Solar powered Pond pump

[eclectic]

Background.

I have built an above-ground pond of c. 1500 litres and I’m setting it up.
I run two Solar panels, each powering a small pump feeding a small biofilter.

I’d like to run the pumps when it’s dark.
I have several 12v Lead acid batteries from 80Ah to 7h Ah,
which can be fully charged indoors.

I own a a fairly well-stocked toy-box
and I’m happy to buy anything else I might need.

Anyone got any experience or ideas, before I start blundering about?

e

 

[papaof2]

A: How much power does the pump draw? I have a Flojet 12 volt pump that delivers one GPM (3.8 LPM) on one amp. I have a Rule 12 volt pump that delivers 4.6 GPM (17.5 LPM) but it needs 4.5 amps.
B: Hpw many hours/day will the pump be running?
From A * B you get the amp hours needed per day. Work out how many hours the sun will be available to power the pump* and how many days/month you have that much sun. Then you can determine how many hours of battery power you need for the pump each month. If your batteries are flooded lead-acid, don't exceed 20% discharge for longest life (10% is even better). If your batteries are AGM or gel, don't exceed 50% discharge for longest life (20-30% is even better). Gell batteries can give 1500+ cycles to 30% discharge, about half that many to 50%. AGM batteries are probably 1100 and 500 cycles, respectively (to their defined "end of life" which is 80% of the original capacity). I have gotten almost 9 years out of a set of AGM batteries in solar service but I watch the discharge/recharge levels carefully (Amp hour/Watt hour meter in the line with the batteries to know how much power has been removed/replaced).
If your solar panel(s) can provide power for the pump during the day, you need enough additional solar capacity over the pump's needs to recharge the battery from several days' use in one day's sun. The daily solar power is NOT a constant figure - I get, at best, 3 hrs/day winter and 5hrs/day summer - and you will have clouds and/or precipitation some days :-( The battery bank should be large enough to handle your typical number of consecutive days without sun. The solar power should be adequate to recharge that much use in one day of sun - you could have four days rain, one day sun, four days rain.
*Go to http://www.solarelectricityhandbook.com/solar-irradiance.aspx to check your location's daily useful solar hours by month, the direction the panel(s) face and the tilt of the panel(s).
The proper solar charge controller could be a lifesaver for your batteries. If the charge controller's Load terminals can handle the pump's power draw, the Load terminals could be set to turn off when the battery drops below a specified voltage (12.2-12.3 vlts) so the battery will not be excessively discharged.
I like the controllers from EPEver/EPSolar, whether PWM or MPPT. They have a wide range of adjustment to match your battery's parameters (boost charge voltage, float charge voltage, load shed voltage, etc.). A few of the "generic" charge controllers also do (banggood.com once sold a very good PWM charge controller for under $20 - SKU300186 - but it's no longer on their website).

Unless the pump(s) need to run 24/7, a PICAXE could control one hour on every four hours or whatever is needed.

 

[eclectic]

Thanks John.

I'll start experimenting tomorrow.

 

[papaof2]

If you want to play with the numbers, there's a basic spreadsheet here: http://jecarter.us/files/My-Small-12Volt-Solar-Pump.xls

 

[eclectic]

I have just done some testing.
Current draw ~ 0.41 A per pump
I would prefer to run each pump 24/7 to maintain the filtration medium.

Looking at the weather forecast,
I think it might be a case of mains charging then switching.

 

[papaof2]

A 100AH AGM battery (max of 50% discharge level) could power two pumps for 2.5 days (60 hours) without any sun. 200 watts of solar panels should keep the battery charged all year. If you have sunlight/daylight most of the time, 100 watts of solar might power the pumps perhaps half the daylight hours and provide some charge to the battery during the peak sun hours (solar noon +/- 2 or 3 hours). 200 watts of solar should be "set and forget" unless you have extended coudy weather. I know from experience that there are days when the best a 250 watt panel can do is 5 watts or less :-(
US solar noon map: https://www.esrl.noaa.gov/gmd/grad/solcalc/ . I haven't tried non-US lat/long figures but the time calculations should work for wherever you are.

