Looking to power an Observatory

asteroids
asteroids Registered Users Posts: 17
This is an off-grid site that I'm looking to power. Initial measurements show that it consumes 48V@4Amps and 12V@8Amps all night long. Not too much power during the day. I'd appreciate suggestions on a design.

Comments

  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory

    welcome asteroids.
    this is somewhat different having an observatory requesting advice for solar. now i do have to ask if the things being powered are critical or not and if you have backup like a generator? it would stand to reason you already have something there to power these items and is most likely a generator.

    you should get a more precise time that these loads are on for and account for worst case scenarios of little to no sun. that would also mean little to no stars for you too.:cry: you should also need to elaborate on how many days of backup you'd like to have if applicable. setting up a reliable system in theory is easy, especially on paper, but are there also cost constraints going on here as this can get real expensive very quickly?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Looking to power an Observatory

    Welcome to the forum.

    There's a missing bit to the equation, and that is time. Coming up with 48 Volts and/or 12 Volts isn't difficult.
    48 Volts @ 4 Amps for one hour is 4 Amp hours @ 48 Volts or 192 Watt hours. You will need to accumulate all the power you need at night in approximately a 4 hour window of sunshine during the day and store it in batteries.
    With the inevitable systems losses, for that 48 Volt load you'd need battery of at least 8 Amp hours @ 48 Volts for every hour of intended operation and about 50 Watts of panel to recharge each "hour's worth".
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Looking to power an Observatory

    Look for threads started by poster "Roy Salisbury". He has a nice little Observatory out of the Las Vegas area...

    See if this Forum Search works:

    Find latest started threads

    Also, a link to Roy building his observatory:

    http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/3568332/page/0/view/collapsed/sb/5/o/all/fpart/1/vc/1

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory
    Welcome to the forum.

    There's a missing bit to the equation, and that is time. worth".

    For Astronomy useful "Night" is about 12hrs in the winter time. About 6hrs in the summer. Center around midnight.
    So, for Solar useful "Day" is about 4hrs, 2hrs either side of local "noon".

    Should I build a 48V system and do a DC/DC to get the 12V I need or try something else?
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory
    asteroids wrote: »
    For Astronomy useful "Night" is about 12hrs in the winter time. About 6hrs in the summer. Center around midnight.
    So, for Solar useful "Day" is about 4hrs, 2hrs either side of local "noon".

    Should I build a 48V system and do a DC/DC to get the 12V I need or try something else?

    as for the 12v, yes, a converter can be used off of a 48v battery bank. i still would like to know if all of these items actually are 12v and 48v dc loads? how are you achieving the power for them now?

    i will do a rough here for you. the 12v current of 8a will have that 8a plus the loss in the conversion and i'll give a rough estimate of 80% efficiency. that is now a 10a draw for the 12v operations and a total wattage of 120w for that area of load. using the minimum of 12hrs of load in the winter this translates to 1440wh or a draw of 2.5a at 48v for a total of 30ah over 12hrs.

    now for the main 48v load. with 4a drawn over 12hrs this is now 48ah at 48v or 2304wh. i'll address the ah here as batteries are based on that rather than wh. adding the 48ah with the 30ah from above gives 78ah total. now you never want to exceed 50% dod on any battery bank so the capacity will minimally be twice the calculated ah for 156ah min. this would only be good for a day and you did not say how critical the loads are or how many days you may need for backup due to inclimate weather, but 1 more day i would add as nobody has every day sunny. this doubles the ah to 312ah and would be a minimum for the battery bank.

    to charge the battery bank would normally go between 5% and 13% for a charge rate, but i'll recommend a high charge rate for better recovery to get the batteries up to full in 1 day from 75% soc from draining the bank down by 25% dod. this means agms like the sunxtender series as they can tolerate a high charge rate. anyway, i will assume about 2.5hrs for the peak solar time in winter (didn't look it up for your area of canada, but will assume it isn't far off from mine) 25% of 312ah is 78ah. over 2.5hrs this is 31.2a delivered. to roughly translate that into pv one must account for the inefficiencies of them and we usually go with an efficiency of 77%. 31.2/.77=40.5a at 48v nominal from pv. you will need a higher voltage than 48v and about 57.5v should be good. 57.5v x 40.5a = 2329w stc of pv as a minimum.

