PDA

View Full Version : Check my math,electrical,solar type question

wierdscience
02-01-2012, 02:23 PM
Got a customer that wants a 12vdc winch and four LED lights to run remotely using a solar panel and storage battery for power.

Winch will draw a maximum 100amp load for about 2 minutes followed by a 20 amp max load some minutes later for maybe 1-1/2 minutes.

It will do this maybe once or twice a week,but maybe twice in the same day or maybe not at all in a given week.

The only other load will be the lights that will run from sun down to sun up everyday,they will draw .9 continuous amps @12vdc. and be switched with a photocell.

I have talked to a local sign company that builds remote lighting systems and winches for bill boards they recommended the following-

Sharp 33 watt panel with a 10amp charge controller along with a 200ah AGM battery pack,solid state relay controls for the winch and lights and four LED all weather universal mount headlamps for the lights.

Does this sound right?Will the system handle the discharge and be able to recover?There is no way to run power out to the system so it will be stand alone solar so it needs to be robust enough to handle the work and last at least 3 years before service.

The sunlight conditions here are pretty good,except this time of year when it's overcast.The sign folks say theirs hold up,but they only use them to service the billboards once a month which involves hoisting tools and materials up several times in an hour.I am skeptical since I don't have much experience with these setups.

Any thoughts?

wb2vsj
02-01-2012, 02:42 PM
A 33 watt panel sounds small to me, but I haven't crunched the numbers.

here's a basic form to try:

http://www.altestore.com/store/calculators/off_grid_calculator/

Need to do a bit of converting here there to get the right units to plug in.

Evan
02-01-2012, 02:51 PM
Look up the hours of sunlight by month for the location in question. Use the lowest month of the year and divide by 2. The reason for dividing by 2 is that fixed panels will only deliver significant power for about six hours per day regardless of how long the sun is up or what direction they are aimed. Also, manufacturers specs are for absolutely ideal conditions. I have a Sharp 80 watt panel and it is very well made. It will deliver 80 watts almost exactly but conditions must be perfect.

If you can't find hours per month then use hours per year which will be available and divide by 24 to give a realistic value for the lowest month. Then divide that by 2.

Using the lower of the two figures then multiply it times the panel rated watts to get watt hours. Divide that by 2 to get a value for the actual energy delivered to charging the batteries and held by the batteries. That gives you watt hours per month on the worst month.

Compare that to the calculated load. Allow for at least a week with no charging, two weeks would be a lot safer. If it is life critical (or million dollar lawsuit critical) I would want the batteries to be able to run the load for at least a month without charging.

2ManyHobbies
02-01-2012, 03:05 PM
Look up the hours of sunlight by month for the location in question. Use the lowest month of the year and divide by 2. The reason for dividing by 2 is that fixed panels will only deliver significant power for about six hours per day regardless of how long the sun is up or what direction they are aimed. Also, manufacturers specs are for absolutely ideal conditions. I have a Sharp 80 watt panel and it is very well made. It will deliver 80 watts almost exactly but conditions must be perfect.

If you can't find hours per month then use hours per year which will be available and divide by 24 to give a realistic value for the lowest month. Then divide that by 2.

Using the lower of the two figures then multiply it times the panel rated watts to get watt hours. Divide that by 2 to get a value for the actual energy delivered to charging the batteries and held by the batteries. That gives you watt hours per month on the worst month.

Compare that to the calculated load. Allow for at least a week with no charging, two weeks would be a lot safer. If it is life critical (or million dollar lawsuit critical) I would want the batteries to be able to run the load for at least a month without charging.Okay, I follow that, but if I've only got a NASA solar irradiance number (average kWhr/day for a year), then what do I use?

rohart
02-01-2012, 03:19 PM
Winching :

100a x 12v = 1200w for 2min - 2400 wm = 2400 x 60 = 144000 joules.

Post winching :

20a x 12v = 240a for 2min = 480 wm = 480 x 60 = 28800 joules.

Total = 172800 or 180000 joules.

Note that this is also 200 + 30 = 230 ampere minutes or 4 ah.

To recover this charge assuming no losses at 33w - 180000 / 33 = 5454 secs
=~ 100 minutes = 1 1/2 hours.

