Matching solar panels to batteries to consumption

[Sauntress]

Originally Posted by thomm225

Actually, his boat is really something. Here's a picture he sent me when I asked about it.

Sauntress,

If there is a problem with me posting this, I'm sure we can have the moderators delete it.

It's an awesome picture though of a beautiful boat......

Tom

That is taken in the inner basin in Ferrol harbour NW Spain. We are slipping lines, jib backed to take us through the wind and then sheeted in to take us clear. Nothing to do with electonics.

[thomm225]

Originally Posted by Sauntress

Nothing to do with electonics.

Let's get back to the electronics then.

I may actually upgrade to this PWM Controller that was mentioned earlier. I like the readouts

https://www.amazon.com/WindyNation-Regulator-Controller-Adjustable-Settings/dp/B015S39PTU

It will replace the $12.00 PWM Controller I installed 3 years ago that's been regulating my 100 watt panel. This panel charges my two 12 volt batteries that I have in parallel

If you don't understand electronics though, you may not want to order this one as many times it doesn't come with a manual

https://www.amazon.com/Sun-YOBA-Cont...41EG?th=1gauge

I think I'm going to do some checks this week with my meter just to verify exactly how much current is flowing through the system at different times. (under different loads)

My system still isn't hooked up permanently

The point is that you can experiment with various systems and setups to see which you like best. I have 100 watts of solar now but may add the other 40 watts. (two 20 watt panels)

[Sauntress]

Originally Posted by thomm225

Let's get back to the electronics then.

I may actually upgrade to this PWM Controller that was mentioned earlier. I like the readouts

https://www.amazon.com/WindyNation-Regulator-Controller-Adjustable-Settings/dp/B015S39PTU

Yes. I went through enough of the customer reviews to be persuaded.

[thomm225]

I haven't read every post so I don't know if it's been mentioned, but you may want to buy two controllers.

I had a weird problem this past summer where my autopilot would fail randomly. Then my GPS would flash.

I thought it was a connection problem at first. It turned out not to be a connection problem, but my controller was acting very strange.

It's easy to get to so I swapped it out with no change quickly eliminating it as the problem.

It ended up being a bad 5 year old battery. I troubleshot this problem about 10 miles offshore and came in on the one remaining battery on autopilot

If you take long voyages, spares are nice in this situation

[Pete7]

So a couple of Lensun GRP backed solar panels a widget to control the charge rated and stop discharges at night and Bobs your uncle, problem solved.

Lensun Solar Energy Store

Pete

[Sauntress]

Just controller or controller and battery monitor?

[Pete7]

Battery monitor really useful for us as I can see if we need to start the engine if its been raining for a couple of days. Not sure you could do much in a similar situation, but at least you would know about it. Guess fire up the oil lamps.

Pete

[Sauntress]

Which about winds it up for me. Thank you everybody.

[Copacabana]

Sauntress, I think you could get by fine with just an inexpensive volt meter to follow your battery voltage. No need for an expensive battery monitor. I still think a 100w panel, a decent PWM controller and about 100ah of batteries is all you need for your power needs. You would have a decent reserve of power and the panel would keep your battery at close to 100% SOC at all times, giving you a long battery life.

[Sauntress]

Originally Posted by Copacabana

Sauntress, I think you could get by fine with just an inexpensive volt meter to follow your battery voltage. No need for an expensive battery monitor. I still think a 100w panel, a decent PWM controller and about 100ah of batteries is all you need for your power needs. You would have a decent reserve of power and the panel would keep your battery at close to 100% SOC at all times, giving you a long battery life.

Yes, someone talked about "murdering" batteries, so that makes sense. They cost quite a bit in the first place. We will start, this spring, with 50w or 60w, and if that is not enough we can add.

[botanybay]

I can only say for myself but on Los Angeles to Hawaii (21 days) and saw two ships, Hawaii to Marshall Islands (15 days) and no ships seen, Marshall Islands to midway (14 days) no ships seen, and midway to Los Angeles (35 days) and many ships seen after half way point.

I still religiously ran a masthead tricolor. I would not consider running without except in a dire condition.

Just one sailors opinion....

Sent from my SM-G935V using Cruisers Sailing Forum mobile app

[Bleemus]

I did the Twostar Race and the delivery crew to England said they had the batteries tested and were perfect. We realized one day into the race that was not true. We ran the whole race without running lights so the wind generator and two meager solar panels could get us a daily weatherfax and position. That was before AIS. I wasn't proud of it but it was what it was at the time. Would I want to do it again? Nope.

[Adelie]

Originally Posted by Sauntress

Just controller or controller and battery monitor?

I would probably pass on the battery monitor.

If your loads were higher and more variable and you had more significant charging sources, yeah maybe.