    now this is not written in stone as you may not need it to go for days, but rather hours until say more fuel for the generator can be had for example. i ran through a very rough calculation of how it could look. of course the system would involve far more as a good mppt controller would be in order and all electrical items like combiners, disconnects, wiring, fuses/circuit breakers, etc. etc. etc. if you intend any ac power through an inverter for other loads like lights then we would need to revisit what i did to add this to it too.

    this is pretty much why we ask for more info as we have a difficult time making some assumptions and when we ask for more info it should be given or we are taking possible wrong stabs at what you may need or want.
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory
    niel wrote: »
    as for the 12v, yes, a converter can be used off of a 48v battery bank. i still would like to know if all of these items actually are 12v and 48v dc loads? how are you achieving the power for them now?
    i.

    The observatory is currently operating and in production in southern AZ. The current observatory is on-grid. I'm contemplating moving it to a darker site. The 12V load consists of the main science camera which draws about 5A in operation and a variety of other gear that draws the rest including computers and networking gear. The 48V load is the telescope itself. The current loads were measured 6months ago with an amp meter or kill-a-watt and are conservative. Interestingly enough providing a clean 12V and 48V supply from battery is much better than the 120V switching power supplies that are in use.

    The other neat thing is that if it's cloudy, I won't be observing. Cloudy days usually precede cloudy nights.
    Sunny days usually precede clear nights. I'd prefer not to have a generator but it seems prudent to include one in the design.

    During the day almost all the gear is turned off and loads are minimal. Just network gear and security cameras.

    Experience tells me to overbuild everything. It' s a bit more expensive, but usually worth it.
    So, based on your comments, a 300ah battery bank should more than suffice.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory

    300ah is good up to about 2 days and i probably would not go quite that far with it to be on the safe side. remember batteries can age a bit too. you could up the capacity too to account for any possible upgrades in the equipment or added extra loads. of course with a generator on hand you may opt to not go for 2 days of backup, but if it were me i would keep the 2 day minimum and possibly opt for 3 days. i suppose you may need to see what battery capacities are available and if the batteries are to be sealed agm or flooded lead acid (fla) that will gas a bit. just be sure that you know a change in the ah capacity of the battery bank could change the charge requirements from the pvs too.

    the pv setup, i am gathering, will be the primary power as the move will put it away from the grid. this would undoubtedly make a backup generator a necessity as stuff happens.
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory
    niel wrote: »
    300ah is good up to about 2 days and i probably would not go quite that far with it to be on the safe side. remember batteries can age a bit too..

    Is the next step now to consider charge a charge controller?
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Looking to power an Observatory
    asteroids wrote: »
    Is the next step now to consider charge a charge controller?

    Yes, let's assume you are going with a 48V battery bank. It's voltage, while charging, can range from 48V - 62V. Your DC gear will need to be able to manage with that, because, someday, someone will forget to turn something off in the AM, and the panels will boost the batteries way up.

    When you have the bank size settled, you can select the watts you need, based on the battery min - max recharge amps. The PV array must be oversized about 20% , to allow for losses from the panel spec, these losses are "real life", much like EPA mileage stickers on cars. Then you will know what size controller you need .
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory

    before we address the rest here, can you say what batteries in what arrangement you are settling on? few work out to have exacting specs that one may calculate for on the fly.

    in general you will want an mppt controller with a battery temperature sensor (bts). the placement of the controller and pvs can be critical as v drop losses could thwart your efforts. the distances involved do heavily come into play for the wires and what size you are using. if the distance from the pvs to the controller are long it may pay to run higher voltages and there are some differences on what a controller can work to with many below the 150vdc threshold. this would be the open circuit voltage of all of the seriesed pvs under the coldest temp you would ever expect.

    i guess i am getting a bit ahead of things here, but in general tell us what length of wire you'd be looking at from the pvs to the controller and what length of wire from the controller to the batteries you'd expect to see. do not underestimate this or v drop losses could be larger than expected and this would throw off calculations on what size wire to use.
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory
    niel wrote: »
    before we address the rest here, can you say what batteries in what arrangement you are settling on? few work out to have exacting specs that one may calculate for on the fly.
    .

