over if mains goes out.

The battery or battery bank would be charged by solar with a trickle charger from mains. I'm looking at the Windy Nation 2-100 Watt panel kit with P20L controller.
https://www.windynation.com/Polycry...-for-12-Volt-Battery-Systems/-/362?p=YzE9NDY=
https://www.windynation.com/cm/Complete_Solar_Kit_Manual_R2.1.pdf
and https://www.windynation.com/cm/P20L Controller Manual_R1.pdf
P20L Controller.jpg

I have a 100Ah battery and plan to get a 1500W pure sine wave inverter.

The controller will shut off solar charging above 13.8V and load below 10.7V (both settable). After the load is turned off, it will restart if the battery voltage reaches 12.6 (also settable). The controller takes 6 wires--two from the panels, two to the load, and two to/from the battery.

I plan to inject mains trickle charging at the battery terminals, using a 5A DC-DC buck regulator set at 13.8V with a 7.5-amp MBR745G Schottky diode and an appropriate current-limiting resistor.
DC to DC converter 5A.jpg

I had thought I would use a picaxe to provide low-voltage cutout, but with these components, especially the controller and transfer switch, it seems to me that these parts are "self-leveling".

My main question is about an appropriate value for the current-limiting resistor. Since the controller assures that the load is disconnected if the battery voltage drops below (say) 10.7, the maximum difference between the buck regulator and the battery is 13.8-10.7 or 3.1 volts. With a 2-ohm resistor, that would be 3.1/2 or 1.55 amps (times 3.1 equals 4.81 watts). Is my math correct?

4.8 watts sounds like a lot for a 5W resistor, even with heat sinking. If there are 2 1-ohm 5W resistors in series, does that cut the heat in half for each? Or do I need 10W resistors.
Code:
.___________.       .____________.      .____________.
|           | ~15V  |            | ~13V |            |
| 200W, 12V |-------|   P20L     |------|  Inverter  |
| Solar     |       | Controller |      |            |
| Panel     |-------|            |------|            |
|           |  0V   |            |  0V  |            |
'-----------'       '------------'      '------------'
                       |      |               | 120VAC              Load
.___________.          |      |         .____________.         .____________.
|           | ~13V8    |      |         |            |         |            |
| DC-DC reg |----------o      |         |  Transfer  |  120VAC |    Pump    |
| Diode     |          |      |         |   Switch   |---------|   Fridge   |
| Resistor  |----------|------o         |            |         |   Freezer  |
|           |  0V      |      |         |            |         |            |
'-----------'          |      |         '------------'         '------------'
                       |      |               | 120VAC
                     .____________.      .____________.
                     |            |      |            |
                     |   Battery  |      |            |
                     |    100Ah   |      |   Mains    |
                     |            |      |            |
                     |            |      |            |
                     '------------'      '------------'
I haven't done a complete analysis to make sure to not overload the inverter, but I think if I can make sure that the freezer and fridge don't run at the same time, I'll be fine. On a normal day, the water pump may run 10-12 times for 40 seconds--not many amp-hours in total.

Does this make sense?
 
Last edited:

Pongo

Senior Member
#2
Math looks OK. "If there are 2 1-ohm 5W resistors in series, does that cut the heat in half for each?" - That will split the power between them. Personally I grew up on .gov electronics rules and would never run a 5W resistor at 5 Watts 'cause it will get pretty hot. My choice would be this metal case type of resistor running well below its rated power. I also would use a higher rated DC-DC module, those little ones aren't very happy running near their specified max.

Are you sure that applying the trickle charge voltage won't confuse the solar controller into thinking the battery is fully charged?

I used a similar AC charger/battery/inverter scheme to back up salt water aquarium pumps. I found that I needed to oversize the inverter to reliably start the pumps.
 

premelec

Senior Member
#3
As Pongo says there may be a conflict between PV and trickle - I would suggest that trickle only be connected when battery voltage is less than a certain value [perhaps 11 volts]... you could add in PV state as well - these systems always require a certain amount of compromise and knowledge of what load will be used when etc...
 
#4
As Pongo says there may be a conflict between PV and trickle - I would suggest that trickle only be connected when battery voltage is less than a certain value [perhaps 11 volts]... you could add in PV state as well - these systems always require a certain amount of compromise and knowledge of what load will be used when etc...
While I agree with the principle of trickle charging, you should never let your battery/s get that low because internal damage starts at around 12 volts. I would start trickle charging if the voltage drops to 12 volts. 12v is about half of the capacity of lead-acid technology and that is all you can safely use.
 

techElder

Well-known member
#5
Well, Lance didn't exactly say he was choosing "lead-acid technology" (there's an oxymoron!), but his choice of voltage levels seems to suggest that.