Lighting :

Night lamps need 1a x 12v = 12w for 12 hrs = 144 wh = 144 x 60 x 60
= 518400 joules.

Note that this is also 1a x 12 h = 12 ah.

To recover this charge needs 300 min = 5 hours.

Charging :

1 full day's charging (full 33 watts - degrade for low sun) = 33va x 10 hours
= 2.5a into 12v x 10 hours = 25 ah.

The 200ah battery will easily provide the 100a for 2 or so minutes, and will sneeze at the 4ah that one cycle needs.#

No sun case, battery fully charged:

Let's assume there's no sun. 1 full day needs 12 ah for the lights and 8 ah for two winching sessions = 20 ah. You'll get 3 or 4 days out of a 1 year old battery, so t6he battery is good enough.

Steady state case, sun every day :

Each day's charging will put back 25 ah if all goes well. With low sun angles, you maybe have to degrade this to 10 ah per day. 10 ah will not cover the 12 ah of lighting needed, let alone the 4 ah needed for winching.

Conclusion :

So I conclude that it depends on what the 33w means. If it means you'll get an average of 33 w for 12 hours during a day, possibly by servo sun following, then it's fine.

But if 33 w means that's what you get at noon, you need to up your solar cells by a factor of two or three.

Phew ! That was fun, but I hope others agree !

danlb
02-01-2012, 04:10 PM
Rohart did well.

The only part I would not necessarily agree with is the 2 minute 100 amp draw being insignificant. The 200ah battery is probably rated at X hours at y amps, but the value of y can be chosen to favor the best outcome. The higher the amps, the lower the capacity in amp-hours. For many batteries, the higher the draw, the lower the battery life will be.

In other words, I've seen batteries claim 10 ah at .001 amp when they are 1 ah at .5 amp.

The other point to keep in mind is that the higher the latitude (or in valleys) the effective hours of night increase, so the LED lights may run several hours longer and the daily charge cycle will be shorter.

Since it will be automated, the daily draw will happen even if you have a week or two of rain.

Other points that just popped into my head. Snow will obscure the panel. So will bird crap. Solar panels are nice and warm. Mine attracts some varmint that craps on the shed roof. Some panels do not work if they are even partially shaded, so tree branches are a concern too.

Sounds like a fun project.

Dan

Evan
02-01-2012, 07:46 PM
Okay, I follow that, but if I've only got a NASA solar irradiance number (average kWhr/day for a year), then what do I use?

It won't be hard to figure it but you need to know how that number is determined. They should have an explanation somewhere.

Evan
02-01-2012, 08:06 PM
The only part I would not necessarily agree with is the 2 minute 100 amp draw being insignificant. The 200ah battery is probably rated at X hours at y amps, but the value of y can be chosen to favor the best outcome. The higher the amps, the lower the capacity in amp-hours. For many batteries, the higher the draw, the lower the battery life will be.

Absolutely correct. The reputable battery manufacturers will have a table or a curve chart that shows the relationship between rate of draw and capacity. You have to figure out the amount of time spent at high current and the capacity at that draw versus the amount of time spent at low draw and the capacity at that rate. Then you normalize the numbers to calculate the effective capacity based on the ratio of the different rates.

You can avoid the above by simply assuming the battery capacity as half what is stated. That will avoid discharging too far which shouldn't be done anyway. The best lead acid batteries for the application are AGM VLRA type which is what you have specified. They are usually rated to about 300 full discharge cycles with less than 20% loss of capacity. However, they will still last a lot longer if they are not fully discharged.

The other point to keep in mind is that the higher the latitude (or in valleys) the effective hours of night increase, so the LED lights may run several hours longer and the daily charge cycle will be shorter.

Huh? The average time between sunrise and sunset is ~12 hours per day over a year no matter where you are. In winter it will be shorter but in summer it will be longer.

J Tiers
02-01-2012, 08:11 PM
There are other factors......

WHAT charge controller? Many totally waste 10% to 30% or so of the panel power, by directly connecting to the battery, not stepping down voltage.

Charging itself is about 80% efficient chemically. The result of THAT is that the panel may need to be upsized by about 65% to account for those losses on a power basis.

