Off Topic (Sort of): 12V Battery backup for house systems

lbenson

Senior Member
Apologies for a long post with only a bit picaxe-related.

I don't want to hijack SolarMike's "08M2 PV MPPT Charge Controller", but papaof2's post has me wanting followup:
I've been "dabbling" with solar for several years, primarily as a "Wait until daylight" option for the frequent but usually short grid outages we have and I've tried a number of commercial solar charge controllers - PWM and MPPT - some are much better designs than others. I'm interested in how "real world" usage works for your design over time.
The solar irradiation for Atlanta works as a rule of thumb for 50-100 miles out from there so we basically have 3 useful sun hours in winter and 5 useful hours in summer. The current system is a 12 volt, 420AH battery bank which is kept charged by a single 250 watt panel via an MPPT controller. With a 2000 watt pure sine wave inverter, the system powers fridge, freezer, a few LEDs and the internet for 8-10 hours - long enough for a storm to pass or until daylight and I could then go out to set up a small (1600 watt) inverter generator. The average running load is under 300 watts but you must allow for the surge watts of starting motors (fridge, freezer, etc.) which is typically 5 to 7 times the running watts but can be 10 times the running watts.
For a number of reasons I feel there's no hope for solar for me in Nova Scotia, but I would like to have backup for oil boiler, well pump, sump pump, fridge, and freezer this winter. Normally, we're snowbirds, but covid (<1 active case per 100,000 in NS over the past 4 months vs for Florida, 3,325 total per 100,000 people) and the likelihood that my wife couldn't get travel health insurance means that we'll overwinter here.

My current 12V battery bank amounts to about 350Ah, but they're all mix and match batteries, so I'm assuming I shouldn't charge them simultaneously, but should disconnect (under picaxe control) them one by one and charge. The very knowledgeable Warpspeed from TheBackShed.com says that for longest life, you shouldn't keep lead acid batteries under continuous maintenance charge, but should let them drop to around 12V2 (self-discharging) and then recharge. That seems to me not to leave enough headroom should mains go out.

We are within a mile of the distribution center for the electricty generated by 5 dams up the Mersey River, so we usually get back pretty quickly after an outage (we were out for 23 and a half hours after a hurricane last fall). If the batteries gave out I can recharge from a 3KW propane generator.

I had figured out and have coded and tested a round-robin system using SSRs under picaxe control where each appliance (sump pump, boiler, well pump, fridge, and freezer) is switched to in turn, but papaof2's setup seems to suggest that my 3kW pure sine wave inverter might be able to handle all at once. I'm sure it could once startup was achieved, but I've feared that if some decided to turn on simultaneously, over-current shutoff could occur.

I don't know the running or start-up draw of the appliances--except for the freezer, all are hard to get to, and boiler and water pump are single-circuit hard-wired. I have wiring run now to the sump pump, but I'd have to figure out how to flood the sump pump well in order to measure it's draw. As far as usage goes, the freezer comes on for 15 minutes about every hour and a half. The well pump runs for less than a minute for about 15 times a day. The boiler is entirely dependent on how cold it is. The sump pump ran nearly continuously until the power gave out in the hurricane last fall, but didn't ever come on when Hurricane Teddy swiped us last weekend. (The house foundation is piled flat rocks and the floor is gravel--there's no way to keep water out in a deluge.) The fridge would have to be moved for me to put the plug-in power meter on it.

I'm sure there's more, but it's a long post. I'd be interested to know what people with experience or ideas think.
 

premelec

Senior Member
This is pretty much a personal thing... so many ways to go... for the life of the publication I read HomePower magazine to see all the things people had done off grid, micropower etc... you can see their archive at www.homepower.com. Judging your load and actual needs are primary... with enough money anything is possible... and batteries are improving with surplus from the crashed / old electric car industry. My personal decision has been minimum battery storage with small inverter which will run my boiler for a day but not refrigerator etc... last 24 hour grid outage [early snow] forced defrost of 1948 fridge I use because it only draws 12 KWHr/mo. - Note that in general larger inverters draw larger idle currents which is why I use a 250 watt unit [I had a 2Kw unit running before... ]. I like sine waves though they are pricier ... so many balances to be considered... and in winter the whole outside is a refrigerator and sometimes a freezer... good luck...
 