There are many solar / battery systems that happily use something besides lead-acid. I'm remembering Tesla. :D

Lance, since you aren't trying to power up the neighborhood, have you investigated the "lithium" style of batteries? There are many examples and 100Ah is not very large for them.

Chargers for "lithium" style systems are quite sophisticated, too. They may simplify your solar system significantly.
 

hippy

Technical Support
Staff member
#6
it seemed to me easier to just provide backup to the 120V mains by putting what I want to keep running (pump, fridge, freezer) on a separate branch with a transfer switch to automatically switch over if mains goes out.
Sounds sensible to me. But why complicate things with solar panels rather than just charging and trickle charging from mains when that is available ?

Adding solar panels and controllers, figuring it all out, seems to me to be a lot of extra cost and complexity with little gain.

I'm not entirely sure what's wrong with just adding a separate UPS powered mains ring. After wiring that in and buying a UPS; job done.
 

lbenson

Senior Member
#7
Thanks, all. Those 25W resistors are hefty, Pongo--just the ticket. IP--I was being lazy and using the low-voltage shutoff value shown in the P20L controller manual--on a similar but much smaller project I used <11.9 as the shutoff, but 12V it is.

Tex--From what I can see, a 7kWh Tesla powerwall costs $3,000 plus probably another 2K to 4K for installation. I'm too afraid of lithium to try anything DIY--especially at those voltages and currents (7kWh is much more than I need).

I'm not entirely sure what's wrong with just adding a separate UPS powered mains ring. After wiring that in and buying a UPS; job done.
hippy--I'll look into the UPS option. It would simplify things. One reason for the solar was to provide electricity in the event of an extended power outage. But in my 16 years in this house, we've never experienced one, perhaps in part because we're only a mile downstream from the first of five power-generating dams on the Mersey river, and less than a mile from the collector station--it's hard for them to get power working for locations downstream from us without getting it to us. I'll have to assess power needs more rigorously.

Are you sure that applying the trickle charge voltage won't confuse the solar controller into thinking the battery is fully charged?
My experience with my smaller (7Ah SLA) system was that when the buck regulator was set to 13.8 volts when disconnected, when connected to the battery the measurement was whatever the battery was. I mean to ask on the WindyNation forum, but so far I haven't gotten the registration emails which their system says they sent (not in spam either).
 
Last edited:

RexLan

Senior Member
#11
You may wish to do some more math and I believe you fill find that that 100 Ah battery probably won't even start your freezer much less run all that stuff. The locked rotor current on the pump motor probably won't get what it needs to start. I had 1200 Ah batter bank at our cabin and it would only run ~ 3 days with a premium 97% efficient Trace inverter. Just a freezer, TV, computer and ocassional lights.

As for the propane bottles you have. They are only filled to 16#'s because of the safety valves and they won't last except some hours and I would be quite surprised if it would run as advertised. Additionally, the generator is MUCH more than you need, but it will run everything.

A small 1000 watt Honda generator will run a very long time on a small amount of gasoline, quiet and very reliable. Buy a 5 gallon jug and put STA-Bilo in it and it is good for the generator about a year. Put it outside in the shade. Rotate it out in the car annually. Check the pump and be sure you have start current - not just the run current. May need a 1500 watt unit.

Like you said though ... probably don't need anything and just keep the fridge and freezer doors shut. Freezer will last a solid week. Start the pump and fill the tub then good a 2 days.
 

lbenson

Senior Member
#12
More reading has only made me more confused. For instance, the rating of UPS devices: I see a CyberPower CP1500PFCLCD PFC Sinewave UPS System, 1500VA/900W UPS tower. 1500VA sounds good. So a question at Amazon about how long a cable modem and router drawing a total of 30 watts can run is answered, maybe two hours. If "VA" means "volt amps" (and volt amps are watts), why isn't the answer 1500/30, or 50 hours--or 900W / 30 or 30 hours? So it is said to have to do with the C/20 rate--but is this then 900 watts at 120 volts, or 900/120=7.5 total amps divided by 20 equals .375 amps at C/20. Multiply this by 120 volts, and at C20 you should have 45 watts available for 20 hours (less efficiency losses). How can you get only 2 hours worth of 30-watt draw?