 

[eclectic]

Thanks again.
I'll investigate tomorrow.

 

[inglewoodpete]

.... If you have sunlight/daylight most of the time, 100 watts of solar might power the pumps perhaps half the daylight hours and provide some charge to the battery during the peak sun hours (solar noon +/- 2 or 3 hours). 200 watts of solar should be "set and forget" unless you have extended coudy weather....

A couple of factors that are likely to affect the performance of solar systems in the UK. As a (naive?) Australian landing in Dublin (similar lattitude) at midday in mid-winter a few years ago, I was surprised at how low the sun was in the sky (less than 14 degrees above the horizon). So, winter will need more solar. The other factor is the number of days where the sky is mostly, if not totally overcast. Blue skies are not all that common for many months of the year.

From my experience in sunny Oz, 200 Watts of solar panel will not be sufficient for 100Ah AGM in the UK. Also, I'm not sure on how freezing conditions will affect the overall system, particularly the battery/s.

These questions would probably be best answered on UK, EU or Canadian forums.

 

[papaof2]

http://www.solarelectricityhandbook.com/solar-irradiance.aspx can provide the useful hours of sun by month, by panel direction. London's worst and best cases are 1.05 hours in December and 4.25 hours in July. You're correct - that might need 500 watts of solar in December. As you go North from London, there will be even less sun. A charged AGM battery should be OK down to -5F (-20.5C) - check the manufacturer's specs for the batteries you are using.
Someone who has been there and done that will have better knowledge of the edge limits.

 

[eclectic]

Thanks both for clarifying my thoughts.
It's 08.00 here and a dull grey sky at my home 53' North.

The Forecast for the next two weeks shows cloud
with highs of 16'C and lows of 6'C plus variable wind. : -(

I'll use my A3 sized panels for "bonus" power,
but rely on battery power otherwise.
I'll work on using the Picaxe to switch between sources.

e

 

[AllyCat]

Hi,

I'll use my A3 sized panels for "bonus" power,

Yes, and probably quite a small "bonus", for the reasons outlined by Pete. An A3 sheet has an area of 1/8 square metre, so two will give 0.25 m2. The efficiency of PV panels is close to 20%, so the "Solar Insolation" figures from post #2 need to be divided (also) by 5.

About 6 months ago we had domestic panels installed on our (almost optimally located) roof, primarily to gain some real experience to confirm (or disprove) my prejudice that the UK is NOT the most worthwhile place to install solar panels. IMHO Solar Panels are far more suited to produce power to run Air Conditioning rather than a Heating System! However, there aren't many other practical urban/domestic options to generate "Green" energy and the recent enormous increase in fuel prices has now at least made the economics look more sensible.

Of course any Solar Energy generation has a "daily" cycle, but much more significantly a yearly cycle, which at latitudes above, say, 50 degrees reduces the peak December (winter) generation to around 25% of the summer rating. But even worse is the "cloud" cycle (in our Maritime Climate), where there can be periods of many days, or even weeks, when little or no useful energy is generated. My ballpark estimate is that full cloud cover, but still quite "bright", gives around 20% of the direct sunshine value and a "dull" or "overcast" day (let alone raining or snow) will be less than 10%.

To give some actual numbers: Our 8 panels (as many as would fit on a rather inconveniently-shaped roof) total about 14 metres2 to max at about 2.5 kWatts, i.e. more than 100 times an A3 panel. Generally their production has been within a few percent of monthly predictions, but last December was particularly "unfortunate": The whole month generated only 43 kW.hours (compared with just over 100 kWh predicted), with just 8 kWh during the whole of the last two weeks, and some days below 100 Watt.hours.

The system also has a battery pack, in particular for "surge" use, such as boiling a Kettle, Microwave Oven or Washing Machine, etc., if the sun isn't shining brightly, and also "overnight" (assuming of course that there was a surplus during the day). The Lithium Iron Phosphate (LiFePO4) cells are rated for up to 6000 cycles at 80% - 90% DOD, but if I understood the manual correctly, their "protection" system shuts them down totally below zero degrees C!

