Running an Air Conditioner From The Sun - The Holy Grail

Discussion in 'Boat Design' started by CatBuilder, Mar 6, 2012.

  1. CatBuilder

    CatBuilder Previous Member

    I had started asking about this in the battery thread, but it is better to give it a thread than to drift the other.

    Objective: To power a 3 ton air conditioner from solar panels and a large battery bank.

    Boat Description: The boat, in question, is a catamaran with a 45' LWL and a 25' beam. The deckhouse roof is approximately an 20' x 20' square, all available to solar panels. The design goals for the boat were:

    *High Performance Cruising (similar to Gunboat)
    *Comfort for Charter Guests (water, hot showers, HVAC)
    *Light Weight
    *Self sufficiency away from the dock (it will never dock, except momentarily for fuel)
    *Outboard engines have been chosen to keep the boat light and to raise them, allowing the boat to slip through the water more easily, reducing drag.
    *Lack of metal below the waterline also means no zincs to replace

    Usage of Boat: Boat is a term charter vessel, taking people on vacation for a week at a time. Charters run from November to March in the Bahamas and from June to September in the Northeast, USA.

    Electrical Design Strategy: I am using the design strategy of using the power hog (HVAC) to size the system, then adding the small items (anchor lights, etc...) later on. This is a good strategy for an initial pass designing an electrical system. It is not necessary, at this point, with the HVAC load size, to add up every tiny little item. That will come later.

    HVAC System: Based on another thread and a lot of thinking on my own, I have decided to go with a 3-4 ton split, reverse cycle air conditioning system, like you would see in a home. Outside is an air cooled condenser. Inside are 3 evaporators for different zones. They are connected by refrigerant lines. SEER Rating should be about 23 SEER or as high as possible. For the sake of the math, let's assume a 3 ton, or 36,000 BTU system. (I apologize about the imperial units).

    36,000 BTU's of AC/Heat divided by 23 SEER = 1,565 Watts of Power

    This means the air conditioner will use approximately 1.5 kW each hour. Since kVa is not often used in the USA, and I do want to stick to proper units, we now have to find out how many kWh I will use in an average day of air conditioning.

    Does anyone have any thoughts on that? My technique is to overestimate and imagine the worst heat wave, though that is usually not the case on a boat. Heat may be needed more often.

    I will just assume a 10 hour heat or air conditioning run time per day. Does that make sense? If not, we can adjust the amount after figuring out the battery and solar needed to keep this system going.

    So, in summary, the HVAC unit will use 15kWh of power each day. Sorry, Daiquiri. This is how it is stated on our electricity bills here in the States. I hope it is not terribly annoying. :D

    The HVAC unit will also never exceed 1.5KW of power demand at any given instant, except on startup, which really doesn't matter, so long as the inverter can handle the momentary rush. The battery bank can take a quick starting rush like this.

    HVAC on Boats - Real World Usage: I find that in season, I do not typically need heat or air conditioning often. Sometimes if you are in a very hot, windless anchorage in a city, you need to run an air conditioner in the evening, as the sea breeze slows down and you are going to bed. Often, it is not necessary to run it through the night and fans work just fine. Heat, if you are sailing in season, is much the same. You sometimes need to turn it on in the morning to take the chill off. A few hours into daylight, with ports and hatches closed, the boat heats up like the inside of a car. If it gets too hot, you open up the ports and hatches and allow the sea breeze through.

    Is this your experience as well in the real world?

    15kWh or 30kWh of Power? If the solar panels of my proposed system were broken, I would need 15kWh available from the battery bank to run the HVAC for 10 hours. However, there is always some light and some power coming from the panels, even on the cloudiest day. So, in reality, you would never need the full 15kWh of power each day. So how do I size this system? Do I just say, "ok, I need the 15kWh each day. This is my buffer?"

    Solar Panels Assuming I need 15kWh per day to run the air conditioner, let's go through the solar panel sizing.

    Fact: 5 hours of sunlight average, per day is what they use to calculate the number of panels you need.

    Our load here is 15,000 Watt Hours.

    15,000 watt hours divided by 5 hours is 3000 watts.

    3000 watts is the amount of power the solar array needs to produce per hour to give me 15kWh of power each day in use.

    Actual power output random "Astronergy" 240 watt panel, at 28.8VDC, is 8.13 amps x 28.8 VDC = 234 watts, each panel. 28.8VDC is chosen as the charging current because 14.4 is the charging current for a 12V system. Doubling that, you would use 28.8 to charge a 24VDC system.

    To make 3000 watts of power when the panel puts out 234 watts of power, it will require 13 panels. Let's round up to 14 panels so it looks nice on deck.

    So, 14 panels will supply a 3 ton, 36,000 Btu air conditioner with enough power to run 10 hours per day.