If you figure ONLY charging amps, not a power calculation, then you only need about a 20% or so increase to account for chemical efficiency, the amps "take care of themselves".

Essentially, the very least you need is a balance of the load and charge.

I came up with about 16.5 Ah load..... if you figure 4 hours good sun, you need 4.2A charge average over 4 hours x 1.2 or 5A charge for 4 hours per day to keep up. Any extra is better cushion. At least 90W panels, preferably more, since the watt rating is at 18V, and simple controllers connect directly for a voltage of only 14 or so.

I figure 9 days cushion on 200 aH, but after that it will be a LONG time to re-charge, and the battery may lose capacity. Did I mention how LONG it will be to recharge after a max discharge?

Evan
02-01-2012, 08:22 PM
Many totally waste 10% to 30% or so of the panel power, by directly connecting to the battery, not stepping down voltage.

The good ones step up the voltage even when the panel output is below the battery charging voltage.

wierdscience
02-01-2012, 10:25 PM
Thanks all for the replies.Good points all,lots of things to consider.

Rohart's post added some clarity to the numbers.The customer has agreed to buy the winch first so we can get hard numbers on the actual draw before we build the rest of the system.If we get a good number for the max now it should improve as the winch wears in.

The sign company owner was kind enough to have his engineer give me a call this afternoon.The panel rating is based on as he put it"the minimum output at 9:00 am in the morning on a 65% overcast day with dirty glass" which is basically "what you would see from a five year old panel that's never been cleaned on a dull overcast day" The full direct sun rating with clean glass is 215watts.Also said I would need at least a 20 amp controller,but he had a 35amp that was only \$12 more,but was 10% more efficient,so good point on the efficiency question.

That makes me feel better about this deal,the panel is \$419 delivered,so it's not a cheap panel.Think he said it measures 49" x 19" x 1" not counting the mounting frame.It's supposed to be Sharp's commercial line.

J Tiers
02-01-2012, 11:42 PM
The good ones step up the voltage even when the panel output is below the battery charging voltage.

That has been heavily touted in the past, but it is actually of very little value. At those voltages, the panel power output is very small*, and the gain is minimal. In areas where the sun is weak and which are not really solar power areas, it may be of some small benefit. Not worth adding as a special feature, but if it were "free" would be obviously no problem.

That possible gain is far smaller than the losses in the case of direct connecting the panels.

The real benefits are obtained by series connecting panels and using a buck-type SMPS to deliver charge current. The voltage conversion is such as to convert the high voltage low current panel output to a lower battery voltage and higher current output, tracking the maximum power output combination of voltage and current. The conversion can be very high efficiency, as high as 97%, depending.

A side benefit is that even at lower-than-rated panel voltages, the maximum power tracking type controllers will continue to maximize the power, giving essentially the same benefit as the formerly hyped voltage boost controllers, but with no added cost or circuitry.

* the panel appears as a "current source", and the voltage reaches rated voltage even at relatively low output powers. Not much current is available at low light levels, so power output is small until the solar input is higher. As a current source, at any output voltage between rated voltage and zero (short circuited) the current changes very little, no matter what the maximum power available is at the prevailing solar input level. So a heavier load merely lowers the output power, wasting the "difference" between actual and maximum available power as heat in the panels.

rdfeil
02-02-2012, 12:32 AM
Wierdscience,

All of the above info is good and basically accurate (I say basically from real world application, see below).

I have built and maintained solar radio sites here in Washington and have run into a few problems which the designs failed to concider...

First, take Evans advice about doubling capacity, I would actually go 4 times minimum here in the NorthWest due to problem #2 below...

Second (problem #2), do NOT believe the sun hour / day charts as posted. I don't know how they compile the data, but in my area the actual sun hours in the winter are at best 50% of what is claimed. In the summer just the opposite is true so the average might be about right but the winter reliability sucks :o .

Third, get the specs on the solar panel from the manufacturer and believe ONLY that. Every panel I have seen lists maximum results in ideal solar conditions. I believe that a 225 watt panel will actually seem to be more like a 33 watt panel in the real world, but only manufacturers specs count.