Flenser

Senior Member
Ibenson,

Your research found "TheBackShed.com says that for longest life, you shouldn't keep lead acid batteries under continuous maintenance charge, but should let them drop to around 12V2 (self-discharging) and then recharge. "

Lead acid batteries have been used as battery backups for computer systems for a very long time and these battery backups need to be charged, or nearly charged, when the power goes out or else they are not very useful.

I imagine battery backup systems can't let the voltage drop to 12v2 when not in use and the need to keep the battery topped up vs the desire to get the maximum life out of the batteries is likely trade-off everyone designing battery backup systems has to make.

Checkout this link which describes float charging lead acid batteries for standby use and has a chart showing the loss in capacity over the years vs the float charging voltage used;
 
Last edited:

papaof2

Senior Member
If you have access to a clamp-on AC ammeter ($35US at RadioShack this week) and you're sufficiently knowledgable of AC power safety, you can open the AC breaker panel and measure the amp draw of each appliance there. If you were living in an old house with knob and tube wiring, it would be easy to get to the wiring for checks ;-) In the days of non-smart electric meters, you could have ensured all other lights and appliances were off and just counted the meter's revolutions/minute when the appliance of interest was in use.

My previous battery bank was 540AH of used AGM batteries (six x 90AH) that I found on Craig's List for $35 each when they were changed out by age (approaching 6 years) from a local datacenter's big UPS. Anticipated life in that service (always at float) is 5-7 years but all of them tested as new. I got another three years of life (to 80% of original capacity) with them kept at float voltage (13.65V) during the sunny hours and dropping back closer to 13 volts overnight.

I would never let a battery sit idle for an extended period, as I've seen batteries only a year old destroyed by not being checked while stored and the voltage dropping to a point of unrecoverable sulphation. I picked up some 15 month old 100AH batteries that were at 8 volts for $5 each simply to have cores when I must buy new batteries as core charges are around $20 each - two of those batteries could be recovered to 80% of capacity but that is end-of-life in conmmercial service. Most of the recommendations I've seen are to not let the voltage drop below 12.4 while in storage and that would seem to apply to your use.
Remember that any lead-acid battery only has a life of X charge/discharge cycles, with X depending on the level of discharge. For a typical AGM battery (not a high-end Trojan or the like), X is 1200 for 10% discharge and about half that for 50% discharge - which is about where 12.2 volts would put them.

I've read more battery documents than I'll ever remember but I do save useful bits along the way. One of those was from a paper that found batteries regularly discharged 10% lasted longer than those always kept at float voltage. I'm testing that finding on the current AGM bank by doing a measured 10% discharge about twice a month. Powering a 440 watt halogen work light via the 2000 watt inverter for one hour gets a 40AH (+/- a bit as I'm using an alarm to remind me to turn things off and not a timer on the light). If we use the estimated 1200 charge/discharge cycles of battery life at 10% discharge, then 2 cycles per month is 24 cycles per year so 1200 / 24 = 50 years which means any possible increase in battery life from those regular charge/discharge cycles doesn't cost anything in real world battery life. If anyone wants the reference to the 10% discharge being better than constant float I can hunt for it.

I mostly mention AGM batteries because those are what I have the most experience with in standby use, but I also have a few flooded lead-acid batteries (car, truck, riding mower) and I replaced the battery in the truck at age 6 years because it sounded a little slow when cranking. Two years later, that battery is still in the basement for testing inverters and doing jump starts as needed as it still tests at 90% of new. Some might consider me paranoid about battery care but things that should always be there when you need them should get regular and careful attention ;-)
 

lbenson

Senior Member
Flenser--thanks for the link. I've seen a number of references to "Battery University", but have never followed the links. Looks very interesting. And you're right about the trade-offs--availability of sufficient capacity when needed is more important than life of the batteries (within reason).