    Price of Panels These example panels go for $304 each. $304 x 14 panels is - $4,256. Solar Array Cost - $4,256

    Weight of Panels They are heavy. 44lbs each. I will have to see if I can find lighter ones. The total weight of the solar panels is 616lbs.

    Batteries LiFePO4 batteries come in small, 3 volt cells. They have 2000-3000 cycles in them, if put to an 80% discharge. That's a lot of days of air conditioning. To get 15kWh out of a set of batteries per day, running at 24VDC, you would need to have a 625AH bank. But, the batteries can only be discharged to 80%, so we need to add another 20% of AH's on there to make up for this. 625*20% = 125AH. Total required AH @24VDC is 750AH to run the 3 ton, 36,000 BTU air conditioner for 10 hours.

    LiFePO4 batteries I found store energy at a density of 131Wh per kg. Taking the 80% discharge level into account, if I need to store 15000Wh this means I need 136kg of LiFePO4 batteries to make a 15kWh storage bank. Or... 299lbs of batteries.

    LiFePO4 batteries I found also store energy at a volume of 247 wH per liter. This means I need a battery bank that is 73 liters in volume, or 20 US gallons in volume. This equates to 4620 cubic inches of volume, which, taken as a cube, would a a cube that is 2.6 cubic feet, weighing 300lbs. OR... ABOUT THE SAME SIZE AS AN ENGINE. Interesting.

    Cost of batteries: The cost of LiFePO4 batteries is $375 per kWh. I need 18 kWh, so the total cost of the bank would be $6,750.

    Solar Benefits

    *Total Cost - $11,000
    *Total Weight - 1,000 lbs
    *Runs FOREVER without buying fuel or doing any maintenance - no additional cost
    1 person likes this.
  2. hoytedow
    Joined: Sep 2009
    Posts: 5,828
    Likes: 385, Points: 93, Legacy Rep: 2489
    Location: North of Cuba

    hoytedow Carbon Based Life Form

    How much weight will all those additional solar cells add to your boat?
  3. CatBuilder

    CatBuilder Previous Member

    Please read my post. That question is answered already in bold in the post.
  4. hoytedow
    Joined: Sep 2009
    Posts: 5,828
    Likes: 385, Points: 93, Legacy Rep: 2489
    Location: North of Cuba

    hoytedow Carbon Based Life Form

    Oops! I must read more carefully. My humblest apology.

    "616 lbs." Sorry I missed that.
  5. rwatson
    Joined: Aug 2007
    Posts: 5,873
    Likes: 311, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    I notice that you don't do a detailed analysis of the alternative systems. The cost of alternate systems has the advantage that if you don't have any charters, you don't have to buy the fuel - where solar means you have all the costs upfront no matter what business you do in the next few years.

    And as far as "runs forever", Solar Cells have a limited life (their efficiency drops off over the years) and batteries will only last so long. (maybe as low as 4 years, depending on usage)
  6. CatBuilder

    CatBuilder Previous Member

    It's ok. I got all testy and wrote a snippy response for a moment, but I deleted it. Was still torqued a little from doing all that work in the above post. :D

    I think I can possibly find lighter weight panels. I am assuming it's the frames that weigh so much @ 44lbs each.

    However, a 5KW generator (to handle starting current and other loads) will use 1 gallon of gasoline an hour. (I am stuck with gasoline - outboards)

    So, for me to use a generator, I need to either have a very large battery bank and inverter coupled to a 5KW generator that runs for 4 hours (assuming charging losses) or a 5KW generator just running for all 10 hours which is more likely.

    That's 10 gallons of gasoline per day - or $37 in fuel *per day*! Not to mention 70 lbs of gasoline per day.

    Want to run this for 5 days? That's my entire 50 gallon gasoline tank gone and 350lbs of fuel aboard, plus the 120lbs generator.

    Certainly, it weighs less to use gasoline, but then if you start to add in the house battery bank, plus the solar you need for other things, I think the weight starts to even right back out.

    Only difference? $0 for 5 days of air conditioning with solar, $185 for 5 days of air conditioning with a generator, plus oil changes and engine breakdowns.
  7. CatBuilder

    CatBuilder Previous Member

    Actually, we do still have to pay. We live aboard 24/7/365 at anchor. So, we still need the fuel and power generation capability.

    Ok, yes, my English is not perfectly accurate. That is not the point. How long does a gasoline generator run if it is used for 10 hours every day? Certainly not 20 years, as an average solar panel does. The gasoline generator would run more like 200 days if used the same way the panels are. They have warranties to 20 years on all the good panels.

    Batteries? Again, please re-read the first post. The batteries are capable of 2000-3000 cycles to 80% discharge. How many 80% discharges do you get in a single day if you run the air conditioner above for 10 hours?

    Ok, I'll tell you: 1 cycle.