A calculator/tool that I've found useful is HERE, which includes options for all locations and orientations, on/off grid and even steerable panels, etc. (I still have some plans for a PICaxe M2 solar-tracker program). It appears to include cloud cover data because the predictions (Click on "Visualize Results") for Brighton are significantly higher than for Southampton, although both are local south-coastal towns with a very similar Latitude (51 degrees).

Cheers, Alan.

 

[erco]

Every time I consider an outdoor solar project, it makes me appreciate mains power all the more. Cheapest, simplest power anywhere. I also drive an EV, which is infinitely simpler (and far cheaper to operate) than gas cars. I've definitely become an EV snob. That said, I'm busy restoring my first car, a '67 Corvair and can't wait to get that gross polluter clattering down the road... for a small percentage of my driving.

 

[julianE]

That said, I'm busy restoring my first car, a '67 Corvair and can't wait to get that gross polluter clattering down the road... for a small percentage of my driving.

why not make the corvair electric, the extra weight might actually make it drive better.

 

[lbenson]

Every time I consider an outdoor solar project, it makes me appreciate mains power all the more. Cheapest, simplest power anywhere.

True that. I'm looking to be able to provide up to two days worth of battery backup for my boiler and water pump in not-so-sunny (in wintertime) Nova Scotia. Best price I've found for LiFePO4 battery (not individual cells) without charger is about $300US per kilowatt hour. In Nova Scotia, a kilowatt hour costs about 13 cents Canadian, bit more than a dime U.S.

Worth it for me for the convenience and the learning experience.

Also looking to provide some solar for the learning experience. Perhaps surprising to some--Milton, Nova Scotia, at 44 degrees latitude is well south of Eclectic--south, in fact, of Genoa.

 

[papaof2]

Per the computations at http://www.solarelectricityhandbook.com/solar-irradiance.aspx, Halifax gets 2.27 useful hours of sun in December and 4.9 useful hours of sun in July. With Milton being at 44.08236 and Halifax being at 44.69473, Milton would get a bit more sun than Halifax.
I've found the solarelectricityhandbook site's numbers to be comparable to what I've measured in backyard sun for my location (southeastern US) - but they do NOT include a "fudge factor" for the average number of cloudy days.
I started dabbling with solar about 5 years ago - also "for the learning experience". I've since built a "big UPS" as my "Wait until daylight" version of a "solar generator" (420AH of AGM batteries, 2000 watt pure sine wave inverter, 8 to 20 hours of limited backup power for fridge, freezer, blower on gas-fired central heat, a little LED lighting and charging phones). Walking down the steps to the basement to flip a few breakers sure beats trying to get a gas generator started during a thunderstorm or shoveling 7" (18cm) of snow at 30F (-1C) to get that generator out of a shed and then trying to get a cold generator started (been there; done that - but only once: it was incentive to find a better way ;-) Along the way, I've discovered several solar charge controller designs that use a PICAXE - it's an incredibly versatile little device.

 

[lbenson]

Yes, SolarMike's PICAXE solar charge controllers are pretty nifty. Not having to start a generator in the worst of conditions is also a factor for me.

So by your numbers, the two (pathetic, but probably accurately rated) 100 watt solar panels I have might give me a kilowatt per day in July (which is usually very sunny). Certainly enough for good 12V lighting after dark. I'll be looking for larger (440W+) panels, but am having trouble finding a vendor in Nova Scotia (effectively in Halifax) who will sell just a few panels (I only have a good spot for 2)--otherwise, shipping is a killing cost.

 

[papaof2]

Do you have craigslist.org in your area? I've found used 300W panels there for reasonable prices - even with a 40 mile drive one way. I don;t know if anyone is offering larger panels in this area as I haven't checked that section of Craig's List in recent months - I have all the panels I have space for :-(