Papaof2--I was hoping to hear from you. Thanks very much for your practical experience built on years of research. Exactly the kind of information I was looking for to continue my own experiments and to get my practical setup started. I've also used a halogen work light for my round-robin testing--300W--and also a 800W electric heater.

I'm sufficiently comfortable with AC wiring to have wired new circuits in a number of houses, wired my barn, and replaced/added breakers, but my breaker box is difficult of access for me, so I'll wait to do testing until I've got my new wiring in for my 5 appliances to be backed up. I have a plug-in 120V voltage and amperage monitor, but I can't easily get to my fridge plug, and the boiler and well pump are hard-wired. I'm working on monitoring those.

Among my collection of inverters I have a 1kW automatic transfer inverter, but I don't have anything which needs instantaneous switchover.

Any thoughts from anyone on whether I'm better off with a single bank of 6 batteries plus a 3kW pure sine wave inverter to handle all 5 appliances, or, say, 3 pairs of batteries plus inverter to run 1) boiler; 2) fridge and freezer; and 3) well pump and sump pump?

I am confident that I can control amount of discharge and recharging with a picaxe plus ESP-01 for wifi reporting.
 
Last edited:

Eng460

Well-known member
Hi Lbenson,

I would suggest the single bank of batteries plus a large inverter, as you have several varying loads which would each need a bit of extra reserve capacity in their own separate banks, whereas with one bank, the reserve can be used by which ever load needs it, so overall less battery capacity required.

Similarly, a single large inverter, as the inverter needs current and power capacity to start each of the loads. Separate inverters means a lot of spare capacity so each can be started if in separate systems. In the same vein, I really like your suggestion of using the Picaxe to sequence at least the starts, if not the run time. Every time you sequence two loads to start or run separately, you reduce the total inverter capacity required. Diversity is your friend.

It Is important to understand how the various loads operate. The fridge loads can be sequenced, as as it is most likely that they each operate for less than half the time, so never need to run together, just alternate them. The oil pump for the heater may run continuously, or may operate in an on/off manner. But either way, it should be reasonable to at least queue the starts so two or more motors can not just by chance start together as I do not expect the start tine to be critical.

Again, when an electric motor starts, the current can be 7 to 10 times full load current, but the power factor is low, so the power reaches a peak of about twice full load just before it reaches its running speed which is a few percent below synchronous speed for the power supply frequency and number of poles in the motor windings. It only lasts perhaps two seconds, but long enough to blow a fuse. All motors draw full available power due to accelerating due to inertia, so the run up time can be longer for a load which increases the inertia of the rotating system. Your heater oil pump is probably a gear pump which is positive displacement, and so requires full torque at all speeds, so a higher starting current for a bit longer.

The sump pump is probably centrifugal which presents no problems in starting because the load is very low (apart from inertia during acceleration) until near full speed is reached. But when it is needed, you want it to have some priority in the start queue to avoid flooding. Again, it would normally be sized to run less than half the time. If it needs more than that, you really need two pumps, or a larger one.

The fridge compressors are more interesting as the compression load is low immediately in starting due to low differential pressure, but reaches a maximum while the refrigerant balances out to its operating location in the system before dropping to normal load. The motor is probably rated for the starting load transient. Again it is a very brief period in a domestic fridge, but it’s worth being aware of that general approximation, seven to ten times current for starting, and a peak starting load of about double full load. Hardly an issue if only one starts at a time, and even an advantage to have others running when one starts, but separate systems, or allowing the probability of two or more starting together means a lot of spare inverter capacity is required.

It would also be a good idea to set up a spreadsheet with each of your users and expected run time to estimate how many kW.hr you will need to keep things running for say eight hours until you can start the generator of required. This will highlight the big users and which figures need more refinement. Remember also that you only get out of a battery about 75% of the energy you put in from the solar cells. The rest goes in heat and other losses. And you may want to allow for some lighting.