    So, if you run the air conditioner every single day (not likely) and you don't have any sunshine coming in to take up some of the load of the air conditioner, and you discharge 1 full cycle each day, how many years is 2000 days?

    That's 5 1/2 years before the batteries start getting tired, running that air conditioner every single day. Odds are, it runs only 30 days out of a year.

    You may not be aware that when you have a battery bank hooked up to a solar array, it is not even cycled most of the time. It just sits there, fully topped off. If the air conditioner takes 1.5KW at any moment and the solar panels are putting out 3000 watts of power when it is running, there isn't even a single cycle. ZERO battery life gone on a sunny day like that.

    I should probably mention at this point, that my last boat ran entirely off solar. There was no other power running my refrigerator, freezer, computers, navigation, phones, lights, etc....
  8. CatBuilder

    CatBuilder Previous Member

    I'm hoping this thread can go the way of the HVAC thread or the "drag of a saildrive" thread.

    Less opinions that have no numbers in them and more numbers, please.

    Thank you in advance.
  9. Brian@BNE
    Joined: Jan 2010
    Posts: 262
    Likes: 13, Points: 18, Legacy Rep: 151
    Location: Brisbane, Australia

    Brian@BNE Senior Member

    OK, first some numbers. My home solar array is rated at 1.5 kW (9 x 167 W panels). In just under 4 years it has generated just over 10 MWh, average is 7.06 kWh/day.

    On a boat in Brisbane I would not have got as much, for two reasons:
    1. Ideally panels should be normal to the solar incidence. With fixed panels, general rule of thumb is to incline at an angle equal to half your latitude. Boat cabin tops are effectively flat surfaces - a significant loss will occur due to sub-optimal panel orientation (reflection).
    2. A lot of panels are impacted by even partial shade, generating nil. My roof was never shaded, but your boom and maybe sails will cause shading problems.

    These factors could easily reduce performance by 50%. I don't think you have anything like enough roof area for panels to be fully solar. But you can top up with genny.

    I'd use the panels with the highest wattage/sq m you can get, that can tolerate partial shading (forget the terminology, but it also involves MPPT controllers) and fill the roof up. You will need a good battery management system for the LiFePO4 bank anyway. Keep a frequent watch on it, run genny to supplement battery state.

    If you are only using solar for heating/cooling, then look into solar thermal options. Probably greater efficiency than solar PV.
  10. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 255, Points: 93, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    There's an error here, Cat. I believe you have misunderstood the concept of Depth of Discharge (or maybe I have misunderstood your calculations). DoD is explained well in this page:

    Batteries can last up to 2000 cycles if they are discharged down to 80% of their capacity. If you drain them down to 20% of their capacity, you reduce their lifetime to just around 600-650 cycles. In other words, in order to get a lifespan of 2000 cycles, you can use only 20% of the battery capacity. If you drain each time 80% of the stored energy, the battery will last 600-650 cycles. That's a decision you'll have to do.

    If you opt for 2000+ cycles, then the battery bank capacity is obtained by multiplying the required Amp-hours by 5 (or dividing it by 0.2)

    I'll read the rest of your numbers with more attention tomorrow, and see if I can be of any help. :)

  11. CatBuilder

    CatBuilder Previous Member

    Thank you, Brian. This is good.

    Where is your home? What latitude? Do you have winters or rainy seasons? My boat will be in the "summer" at all times. Summer in the north and summer in the south. That might make a little difference on the average.

    Also, it is not necessary to generate the 15kWh per day I am using as an example unless I am running the HVAC every day. As I was saying, usually it's only run for a bit in the evening going to bed or in the early morning. Just a few hours here and there until the outside temperatures are more comfortable. I could run it for 10 hours straight through the night if I wanted, but then I would have to wait the next day while the batteries charged back up... or possibly wait two days. Typically, you would not run the system every single day. Just on peak times when it's too hot or cold. Our boat will never see winter.

    Again, these points are true. Had this experience with my last boat. I could pull the topping lift and get the boom out of the way to leave the panels unshaded for the majority of the time. I know this is an issue and it was on my last boat. I had to be careful about that, to some extent, moving the boom around during the day if at anchor. If sailing, there are some shadows and some losses, but it is usually made up by motoring in and out of port anyway. Still, it's a factor to consider.

    Absolutely. I was looking at 240 watt per panel and looking at having 14 of them from the calculations above. I do know what you mean about the kind that tolerate partial shading. I will definitely try to find the lightest weight ones I can that will tolerate partial shading if the numbers all work out here. I had an MPPT on my last boat (they were just coming out) and it made a world of difference. Great charge controllers.

    I plan to go without a generator. That is the whole idea of this thread.

    The elephant in my boat is the HVAC system. The entire boat could run of much *much* less solar if I didn't need to keep the guests cool and warm.