Fourth, consider all of the following and make your decisions based on what you think... 1) actual power required, 2) maximum discharge rate for the batteries, 3) maximum charge rate for the batteries, 4) realistic expectations from the solar panels, 5) autonomy (sp?) the time the system must be able to work properly without charging properly 6) recovery time... the time it will take to recharge fully after a bad charge event (overcast for 2 weeks or longer), this may not be as important in your area, but in my area we can get 2+ weeks with almost no solar day time so the batteries get pulled way down before recharge, then we may get a week with normal solar time, then the bad two week thing happens again, you see the problem.

Anyway, what I am trying to say is More battery and More panel than you think you need then you might be close :rolleyes: .

For example I have one site that has about a 250 amp/day load on average. This site has over 7500 Amp/Hour of AGM batteries and over 2000 watts of solar panel and it will struggle some winters. In the summer we black tarp 2/3 of the panels to avoid high temp charging.

Evan
02-02-2012, 12:44 AM
If you look at all my recommendations they amount to providing 4 times the capacity.

That makes me feel better about this deal,the panel is \$419 delivered,so it's not a cheap panel.

That is extremely cheap for that particular panel. Panel prices are rock bottom right now because of over production right when subsidies are being cut by governments. The market is flooded with panels and prices are half what they were a couple of years ago. It's a good time to go solar if your conditions are right. It's actually possible that they may pay back before they need replacement.

danlb
02-02-2012, 12:59 AM
Huh? The average time between sunrise and sunset is ~12 hours per day over a year no matter where you are. In winter it will be shorter but in summer it will be longer.

I could have stated it a bit better. Higher latitudes have wider variances between the number of daylight hours in various seasons.

When it comes to charging a battery pack that must stay charged, you have to design for the worst possible charging conditions, not the average.

It might be worth asking how often the winch can be out of service. If it can be out of service for a few days at a time, then the reserve can be much, much lower.

BTW, you want to add a low voltage cut-out so that the winch does not over discharge the battery when it's mostly depleted. I can see a situation where the winch gets 1/2 way through the cycle and runs out of juice, then ruins the battery.

Dan

Evan
02-02-2012, 01:15 AM
Good point on the cutout and pretty easy to implement.

wb2vsj
02-02-2012, 06:55 AM
...

That is extremely cheap for that particular panel. Panel prices are rock bottom right now because of over production right when subsidies are being cut by governments. The market is flooded with panels and prices are half what they were a couple of years ago. It's a good time to go solar if your conditions are right. It's actually possible that they may pay back before they need replacement.

Indeed! Mid 1990's I bought ten 120W panels, for \$600 each. so \$419 for a 225W rated panel is a good buy.

J Tiers
02-02-2012, 08:47 AM
"Rules of thumb" like "4x the rating" or whatever are Ok to get "in the ballpark". They are often used to discredit solar power, "you need 5 times the power of panel than your load should need....."

Nothing substitutes for actually doing the calculations for the load and charge system, and accurate estimates of the local solar availability. After that you can figure out how many / how big the panels need to be, whether they must track, etc, etc.

Solar power has some downsides, but it is very reliable in the correct geographical area.

In the "average" area it is likely more reliable than wind power. And usually easier to get past the local "appearance police" if you are unfortunate enough to have them.

fjk
02-02-2012, 09:44 AM
I did the calculations to solarize our off-the-grid cabin in Maine.

All the previous math seems pretty much right on.

The only things I'd do are to 1) make all the numbers "worse" by a factor of 2 (divide capacities by 1/2, multiply loads by 2) in order to account for the random gotchas, decreases in efficiency, etc, etc, etc. Second, find out how long the lights should work if there is minimal sunlight to recharge and then size the battery accordingly (FWIW, on cloudy days up in Maine, our panels
can still provide between 1/3 and 2/3 of their rated output, so you can use
that as a rough guide).

Frank

p.s. the "worsening" of all the numbers also gives some extra room to accomodate the things that get added to the system...

wierdscience
02-02-2012, 01:43 PM
The \$419 price is buying them though the sign company using their volume discounts.If I went at it alone they would be \$470ish,still not bad though.