I hope that helps get you started.

Eng460
 

papaof2

Senior Member
Using a Kill-A-Watt or similar for an extended period is needed for many modern appliances so you can learn their various power requirements. We got a new fridge/freezer (Samsung double door, bottom freezer, dual evaporator) and I left the meter on it for 6 weeks. By checking at randow times, I discovered that actual power draw was between 4 watts (single circulating fan in use) and 560 watts (both fridge and freezer defrosters on). Having 6 weeks of usage, I could work out an average per-hour load to use in my spreadsheet. I did that extended measurement twice (summer and winter) and the average power is 59 watts winter and 76 watts summer. The most likely contributor to the difference by season is the house temperature - 68F in winter, 78F in summer. Other things also have seasonal variations (upright freezer, similar to fridge) and you use more hours of lighting in winter than in summer. A Kill-A-Watt or similar is also useful to find appliances that use more than their nameplate rated power. An 1100 watt microwave oven with a 1750 watt nameplate input rating actually draws closer to 1900 watts - don't want to put that on the 2000 watt inverter which is normally not loaded to more than 75-80% of its rated continuous power.

I would also suggest having a single battery bank and a single large inverter. That combination will provide the highest surge power (for motor starting) for the least investment. Be aware that the "surge" or "peak" ratings of many inverters seem to have no connection to the real world. I saw one listed as "4000 watt" but the Q&A section had a "Do not exceed 550 watts" caution. I have no idea where they came up with "4000 watts peak".

Remember that large loads discharge a lead-acid battery faster than smaller loads, as in 20 amps for 5 hours is less total discharge than 100 amps for 1 hour because of the Peukert effect. There are some interesting charts on battery performance under various loads and some of the AGM mobility batteries (wheelchair and scooter use) have detailed info for various loads and time frames. Interstate Batteries rates their DCM0100 12 volt battery as 100AH at the 20 hour rate (5 amps), 95AH at the 10 hour rate (10 amps), 76AH at the 5 hour rate (15.2 amps) and 52AH at the 1 hour rate of 52 amps. In the amount of discharge versus number of cycles battle, the 20 hour rate wins with the 10 hour rate being a close second. For the few batteries that have a 100 hour discharge rating, that rate wins ;-)

In the better inverters, you find the Outback VFXR2812A with ratings for continuous (2800VA), peak (3200VA for 30 minutes), surge (4500VA for 5 minutes) and instantaneous (4800VA for 0.1 second) power. For me, listing a time frame for surge power is what really tells you an inverter's motor starting capability. Many less expensive pure sine wave inverters list "surge" or "peak" as twice the rated power but that's meaningless unless you have the actual time that amount of power is available. I have a small (counter height) fridge that draws 55 watts but a 500 watt pure sine wave inverter can't start it, even though that inverter lists its peak power as 1000 watts. I've never put a 'scope on the fridge to determine how much power it draws for how long at startup but the 500 watt PSW inverter obviously can't provide its rated "1000 watts peak" long enough to start that fridge. A 1000 watt modified sine wave inverter has no problem starting that fridge so 1000 watts for long enough will work. Caveat emptor.
 

premelec

Senior Member
I have gone more in the direction of how to use less energy rather than supporting more - the energy flow charts shown at https://flowcharts.llnl.gov/ show what a high percentage of energy is simply "wasted" . There are 2655223 foot pounds in a KWHr and
1 watt 24/7 = 8.766 KWHr or 23.8 Million foot pounds - worth paying attention to every watt... ;-0
 

lbenson

Senior Member
Eng460--Thanks very much for the thoughts and explanations. I had not previously considered that in the round-robin scheme I'm proposing, once something has gotten through its start-up surge and is running normally, I can allow something else to start up, assuming sufficient capacity (which I should have with a 3000 watt pure-sine wave inverter, with stated allowed overload to 120% for 10 seconds and 150% for 1 second).

I will be sending voltage and current reading to my network accumulator for analysis.