    I did look into solar thermal options, but they don't exist for cooling. Any thoughts there?
  12. CatBuilder

    CatBuilder Previous Member

    No, you have not misunderstood my figures or the wording. What is not making sense to you (I think) is that you are thinking of lead acid batteries. LiFePO4 batteries are completely different. Let me say it in two ways, to make sure we both understand the LiFePO4 life cycle.

    1) At 80% DOD, you can get 2000-3000 cycles from a LiFePO4 battery.
    2) At just 20% of C left in the battery at the end of discharge, you can get 2000-3000 cycles from a LiFePO4 battery.

    These are not lead batteries. They behave very differently. These are the batteries used in modern electric and hybrid cars.

    I look forward to any input you feel like having. You were a great help on the HVAC thread and your numbers were spot on. Thank you again for that.
  13. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 255, Points: 93, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    I'll give you a real-life example, Cat: my iphone is regularily dicharged down to 15-20% of the capacity (which gives 80% DoD) and now, after two years of use, the battery lifetime is roughly half what it was when the device was new. In the beginning I used to recharge it once every two days, now I have to do it nearly each night (though I am using it much less now). So we are talking about some 500 cycles here and not more than 65-70% of the retained capacity up to this moment. By retained capacity I mean the percentage of the original (new battery) capacity which the battery attains after a full recharge.

    So, based on this and on what I know about battery technology (but I'm definitely not an expert) your estimate and numbers sound way too optimistic to me. Can you please provide some references for the 80% DoD, 2000 cycles data? It is important to understand well the retained capacity criterion used for estimating the number of cycles at each DoD. for example, my iphone will probably arrive at 2000 cycles if I accept the fact that it will have some 30-40% of the retained capacity at that time... :)


    I am not trying to dismiss your numbers or blow up your dreams. Just trying to save you from a possible 300-400% error in capacity calculations. If you prove that I'm wrong, so much better both for you and for me (I will have learned something new). ;)
  14. Brian@BNE
    Joined: Jan 2010
    Posts: 262
    Likes: 13, Points: 18, Legacy Rep: 151
    Location: Brisbane, Australia

    Brian@BNE Senior Member

    I am in Brisbane, latitude 27deg25min South. We get wet summers and dry winters.

    Seasonally over the 4 years (kWh per day) my highest was Spring 2009 at 8.6. My lowest was Autumn 2011 (La Nina, year of floods...) at 6.45.

    I am a bit surprised by the narrow range. Yes, we do have a lot gorgeous blue sky days in winter. But I expected summer to be better. Possibly its the panel inclination as well as weather. I just put them on the roof, which is at about 36 deg. They probably should be flat in summer here, but no doubt I benefit from the angle in winter.

    With your 2 summers and flat angle you could be in good shape.

    As far as solar thermal, I can't point you to anything that would work for a boat. A couple of country towns here have a dish system which tracks the sun and generates most of the town's electricity needs. Not practical for you. I recall a Sydney hi-rise office building converting some solar energy into ice, stored in the basement which was then used for cooling. Only had a quick look for this, but can't find any details of it. No recent news might mean it was a dud. Probably needs too much space for you also.

  15. Squidly-Diddly
    Joined: Sep 2007
    Posts: 1,810
    Likes: 140, Points: 63, Legacy Rep: 304
    Location: SF bay

    Squidly-Diddly Senior Member

    bit off topic, how about ice-based cold-ness storage?

    1)always good to have lots of ice on hand, for all sorts of things.

    2)ice is made of fresh potable water, and you can never have too much of that on a boat.

    3)the 'coldness' stores pretty well in ice, and for over a week in a king sized ice-chest.

    4)a heat(coldness) conducting container, or two, full of ice works pretty well to cool off smaller spaces like a boat.

    5)you can release the coldness of ice either very slowly, or quite fast.

    I've found 1-gal plastic milk jugs will survive several freeze/thaw cycles.

    In Mexico they used to air-condition "1st class" trains by dumping ice on the roofs which had curbs to keep the ice from falling off.

    Have a freezer you can run full blast when docked or in bright sun, and also load the boat with ice you've made at home at start of each charter.

    Take about 5-10 one gallon jugs out and place them on an aluminum trough on the roof with enough sea water to create 1/4" water flow around them, and have trough's aluminum conduct the coldness into the cabin. Probably need a quick evaporating cover(ceiling) to handle the condensation. Make it out of athletic clothing fabrics.

    At the end of the day, use the now thawed water in the jugs for your fresh water.

    I think this might work for 1-week charters, if clients are OK with lugging the 1 gal jugs once or twice a day.

    I've noticed you normally have 20% of the people who 'just can't take the heat' and having 1 gal jugs of ice would allow them to cool their cabin or persons as much as a body could stand, while not requiring full boat super-cooling.

    PS-gallons of milk can also be frozen rock solid and used as the 'ice' for ice-chest and the milk used as it melts.
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.