    The only time I looked at batteries was with respect to how REAL telco switches provided 48V battery backup. That was a long long long time ago. Therefore, I'm starting from scratch. I have no opinion.
    But, what I do want is something that is very low maintenance. This will be a remote and unattended site. Visits may only happen two/three times a year.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory

    ok, how long are your visits for? just the night ? 2 nights? a week?

    if it is so infrequent at 2 or 3 times a year, it may not be a worthy solar investment. solar is quite expensive compared to running a generator for a short time period.
  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    Re: Looking to power an Observatory
    niel wrote: »
    ok, how long are your visits for? just the night ? 2 nights? a week?

    if it is so infrequent at 2 or 3 times a year, it may not be a worthy solar investment. solar is quite expensive compared to running a generator for a short time period.

    I bet he doesn't mean it will only operate 2 or 3 times a year but be remotely operated. If that is the case a good set of monitoring is also in order.
  • mikeo
    mikeo Solar Expert Posts: 386 ✭✭✭
    Re: Looking to power an Observatory
    I bet he doesn't mean it will only operate 2 or 3 times a year but be remotely operated. If that is the case a good set of monitoring is also in order.
    That is the way many observatories are operated via the internet and remote control. Looks like a good application to use AGM batteries. There are several sealed L16 style AGM batteries in the 390 ah range. This could be matched to a good 60 amp MPPT charge controller and about 2400 watts worth of panels. At his location he will probably see 5 hours of sun on an average day. And as he mentions, on cloudy days the observatory will not be used most of the time. I don't know they keep the observatory heated as this would be one problem keeping electronics and any optics at stable temps.
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory

    The current observatory is operated remotely most clear nights. I've become very picky about equipment that needs to be operated remotely. I haven't looked at charge controllers yet but it'll be essential that they have some kind of computer interface. To recap, so far. 300ah or so AGM batteries in a 48V configuration. A 60amp or better MPPT charge controller. 2400W of solar panels???
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Looking to power an Observatory

    Make sure you have your Amp hours right, and then you can get a good array sizing by planning on 10% of that for peak potential charge current. Like this:

    30 Amps * 48 Volts = 1440 Watts, less typical 77% panel & controller derating = 1870 Watt array.

    Some things that affect this include ambient temperatures (high reduces panel output), elevation (high improves panel output), long distances between array and charge controller, et cetera. Since daytime use will be minimal you could also go for the minimal 5% charge rate (excluding load current). Just be sure the available sun will replace the "used" Amp hours.

    Take a look at the MidNite Classic charge controllers: http://www.solar-electric.com/misoclchco.html
    They are really state-of-the-art and have a simple Ethernet computer interface which may be to your liking.
    Their web site: http://www.midnitesolar.com/
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory
    asteroids wrote: »
    The current observatory is operated remotely most clear nights. I've become very picky about equipment that needs to be operated remotely. I haven't looked at charge controllers yet but it'll be essential that they have some kind of computer interface. To recap, so far. 300ah or so AGM batteries in a 48V configuration. A 60amp or better MPPT charge controller. 2400W of solar panels???

    now that i'm clear on there being nobody there 99% of the time and it is being used by remote control i feel more than 2 days of battery power should be available. this could be for 3 days, plus or minus as i'm not sure of long cloudy stretches there. this figures to be 468ah. now battery ah capacities vary and don't work exactly to what one may want.

    on the smaller side is this battery of which you would need 8 in series to get 48v and it's rated at 405ah over 24hr,
    http://www.solar-electric.com/pvx-4050.html
    and it would seem to be in the neighborhood, but they get derated by temp by at least 25% at 77 degrees f so this would've fit the bill for the original 300ah. it is the largest 6v battery so they would need another string paralleled to use their 6v batteries. paralleling another string can be done, but it is preferred to have 1 string if possible. that leaves the 2v batteries available with more ah capacity, but for 48v one would need 24 of them. now the 468ah (3 day requirement) x 1.25 = 585ah. there are 2 batteries in that ah area with one at 534ah/24hr,
    http://www.solar-electric.com/pvx-5340.html
    and the other is 648ah/24hr,
    http://www.solar-electric.com/pvx-6480.html
    the 2nd one is too much imo.
    that would mean 24-534ah batteries. if you go less than this you must realize that it will not last as long without going beyond 50% dod and could shorten battery life a bit more than you'd want. again it's up to you as you would know what's possible there and is your decision. i am only following guidelines we have settled upon here for this rough outline of a system.