Papaof2--I have a Kill-A-Watt clone, but the freezer is the only plugged-in thing I can conveniently get to. Your thoughts about summer/winter temperatures are apt--our freezer is in the unheated basement, where the temperature in midwinter is only a little above 40F--I could probably let it go for days without much worry. I'll be interested to see how it changes from the summer usage I have monitored--on for 10 minutes or so about every hour and a half.

I'm aware of the C/20 optimum usage recommendation. It's good to see that C/10 only gives you about a 5% loss (for that battery)--I hadn't previously seen that very useful piece of information.

Very interesting chart, premelec, with coal still making up nearly a third of electrical generation. I understand your desire to use less energy, but I think getting buy-in from my wife for cutbacks would be difficult (not that I think she is at all wasteful).
 

Eng460

Well-known member
It interesting to think about how you will interface a Picaxe to all those loads in order to prevent starts occuring at the same instant. For my part, I don’t think I would want to modify wiring inside the cabinet, so I would be thinking of some sort of external switching device, but I have no idea what would be available and suitable.

But it would be a “real job” for the Picaxe, and those projects which interface with external devices are always the most interesting. It will be interesting to see what you come up with. I am thinking that apart from perhaps alternating fridge and freezer, it is probably necessary to have the capacity to run at least some of the other loads simultaneously, but concentrate on preventing simultaneous starts. Still a very interesting problem.

It will also be interesting to see what your network accumulator analysis shows. (By the way, what exactly is a network accumulator? I don’t think I have heard that term before.)

Eng460
 

lbenson

Senior Member
what exactly is a network accumulator
Just my name for the two devices to which all my monitoring devices sent duplicate TCP messages. Both run the Openwrt linux operating system for small devices. One is a Seagate Dockstar (very powerful, long out of production--I have 6 of them) and the other an ancient ASUS wl-500gPv2 router.

Raspberry Pi Zeroes would do the job cheaply, but I haven't yet solved to my satisfaction the issues of power outages and SD corruption. The Dockstar and wl-500 have everything in flash--if power goes out, they just come back up when it returns.

My plan for round-robin switching is roughly like this--using these 40A SSRs: https://www.aliexpress.com/item/32880595456.html

If power company mains power is available, an SSR is turned on and 5 other SSRs are turned on, each powering one device. If mains power goes out, that SSR is turned off, all 5 individual SSRs are turned off, another feeding the inverter output is turned on, and then the round-robin algo sequentially switches in the other devices, moving to the next one if it's not looking to be powered (the sump pump very rarely is--not since April or May, I believe, and unlikely to go on in the dead of winter; the water pump runs for no more than a minute at a time less than 20 times a day).

When mains power returns, the inverter SSR is turned off, the power company SSR is turned on, and the individual SSRs are turned on.

This is theoretical at present. Any critiques/suggestions are welcome.
 

neiltechspec

Senior Member
For what it's worth.

I've used a few SSR's. Very few if any are suitable for inductive loads & will just blow straight away.
Those you cheapo ones will just go pop !. You have to spend lots for inductive capable.

On a Hot Tub controller project I did for someone (PICAXE based),I tried similar (different label, same product) & blew them straight away.
Ended up using proper mains rated relays with 12v DC coil activating 1HP water circulation pumps & 1/4HP Air blower.
Just using an SSR for Water Heater (resistive load). Unit still works after over 5 years.

I now would only consider them for resistive loads.

Neil.
 

lbenson

Senior Member
Thanks for letting me know of your experience, but the ones I tested round-robin fashion worked fine for switching a 300W work light and an 800W electric heater ( ok, resistive loads ). I'll test further with my river water pump, now pulled for the winter.
 