My experience with different sized solar panels is mixed, with a number of new "100 watt" panels never delivering more than 75 watts in backyard sun: Grape. Sonali, Eco-Worthy. The Mighty Max panels did a bit better. I have a pair of HQST 100W panels to test when my back is up to installing them above the garage door. The advertised panel ratings are their STC (Standard Test Conditions) rating - in the factory, controlled lighting, panels in air conditioing at 25C - solar panels lose capacity as they heat up and dark panels will always get hot in the sun. The NOCT rating (done under more realistic conditions) is about 73% of the STC rating. My solar spreadsheets typically have both numbers but I plan my available solar power based on the NOCT ratings. The solarelectricityhandbook site shows fewer useful sun hours than most of the sites selling solar power but I think it reasonable to build on the pessimistic side - I can't run down to the local "solar station" and load "2 hours of sun" in the back of my truck when needed :-(

The exceptions I've found to the "never makes rated power" rule are some used 250 watt Jinko panels which delivered 235 watts in backyard sun (measured at the output of the MPPT charge controller). Those are probably the "most for the money" panels that I've tested.

You mentioned LiFePO4 batteries. Typically, they can't be recharged below 32F/0C and a good BMS will enforce that limit. AGM's are bigger for the same capacity but can be charged at 14F/-10C and discharged at -4F/-20C. Some of the higher end AGM vehicle batteries (Odyssey) are rated for -40 (one place where F and C agree ;-) but only 400 cycles to 80% DOD. AGM deep cycle batteries may not have that temperature rating but do have lots more cycle life. Lots of tradeoffs in choosing batteries for solar use - where you are is a factor plus how deep your pockets are. I have solar powered lighting in an equipment shed out back. It uses an AGM battery even though I could build a LiFePO4 of similar capaicty for about the same price. Why AGM? We get temps of 20F or lower in the winter and I want the lights to work whenever I turn the switch (spring wound timer so I don't forget and leave the lights on).
You need to have the full datasheet for any battery you're thinking about using.

 

[inglewoodpete]

So many solar panel 'specifications' make claims of unrealistic wattages. Yes, I've been caught out too!

Regardless of the latitude of where the panel is to be used, you need to be sure of its actual output capability in ideal real-world conditions.

What I've learned is a 'rule of thumb' you can expect: 155W per square metre* from solar panels. (Duds and low efficiency ones excluded). This assumes that the panel can get 1kW/square metre irradiation*, so generally must be located in latitudes between between 0 and +/-30 degrees. Ie 15.5% efficiency.

Of course, you should measure the solar panel area, excluding the border - unsurprisingly, the aluminium frame does not contribute to the output! So, if you're looking at an advertisment or discussing a panel with a salesman, always determine the actual size, which is typically 50mm or more smaller than the overall size. You will quickly discover that there are a lot of unrealistic specs, ignorant sales people and outright scams!

 

[papaof2]

My "learning experience" included a lot of reading datasheets and comparing real world performance to advertising claims. The Jinko panels I mentioned before included a NOCT rating on the datasheet - the 250 watt panel lists 183 watts NOCT. If you have limited sun, whether from your latitude or your neighbors' trees, knowing the worst case in full sun can help you make the best choices for your needs.
I've seen multiple listings for the "useful sun hours" at my location but the ones from the sellers/installers of solar power are inflated by 20% or more over those from the solarelectricityhandbook site. And some list the "yearly average" sun hours - claiming "4.9 hours average sun" does not improve a real world number of 3.2 hours of useful sun in December. Maybe those solar claims are written by the same people who rate banggood.com's electrical devices? Such as a "4000 watt peak" inverter that lists "do not exceed 300 watts rated power" in the fine print? Or their solar panels that were rated by multiplying the open circuit voltage (VOC) by the short circuit current (ISC) and the person doing that not knowing that those two things can NOT occur at the same time? I guess ignorance is bliss, as long as it sells products...
I remember all the hype about CB in the US in its early days but I never promised anyone more than "At best, you can get this far from vehicle to vehicle and that far from vehicle to base station." No "You have a private channel." or "You can talk 50 miles between cars." That hype now seems to have infected some sellers of solar power. You've seen the ads for "Get paid $2000 for installing solar power"?

 

[papaof2]

Another site to make calculations at - someone not in the business of selling solar ;-)

PVWatts

Estimates the energy production and cost of energy of grid-connected photovoltaic (PV) energy systems throughout the world. It allows homeowners, small building owners, installers and manufacturers to easily develop estimates of the performance of potential PV installations

pvwatts.nrel.gov

You will need your lat/lon as the site doesn't know where Milton is - just Nova Scotia. However, it does include weather as a factor in computing solar.