Is my math wrong? Or my understanding of the meaning of the terms?

Further I read that sometimes an electrical "Authority Having Jurisdiction" (AHJ) will forbid the switching of normally hardwired appliances like well water pump or boiler from being converted to plug-in so that a UPS could be used on the grounds that it may disconnect the appliance from the normal grounding. (This would seem also to apply to plugging into a generator.) (I'm not sure why this disconnection would necessarily be the case, but perhaps it depends on the UPS.) Having a branch circuit with a transfer switch obviates this, I guess.

I also found that I had posted a related question (about boiler backup--not implemented) 4 years ago, and picaxe member goom had recommended the AIMS Power Pure Sine Inverter with Charger & Automatic Transfer Switch in various wattages (batteries not included). In the higher wattage range with suitable batteries, this might be good for a non-solar option.

(Rexlan--thanks for the advice on sizing and other issues.)

I still have an interest in experimenting with small-scale solar, since the WindyNation kit is within my price range, and seems understandable.
 

techElder

Well-known member
#13
Is my math wrong? Or my understanding of the meaning of the terms?
... a question at Amazon ... is answered
The second quote answers the first quote! :D

I do know that when my water well pump (240VAC) comes on, it really puts an audible load on my Honda EU6500is generator. I can't imagine a 1500VA UPS being able to run it; let alone start it. (We do have different wells here in Texas. Mine, at around 200 feet, is considered a very shallow well.)
 

RexLan

Senior Member
#14
Keep in mind that most of these "high efficiency" solar panels are actually about 20% efficient and will never produce in real life their advertised hype. All their math is based on lab conditions .. not the real world. You might get 3-4 amps @ 12Vdc from the 200 watt panels and their poor electronics will eat about 50% of that in the conversion to AC.

I believe the electrical issue is likely related to your equipment not being properly grounded while connected to the power mains which could inadvertently cause a catastrophic failure if it failed and the commercial power was returned. However, if the appliance is totally removed from the commercial system, like a freezer with a cord, there is no issue so long as there is a manual disconnect/connect isolating the two systems so they cannot back feed.

I have a generator and a transfer switch here at my new house and had it in Alaska and Florida. Works well and I have used it a few times. I use a switch like this (manual) and it is hardwired into the main distribution center .... all meeting code. It simply rerouted the source to the resective circuits I have chosen like the fridge, freezer, etc.
Transfer Switch

I think you will be quite disappointed in the solar setup and keep in mind that the
ratings" are only good with a perfect and fully charged battery. Drop the voltage a few 10ths and the available power is gone.

The pure sine wave inverter or Sola transformer is good but also keep in mind that most things don't care or need it. Exception is a TV which need the cycle clock freq. to sync the horizontal sweep. Without the picture will roll.
 

hippy

Technical Support
Staff member
#15
Is my math wrong? Or my understanding of the meaning of the terms?.
I suspect the answer on Amazon was wrong, but you have to remember that VA or W rating isn't a time-domain figure. Think of it more as what it can handle without overheating or catching fire, more a rating of whether it's sufficient to power what you want to keep running.

How long a generator will keep delivering power for will depend on the size of its fuel tank.
 

Pongo

Senior Member
#16
The second quote answers the first quote! :D

I do know that when my water well pump (240VAC) comes on, it really puts an audible load on my Honda EU6500is generator. I can't imagine a 1500VA UPS being able to run it; let alone start it. (We do have different wells here in Texas. Mine, at around 200 feet, is considered a very shallow well.)
A "shallow well" is typically 25 feet or less with a relatively small pump. You and I have deep wells, mine is 240 feet and my backup gen is a similar size to yours - a whole different ball game.
 