papaof2

Senior Member
Solid state devices and some AC motors do not play well together. I've seen people (not very knowledgable people, unfortunately) blow up an 8000 watt pure sine wave inverter by trying to start a 10 inch miter saw (brushed motor) repeatedly. You can power that saw from that inverter but you need an inline thermal resistor (NTC thermistor) to manage the near-short-circuit start-up current of 40 amps or more - a standard 20 amp breaker can handle that brief surge when powered from mains or a 5000 watt or larger generator - but you need to hold that peak down to something the electronics can handle if powering from an inverter or inverter generator. An NTC resistor of 5 ohms or so will limit that peak current: 120 volts / 5 ohms = 24 amps so that should protect a 40 amp SSR. The NTC resistor goes down in value as it heats up, so it's effectively "not there" - something like 0.1 ohm - in a few seconds. You can get adequately rated plug-in devices from raymondinnovations.com (GS10, GS11). I have their older (not on the main page) CS10 to allow using a 7 1/4 inch circular saw or a 10 inch table saw from my 1600 watt inverter generator. You can't force either saw through 2 inch wood when on the inverter gen, but a slow, gentle cut works fine and is much faster and easier than a hand saw which is something my 70+ year old body appreciates.

I have several ideas for projects using NTC thermistors so I bought a bag of 5 ohm, 10 amp thermistors on Ebay for maybe $15US. Probably more devices than I'll ever use but having extras does allow for the occasional "Oops!" moment. The manufacturer doesn't recommend using the devices in parallel to handle additional current but I don't have many things that draw more than 10 amps when running. If you have something with frequent on/off cycles, you might need to add a fan to cool the thermistor back to its higher resistance state before the next power on event (the CS10 has a thermally controlled fan for that). I can hear the difference when using the CS10 with the table saw - the saw doesn't start with the usual Bang! - it just whirs up to speed. That sounds to me as though the device might extend the life of the saw motor by reducing the abuse on start-up.
 
Last edited:

lbenson

Senior Member
Papaof2--I won't be running any mains-powered saws (or other tools) off of the inverter, LOL (if I had to cut a board for some reason, I have a battery powered circular saw which could cut (gently) maybe a dozen 2x6s before I needed to swap in the second battery (proven yesterday (and also 70+))--and there's also the generator.)

Neil--I also have a handful of these 30A relays--do you think they would be more suitable for boiler, sump pump, water pump, fridge, freezer?

 
Last edited:

premelec

Senior Member
Those relays look good for the job with 5ma drive PICAXE should do it... they do have a little idle current but should be fine in your situation - I like open fame clackers that I can close poking manually with a plastic pen ;-0
 

papaof2

Senior Member
Looking at the foil side of that board, I don't know that the tracks could handle 30 amps but they're probably OK for the loads you will have. Depending on the length of the wiring from the PICAXE to the relay board, that wiring may need additional bypassing and suppression to filter out the motor on/off transients.
 

lbenson

Senior Member
Depending on the length of the wiring from the PICAXE to the relay board, that wiring may need additional bypassing and suppression to filter out the motor on/off transients.
What would you recommend? I expect about 15 feet between relays (in basement) and picaxe (floor above). Was going to use 6-core 24AWG telephone wire but could use CAT-6. The motors themselves will be 6-20 feet from the relays.
 

papaof2

Senior Member
There's an encyclopedia on bypassing and suppression ;-) Set up one motor and see whether it works consistently. If so, there may be bypassing included on the control lines to the relay board. If not, search the forum for bypassing and noise suppression. Often easier to do on control leads at 12 volts or less than on the leads at mains voltage. I don't have a list of links at hand but I'll bet that someone does ;-)
 

lbenson

Senior Member
How critical here is zero-crossing switching of these relays? I have this zero-crossing detecting triac module which I could use (without the triac part) to enable the picaxe to switch the relays on and off at the zero-crossing point: https://www.aliexpress.com/item/32802025086.html

I have successfully used this module to test ramp-up of AC load start-up surge current with "cycle slicing", so I'm confident of the zero-detection part.
 

lbenson

Senior Member
Phil--very directly related to the conditions to be controlled--thanks.