 

[lbenson]

Thanks for the link. (Just got back to Nova Scotia day before yesterday.)

 

[erco]

Dude claims lithium sulphur batteries are the way to go, although more research is needed to explain an "accidental breakthrough".

 

[papaof2]

Remember when the "salt water battery" was the "next great thing"? They even got those into production for a little while.

 

[lbenson]

Cold fusion, anyone?

 

[erco]

Over Unity forever!

 

[Buzby]

These 'new' battery techs have been around for years.

The 'salt water battery' is light-weight, has very high energy density, but a limited life span. This is why it has been for years the perfect choice for 'money no object', one-shot applications, such as torpedo propulsion motors.

I remember the 'sodium/sulphur' battery, it's downside is the sodium has to be molten. A highly reactive, hot liquid metal electrolyte is not something conducive to consumer goods, yet.

Regarding 'over unity', isn't that what a heat pump does ?. You put 1kW in and get 3kW out.

I really do believe there are technologies that will give us 'unlimited' power, but we don't know what they are yet.

 

[erco]

I really do believe there are technologies that will give us 'unlimited' power, but we don't know what they are yet.

 

[papaof2]

Peanut butter and dark chocolate?

 

[julianE]

I bet there will be a "one weird trick" that will allow a car to drive coast to coast on a single lithium sulfur battery.

 

[papaof2]

Search for pictures of the "vehicles" the gasoline companies used in the 50's/60's to get incredible mileage. Aluminum tube frame, spoked bicycle wheels, etc, etc. Might do even better with magnesium or titanium and they're even lighter (at several times the price). Roof the vehicle with 1500-2000 watts of thin film solar panels (2.0 to 2.7HP). Run during the sunny hours on the solar panels only and maybe charge the batteries some from the sun when on level stretches of road or use regenerative braking for charging on downhill stretches of road.. Run on battery to the next place in battery range that's likely to have good sun the next day or wait for the next day's sun before starting out again. Just ignore the fleet of large, diesel-powered support vehicles that trail behind the test vehicle ;-)

 

[lbenson]

@papaof2--works till you get to the Rockies--then, I don't know.

 

[AllyCat]

Hi,

Roof the vehicle with 1500-2000 watts of thin film solar panels (2.0 to 2.7HP).

Spot on with the watts/hp conversion, I still remember 1hp = 746 watts and = 550 foot.pounds/sec from my schooldays, but I think you are rather (or very) optimistic with the Solar Panels. AFAIK, thin film solar sheets (aka Amorphous Silicon) are still significantly less efficient than Monocrystalline or Polycrystalline silicon types of panels, partly because they are less effective at converting the Infra Red frequencies that represent around 50% of useful sunlight.

Even the multilayer ("3 colour") panels (for example here appear to have typically around 6% conversion efficiency and a modern commercial product seems to be nearer 5%. I know American cars are typically larger than ours, but 40 square metres or almost 400 square feet of "roof" for the 2kW ? Of course those efficiencies are not actually quoted in the promotions, but they can be easily calculated from the specified areas and output ratings. These usually assume 1,000 watts/m2, which even with clear skies falls away when the sun moves by more than some tens of degrees away from "Overhead". Of course the sun moves by about 15 degrees/hour, unless you can travel (west) at 1,000 mph.

A typical EV gives around 3 miles/kWhr, which could of course be increased by using "bicycle wheels" and other high tech. methods, but a "passenger" might weigh 200 pounds and a 10kWhr battery much the same again . Incidentally, I found it amusing that an Electric Bicycle is technically more efficient (Greener) than a "Pedal" bicycle, because the conversion: Sun -> Electricity -> Motor -> Wheels is potentially more "Green" than Sun -> Food Production -> Human Muscles -> Pedals -> Wheels.

Cheers, Alan.

 

[papaof2]

There are currently some claimed 20% efficient panels. Still not what you'd need for a solar panel driven road vehicle. And you'd still need pedals and a chain when the weather is clouds/rain/snow for a week.