    anyway, the higher ah battery bank will warrant more pvs to keep them charged. the new battery capacity figure of 534ah could also follow the same lines as what we calculated for the pvs before and one would be tempted to up the by 33% too, but i don't think you need go quite that far.
    the 2400w in pv from before would provide 1848w after the .77 derating. dividing this by 57.6v as a rough max volts for these batteries to charge to full gives us 32.083a. this would represent about a 6% charge rate to the battery bank and is within the usual 5-13% recommended. you could up this to get a better % and a faster ability to charge the batteries to full. your loads should not necessarily warrant the increase in pv for proper function as we are only addressing battery needs at this point, but more certainly would not hurt. to achieve the charge figure of 10% would mean 534ah x .1 = 53.4a. figuring in the 77% efficiency would increase it to 69.35a in pv at at least 70v. you guys can recheck my math on all of this as i think it may be off a tad being the amps to charge are more than double what the 32a would be for a 6% rate, but the basics are here.
  • asteroids
    asteroids Registered Users Posts: 17
    Re: Looking to power an Observatory
    niel wrote: »
    now that i'm clear on there being nobody there 99% of the time and it is being used by remote control i feel more than 2 days of battery power should be available. this could be for 3 days, plus or minus as i'm not sure of long cloudy stretches there. this figures to be 468ah. now battery ah capacities vary and don't work exactly to what one may want.

    on the smaller side is this battery of which you would need 8 in series to get 48v and it's rated at 405ah over 24hr,
    http://www.solar-electric.com/pvx-4050.html
    .

    Thanks for all the suggestions. There's a lot to study and think about. Could I move on to the solar panels themselves. Any suggestions here?
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Looking to power an Observatory

    that is and isn't that tough as there are many out there you can go for in many configurations. now with a large array one may want to have the larger pvs in the 200 and up wattage area. for most applications the upper limit is 150v due to controller max limitations on the coldest recorded temperature for that area for the pvs chosen and they are in series. (there are some controllers that exceed 150v) most pvs will show something like .34% increase in temperature per degree c so you may want to convert f to c. a similar thing happens when the temperatures go high as the voltage drops by .34% per degree c and the base for these is at 25 degrees c or 77 degrees f. the hot temp can come into play for the minimum vmp voltage seen by the cc for you wouldn't want that to go lower than what you would need to charge your 48v battery bank with. for sunxtender batteries this would be 14.4 x 4 = 57.6v plus a good 4-6v extra for losses in the controller and wires.

    now there is a question of the charge controller you'd want and there are a few that would actually work for you, but the latest and greatest is the midnite classic. in fact, they have quite a few things that you may wish to supplement your system with such as combiners, battery boxes, and what have you and our host carries the midnite line. the classic has some advantages over many other controllers and you can look over things here- http://www.solar-electric.com/mnclassic.html in 'details' you can get much more info and you can even go to their website for their forum where many of the aspects of their products are discussed, http://www.midnitesolar.com/ see forum listed on upper right side.

    i know i'm not giving you a direct answer on some pvs, but you should stay with more common names. our host carries some good ones. http://www.solar-electric.com/hiposopa.html
    you will also notice that the price per watt is given as well and can help you save a bit of cash. now the monocrystalline types are best, but the multi and poly crystallines aren't too bad either. the type for you to stay away from would be the amorphous or thin film type. now if you have a pv in mind for us to comment on or explore its feasibility for a system for you, be it from our host or elsewhere, then just say the make and model of the pv with the specs of the voc, vmp, imp, and you'll need the isc too at some point. do also check if it gives voltage change %s per degree c temperature info.

    don't forget the pvs will need to be mounted. i'm not an expert on their mountings, but you may want some access to the pvs without allowing easy access to vandals or thieves. also there should be at least 2x the height for the general max snowfall you'd expect in any given day for that which will accumulate on the ground plus what would slide down off of the pvs. also, you still did not give me an idea of the distances between the pvs to cc and cc to batteries might look like as this could influence your controller as well as wire choices.

    as you are finding out, this stuff gets quite involved and so far i'd hate to even guess the costs of the items we've discussed so far, but it's up there.