I had seen that combination (triac + relay) recommended before, but without the full explanation. Key points:
"On high inrush current inductive loads, the silicon structure is preferred and it can absorb repetitive high current peaks. In addition, a Triac can be triggered in phaseshift mode for a better management of the inrush current"

And

"With the previously explained sequence, the mechanical relay switches when the voltage across it is low, as the Triac keeps on conducting. Moreover, the load inrush current is detected by the silicon switch, safely thanks to its high current capability. It also increases the relay lifetime, which only drives steady-state current. Finally, there is not any mechanical relay bounce."

And other good things for switching pumps and other motors with protection for relays. Very apt indeed.
 

premelec

Senior Member
FWIW I've seen more shorted solid state devices than welded relay contacts [though I've seen burned up relay contacts]. Whatever route taken consider component failures & consequences... and note that a good design lasts for a long time [my 72 year old fridge has a starter relay I keep thinking will fail and just keeps going].
 

lbenson

Senior Member
Yes, thanks, premelec. "component failures & consequences" are very much on my mind, especially since my spouse may not have much patience for failure of these critical (in terms of household functioning) systems. So I'm trying to make it activate only when there has already been a failure of the critical underlying utility mains power.

Part of the issue for me is that the most critical time of likely failure is during a blizzard. My basement is only accessible from outside the house. I think I can get emergency plug-in to the inverter power for the fridge and freezer in a 6-inch-deep mini-pantry backing the bathtub (accessible from the kitchen and with access for cable to the basement). For the fridge that would mean running an extension cord across the kitchen floor, but certain inconveniences will be part of losing power. Boiler, well pump, and sump pump offer more complicated access problems.

My thoughts about configuration continue to evolve.
 

papaof2

Senior Member
I think you mentioned sequencing the start-up of things. Could that also include a delay, such as some lights on immediately but fridge, freezer, pump or the like not powered up until 10 or 15 minutes into the failure? If you get failures as we sometimes do with the power off, power on, power off, power on, power off to stay of a self-resetting breaker somewhere or the powerlines, then perhaps you don't want anything other than lights to be powered up until you're certain it's a long enough failure to justify the switchover.

I'm being reminded to be thankful that my house (only 46 years old) has inside stairs to the basement and a small (4 circuit) transfer switch connected to the breaker panel. We still need an extension cord (an orange 75 foot outdoor rated cord, so it's visible and can be routed out of the walkway) to power the freezer but that's a minor inconvenience when the fridge, furnace and a few lights are on the transfer switch.
 

lbenson

Senior Member
Papaof2--thank you. I am programming in a delay upon loss of mains power, but so far only a minute. We have occasional flickers, sometimes enough to cause old digital clocks to blink, but I'm not sure I can remember an occasion of a 10-15-minute outage. And outages are rare in any case--we might go all winter without one. But we have had two in the past year--neither of which was terribly inconveniencing (because it wasn't 4F or thereabouts outside).

I think no harm done here if the inverter supplies the power for 10 or 15 minutes and then mains comes back on (another delay of a minute to make sure it stays on). For lighting, I'm intending to provide 12V LED lighting.
 

premelec

Senior Member
If refrigerator systems are still subject to not starting compressor under refrigerant high pressure differential [new variable speed drives may not have this consideration] - it would be best to delay start after off either if unit was running before off or more simply JIC restart delay... [JIC = Just In Case ;-0 ]
 

lbenson

Senior Member
Not certain what that means, but I think I will originally try to solve the refrigerator/freezer issue by providing each (sequentially) the opportunity to run once an hour. My fridge is a youngster compared to yours--only 20 years old.
 

premelec

Senior Member
If you unplug your unit while it has been running a while wait a few seconds and plug it in does it start the compressor immediately? Older units don't - until the pressure between condenser and evaporator has equalized so compressor has much less initial surge - otherwise an overload pops and tries to restart in a few minutes but you get big current surges trying to start too early - normally this is handled by the hysteresis in thermostat on/of temperatures...

FWIW I remind you that modern refrigerator freezer appliances likely have defrost timers and mullion heaters as well as the light that goes on when you open the door... so just disconnecting power can cause some unexpected consequences... ;-0
 
Last edited:
Top