AllyCat

Senior Member
#17
Hi,

Is my math wrong? Or my understanding of the meaning of the terms?
Hmm, IMHO there's quite a lot of confusion in this thread. Firstly, the run time probably will be determined mainly by the capacity of the batteries, not by the inverter. For a 100 A.hr, 12 v battery that's (theoretically) around 1000 watt-hours but as said above you don't want to fully discharge the cells (even if they let you) and there will be various inefficiencies (losses) in the system, so maybe 500 watt-hours in practice, perhaps 15 hours @ 30W. That suggests the particular Amazon "review" might be rather inaccurate, but no surprise there. ;)

VA is only equivalent to watts for a resistive load. Motors (particularly smaller ones) may have a significant inductive component, so that the current and voltage are out-of-phase with each other. The "real" power is the (instantaneous) product of voltage and current, so the average "power" is very often (much) less than the (required) VA rating. Again, as mentioned above, the starting (stalled) power of a motor may (need to) be very much higher than the normal running power. Also consider bridge-rectifier circuits which may consume an enormous "gulp" of current just at each peak of the sinewave (after a longer gulp whilst charging up the reservoir capacitors initially), in many "dc" based systems.

A "true sinewave" inverter will use either an enormous (semi-tuned) transformer or (nowadays) a modulated PWM system (that's double-modulation: typically a 60 Hz sinewave modulating an ultrasonic switching output stage). The disadvantage of the former is obvious, but the latter may suffer from severe problems with "transient" effects (i.e. gulps of current load, as described in the previous paragraph).

I believe the "quasi-sinewave" systems are "switching" systems with typically five "dc" levels: Initially they generate a voltage level of (say) 50% of the peak for a few ms, then switch to 100% for rather longer, then back to 50% for a few ms before returning to zero (overall taking 8.3 ms in USA, etc, 10ms in Europe, etc.). This is then repeated for a similar negative half-cycle (i.e. a frequency 60 or 50 Hz as appropriate). Note that either system might be highly accurate (or inaccurate) in absolute frequency terms, but I don't think that TVs have been synchonised to mains frequency since John Logie Baird, over 80 years ago. However, the field (frame) rate is chosen to be close to the mains frequency to avoid nasty "flicker" (low frequency beat) effects, because the (studio) illumination (incandescent lamps, etc.) is probably modulated at mains frequency. The main issue with quasi-sinewave systems is that there are high frequency edges which may cause problems with capacitive loads (I was surprised that the mains-LED that I dismantled, contained nothing much more than a high-voltage series capacitor).

I do agree that "solar" is really not an appropriate technology for "backup" purposes. If you are going to spend lots of money on solar panels then you should use all the power generated. But a rated 1kW panel should generate that power under the rated conditions (basically clear sunlight, directly overhead at the equator), their 20% efficiency is already taken into account. Of course in higher latitudes (e.g. Canada), even with cloudless skies, a "1kW panel" may deliver only a few hundred watts for a few hours a day in Winter. Then, as mentioned above, there will be losses in both the storage and voltage conversion/inversion processes that might well amount to efficiencies of only 70 - 80% each).

Cheers, Alan.
 

Pongo

Senior Member
#18
More reading has only made me more confused. For instance, the rating of UPS devices: I see a CyberPower CP1500PFCLCD PFC Sinewave UPS System, 1500VA/900W UPS tower. 1500VA sounds good.
You're right, consumer UPS's have very small batteries. Not worth to consider for this application unless you hook up a larger external battery, but then the UPS charging circuit is inadequate to recharge it so it's not a great idea.
 

lbenson

Senior Member
#19
Thanks again to all for my ongoing education. Pongo, you are right about my shallow well--the well casing is about 3 feet in diameter and the surface of the water is only 12-15 feet down. I once poked to the bottom of it, but can no longer find the pole I put together to do that--painted black, then white every 6 inches. (In the severe drought this summer, the shallow wells of both neighbors downstream from me went dry, and many others as well, but ours didn't.) The pump is a 1/2hp lift pump, in my basement about 25 feet from the well casing.

I would not, indeed, expect to get much PV power in the Nova Scotia winter (and I'd wonder about how pole-mounted panels would do in a nor'easter). I would be interested to know how much power you would get from 2-100 watt panels. I doubt the pump, older fridge, and older freezer would care much about a modified sine wave, but the newish oil boiler--by far the most crucial and most expensive item--has some microprocessor components. Some years ago I got no help from the boiler provider regarding suitable backup power. The power company might be delighted to sell me a powerwall.
 

premelec

Senior Member
#20
There are lots of things to consider.... I've read Home Power Magazine - www.homepower.com for decades and their site has a lot of useful information... I use 100 watts of old PVs to charge a bunch of parallel 12v batteries which could run my furnace circulation pump and ignitor in a pinch [I worry more about frozen pipes than keeping a fridge going...] - 100 watts four hours a day is a lot of energy if you use it wisely - in the case of water storage of fluid rather than trying to be able to pump may be best solution [I used to have big tank in the attic... now I'm on gravity feed from hillside in different house]. There are likely places to get figures of insolation for your location - have you got any PV dealers or neighbors with PVs. For all of USA there is a PVwatts site on internet...
 

lbenson

Senior Member
#21
Thanks for the encouragement, premelec. I also thought that 100 watts an hour for 4 hours per day would go a good ways in an outage. And it's certainly true that running the boiler is much more important than fridge or freezer (freezer in particular--it's in the basement where the temperature may be around 40F on the coldest winter days).