Let's start from the beginning. As I understand it from the first post you will only run navigation/anchor, compass and chart table lights.

Let's assume 10hr use per day.
Nav/anchor - 2.1amp --> 21ahr/d
Compass - 0.1amp -->1ahr/d

Chart table let's assume 0.8amp x 1 hr/night -->round to 1amp

So net 23amp-hr per cycle. That means on nights after a full recharge, you will be at 77% SoC at your lowest point just before dawn.

If you get no recharge at all the next day you will still be at 54% SoC at the end of the second night.

My take is that you have plenty of reserve storage capacity. Actually you will have more than the nameplate capacity, Peukert effect means that since you are drawing power at the 80hr rate instead of the 20hr rate my guess is that the battery capacity will be somewhere in the vicinity of 130amp-hr. Let's assume 100ahr anyway to make things simple and account for the fact that the batteries will age and lose capacity.

So how much generating capacity do you need? The rule of thumb is that a panel mounted flat using a PWM controller will produce daily amp-hours equal to 25% of panel nameplate capacity in watts. So a 100w panel will produce about 25ahr/d. That would meet your demand with a little reserve.

My guess is that for a series of cloudy days where you can't full recharge you will slowly draw down your batteries and it will be almost a week before you start dropping below 50% SoC. To get back to 100% though will take 14 days of good weather, you don't have enough reserve generating capacity to do it any faster. So for Panel Capacity Option A] you up your panel nameplate capacity to 125w and then it's over a week of clouds to get to 50% and 4 days of perfect conditions to get back to 100%. So that answers your original question at the start of the thread. But there is an option B] for panel size a number of paragraphs further along.

So getting back to the battery monitor. A normal shunt monitor that measures amps in and amps out has a number of inaccuracies. Overtime it can no longer tell what the SoC is. This is like running a dead reckoning plot, after a while the result is not worthless but it isn't very accurate. If the battery were getting to 100% SoC regularly, (say every 5th day as a guess) [getting to 100% is like getting a fix to start a new DR from] a shunt monitor would be keeping adequate track of battery status. I believe there a number of high end monitors that keep track of some of the inaccuracies like battery temp and Peukert effect and give a better running estimate between full charges. You pay for that extra functionality.

The only "monitor" I am aware that doesn't use a shunt is The Balmar Smart Guage which uses voltage sensing and some sort of learning algorithm. I have no opinion on it but it is recommended by MaineSail who is one of the few independent sources for marine electrical/electronics testing. He seems to recommend it over shunt monitors. It's a $275-325 item.

Third option is a voltmeter. Since you need accuracy 2 places past the decimal the voltmeter needs to be a digital readout. If you can get one that has a temp sensor you can plant on the batteries, all the better. If not pick up a kitchen thermometer (dial or digital) that reads to the nearest 1C or better. This is a $20-50 item maybe more with the temp sensor. You also want a specific gravity meter to test battery electrolyte, assuming the batteries are not sealed.

There are a number of sources for voltage vs resting state of charge (i.e. not loaded for the last 2-24hr depending on the authority you are reading). That is not what you need but a good starting place. I also found one for Voltage vs SoC loaded at: batteries - How do I calculate the self discharge rate of a lead acid battery? - Electrical Engineering Stack Exchange See graph below.

If you want to make one specific to your battery set up and your normal loads then do the following:

Sometime when you are sitting at a dock for a number of days, fully charge the batteries and equalize them, [you want to pay the extra bucks for that feature in your charge controller].
1. Let the battery rest all night.
2. In the morning,
a. if you can measure the specific gravity in each cell of the batteries and adjust for temp.
b. Turn on the voltmeter and after about 30min take a reading.
That is resting voltage fully charged. Any significant difference in voltage or specific gravity from readily available graphs will give you an idea that your batteries are starting to loose capacity.
3. Turn on the Navigation light and the compass light but not the chart table light. Measure Loaded voltage and temp.
4. After 5 hr measure battery temp and loaded voltage again then turn off the lights and voltmeter.
5. That night at sunset, check the resting voltage and turn the voltmeter back off.
6. Next morning measure battery temp, specific gravity, and resting voltage.
7. Turn the lights (Nav & Compass) back on. Wait 30min then check loaded voltage.
8. Repeat steps 5 thru 7 until the morning specific gravity and/or resting voltage indicate you have reached 50% or gone below 50% SoC.

If the weather has been consistent battery temp should be too.

If the batteries are in good condition so Peukert kicks in nicely I could see this taking 6 or 7 cycles. If you wanted to shorten the total test period or can't arrange to be at the boat mid-day turn the lights on for 10hr per day and skip step 5. every cycle.

The loaded voltages at the beginning and end of each rest period should be almost the same.