Part of my problem (and my desire for a backup) is that the house is unoccupied in the winter, and kept at around 52F to keep 170-year-old plaster from going through freeze-thaw cycles. This also makes a non-instant-on generator an option only while the house is occupied (but not a bad one). I can determine from temperature readings on supply/return pipes how often the boiler runs and for how long, but to see what the current draw is I'd have to put a plug on the hookup wire and plug it into my kill-a-watt meter.

Or I can just hope that everything goes as it has for the past 18 winters, with no prolonged outages, and that my monitoring system can tell me if I need to get the plumber in because of a leak or stuck zone valve (both of which have happened and I've caught it and gotten it fixed without serious harm being done).

I have no hill from which water may be fed, and I fear that water storage in the attic could result far worse damage than the absence of water. If I had natural gas or cooking/heating LP, I'd probably get a whole-house generator.
 

premelec

Senior Member
#22
Another thing to remember besides insulation.... it's BIG... Thermal mass! Just a bunch of plastic jugs with water near what you don't want to freeze slows the whole thing down - down side is it also takes a lot of time to warm up - but in the un occupied situation that's not so much a problem... and if you have a phone line or cell service to the property you can make a PICAXE phone line dialler [use pulse rather than TT or servo to push the button on cell phone] to call you if it goes below 35 and you can snow shoe in to build a fire.... There are heat tapes for pipes if you don't have a lot to protect and insulate around them so not too many watts to the tapes... I use Taco Sentry zone valves - less problems that the White Rodgers units I started with and don't take ongoing power like Honeywells... gads here comes winter!
 

lbenson

Senior Member
#24
I do have internet service, and my picaxe + linix openWrt router monitoring system sends me email several times a day and texts if a sensor goes off or temperature is too low or too high. UPS will keep these alive for a while if power goes out. Plumber or a neighbor can come in to check--no snowshoeing required.

We have usable cell phone service here, but the picaxable GPRS GSM SMS modules I've gotten (A6 and Air200T) don't find a strong enough signal. If I drive them down to where I can see the cell tower from about a mile away, they're fine, so it's not a configuration problem. Both modules have worked for me in Florida.
 

fernando_g

Senior Member
#25
Like almost everything on Ebay, I would strongly suggest that you validate that the module does do what you want it to do.
That includes maximum load operation at your operating temperature extremes.

Lastly, if you plan to use it outside, do give it a pair of coats of humiseal. Mask the outside of the heatsinks with tape when you are spraying the humiseal coat.
 

lbenson

Senior Member
#26
Like almost everything on Ebay, I would strongly suggest that you validate that the module does do what you want it to do.
That includes maximum load operation at your operating temperature extremes.

Lastly, if you plan to use it outside, do give it a pair of coats of humiseal. Mask the outside of the heatsinks with tape when you are spraying the humiseal coat.
Right, thank you. Everything but the (eventual--not this year) solar panels will be inside.

I bought this hardwired inverter plus transfer switch for backup (for starters) for my boiler, if I can get my electrician to wire it up. Battery and charger are separate: https://www.amazon.ca/gp/product/B00TI1D5JK/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

I did this after reading someone's account of success with a similar system.

I still don't know the exact draw of my boiler, and won't until/unless I put a plug on my hardwired unit and plug it into a kill-a-watt module.

I looked back at my monitoring of temperatures on the zone supply and return pipes. It looks like my lowest daily overall temperature was on March 12, 2017, with a low of -14C and a high of -2C (other days had lower lows, but higher highs). Judging from a rising temperature in the (combined) return pipe, the boiler pump came on 29 times. I don't know for how long. The pump can run without the burner igniting if the water in storage is hot enough. If the burner does come on, the pump may (and probably will) run for a while after it shuts off.

I would guess that on average, the circulating pump ran for about 4 minutes, providing enough warmth to shut off the thermostat relay. The thermostats are set at 6C--which might not be enough to provide a reservoir of heat if the power goes out. I'm hoping that the battery/inverter might provide about a day's worth of runtime--around 4*30 or 120 minutes. But without measurements, I'm guessing.

I heard back from windynation. They said a mains trickle charger (B&D, Nico, etc., instead of the DC-DC module) was fine as shown on my diagram. They even complemented me on the ASCII art diagram; said they'd never seen anything like it--thanks, hippy, for showing the way in keeping ASCII art alive.
 

lbenson

Senior Member
#27
I have further modified my thoughts on this arrangement, and WindyNation has come up with a new solar charging controller which includes remote access to the monitoring results: https://www.windynation.com/cm/TrakMax40BT Controller Manual_R1.pdf

Only a light load can be taken off the controller, so the inverter comes directly off the battery, resulting in a new schema:
Code:
.___________.       .____________.      .____________.
|           | ~15V  |            | ~13V |            |
| 200W, 12V |-------|  TrakMax   |------|    Light   |
| Solar     |       | Controller |      |    Load    |
| Panel     |-------|            |------|            |
|           |  0V   |            |  0V  |            |
'-----------'       '------------'      '------------'
                       |      |                                    Load                   Load
.___________.          |      |         .____________.         .____________.         .____________.
|           | ~13V8    |      |   ~13V8 |            |         |            |         |            |
| Trickle   |----------o------|---------|  Inverter  |  120VAC |  Transfer  |  120VAC |    Pump    |
| Charger+  |          |      |         |            |---------|   Switch   |---------|   Fridge   |
| Diode     |----------|------o---------|            |         |            |         |   Freezer  |
|           |  0V      |      |      0V |            |         |            |         |   Boiler   |
'-----------'          |      |         '------------'         |            |         |            |
                       |      |                                '------------'         '------------'
                       |      |                                      | 120VAC
                     .____________.                             .____________.
                     |            |                             |            |
                     |  Batteries |                             |            |
                     |    200Ah   |                             |   Mains    |
                     |            |                             |            |
                     |            |                             |            |
                     '------------'                             '------------'
I won't do the solar part this year, but hope to do some aspect of the mains-based battery backup system this fall. I have ordered some of these to monitor usage of fridge and freezer via wifi:
https://www.newfrog.com/product/sma...-remote-control-timer-switch-outlet-us-235152

For the period of time in the winter when I don't occupy the house, I'd like to implement a round-robin schedule to power one load circuit at a time, like this:

Load Timing Schedule:
0000-0100 Boiler
0100-0200 Freezer
0200-0300 Boiler
0300-0400 Fridge
0400-0500 Boiler
0500-0530 Water Pump
0530-0900 Boiler
0900-0930 Fridge
0930-1600 Boiler
1600-1630 Fridge
1630-2200 Boiler
2200-2230 Water Pump
2230-2400 Boiler

Of course, the batteries must be sized to suit the heaviest load.

I'm reasonably certain that in our less-than 8C basement, an hour a day for the freezer will be fine.

Monitoring last year showed that the water pump ran only twice for about 40 seconds at a time in the month of March. With water shut off to everything but the boiler, the only thing it does is pump up the pressure tank from whatever losses there are there. If it runs for much longer than that at any time, I'll know there is probably a leak and can call the plumber.

With the house also at around 8C, the fridge won't run very much--I think allowing it to run 3 times a day will be fine.

The boiler is the big question--the circulating pump is 1/25th hp--about 30 watts. As said above, on the coldest day it ran 29 times for about 4 minutes at a time. But I don't have any idea how much the oil heater/atomizer/sprayer takes in the way of watts.

I'd like to find a transfer switch which could cycle through 4 circuits based on a timer, but that may be asking too much. If I have to implement it with outlet modules like the monitoring/timing ones I ordered, I'll have to get Canadian-approved ones, and will also have to approval from authorities and insurance company for use on hard-wired appliances like boiler and water pump. And I don't wish to add so many failure points that the setup is less reliable than the power company.

Perhaps for this winter I can just do non-critical elements like the freezer and fridge to see how the setup would work. And maybe the water pump since it runs so rarely and with a monitored failure I'd be able to get someone to look at it before harm was done. The critical element in the event of a longer power failure is the boiler--burst pipes would be the biggest problem--perhaps another charge of antifreeze in the pipes is in order.
 
Last edited:
Top