Graph %SoC determined by resting voltage or specific gravity against loaded voltage. If you use Excel it can plot that for you and fill in a curve between data points. The curve is good for that battery temp and similar battery temps.

You will want to make a number of graphs. Each for a different battery temp. I would make 2 or maybe 3. Make one in the summer, maybe one in the fall or spring and one in the winter. These need to be made when you will be on a dock long enough and the weather forecast is for consistent high and low temps for the period you will be testing.

If you can get hold of an ammeter that reads to 3 amps accurate to 0.1 amp less you can determine your actual battery capacity. Use it during one 5 hr test cycle. Also make sure you accurately record on and off times for lights.

OK this is a lot of work. As a nerd engineer I might go thru this but probably I would just use the graph below.


Panel Capacity Option B]

Rather than increase panel capacity to meet normal load there is also the option of decreasing load. The compass and chart lights are pretty minimal already, the big user is the Tricolor/Anchor light at 21ahr/d, obviously incandescent. If that were an LED fixture consumption would be on the order of 0.2amps or 2 ahr/d. Now your net consumption is 3-4ahr/d depending on chart-light usage. At 4ahr/d that's 28ahr/wk which is the high end of what you say your panels put out. Let's say you converted the other 2 lights to LED and got their use down to 0.5ahr/d and the total use to 2.5ahr/d then you are looking at 17.5ahr/wk and that's almost the low end of production for your existing panels.

If your compass manufacturer is still in business I expect they have an LED conversion kit. If not maybe you can find something that works.

If you already have the chart-light fixture it probably uses a standard bulb of some sort. LED bulbs have been made to replace all standard marine bulbs I am aware of. If you don't already have the fixture then you are wide open in the choice of fixture and bulb. In choosing for the chart table my reading suggests that the best choice is a white dimmable light set to the lowest usable brightness. The reason for this is that the lowest usable white light is significantly dimmer than the lowest usable red light and your eyes recover night vision faster in the end. The paper on this I read indicated there were times when red was better but the particulars did not pertain to reading charts and recognizing colored lights in the dark outside after chart reading.

For the tricolor/anchor light if you have the fixture already, then find an appropriate LED bulb to fit it and keep the incandescents as spares. If you don't have the fixture already then I have no recommendation for a fixture.

For my boat I intend to mount the NASA/Clipper SuperNova Combi. I am choosing if for a combination of price (£135.46 UK and $230 US), and low weight (I have a very small boat) and compliance with color/brightness/vertical spread requirements. About 8-10yr ago one of the UK mags tested a number of LED tricolors and praised the Supernova for having a full 25* vertical spread on the light so it would be visible heeled. I don't know if other lights have come more into compliance, I assume they have but don't know one way or the other. Early reviews of the product were mixed. More recently reviews have been better but fewer of them maybe indicating better production quality. I do not recommend this product but I am going to role the dice and see.

I looked up the model number you listed for your solar panel. What I found indicated it was an 11w panel that could produce 4.3ahr/d at the latitude of NewYork in the summer. Let's say you are getting 3ahr/d or 21ahr/wk to be safe.

Let's further assume that you go on passage, it's cloudy and you are getting 1.5ahr/d from the panels and that you have only converted the Tricolor to LED so your daily use is 4ahr/d so you have a net loss of 2.5ahr/d. Because of your low use rate your battery capacity would be up near 150ahr but the self-discharge of the battery cuts into that so you only have 130ahr effective. That means it will be 26 day before you draw the battery down to 50% SoC. If you ditch the chartlight and dim one of the oil-lamps way down you could probably stretch to near 40 days, perhaps enough to cross the Atlantic.

If you already have fixtures converting to LED is probably cheaper than getting a bigger panel. You want to get a charge controller that can equalize regardless so that doesn't come into the money equation for comparison purposes.

I would be a little uncomfortable with the close margin between existing output and anticipated load unless everything was converted to LED. The supplier I got the panel I use for camping has a 20w panel that is 14"x19"x1" for $50. So Solar Panel Capacity Option B] is 20w. I would think about finding a place to permanantly mount that and bring out the existing panel when the new panel needed help on cloudy days.

Going back to the original question at the top this post. Get a PWM controller and a digital voltmeter and use the chart below rather than getting a battery monitor.

[mitiempo]

I agree in this case a battery monitor is not necessary.

I would definitely convert the tricolor to Led, as well as any lights that are still incandescent.

I would get a larger panel - say 30 to 40 watts. A flexible panel is an option and is easier to store below when not needed. And yes a Pwm controller will be fine for your usage.

The loads you will have, assuming all lights are Led, will be low enough that you shouldn't have to go through what Adelie posted above. The chart below should be a good enough guide. With all Led usage the draw will be low enough that I wouldn't worry too much about resting voltage. With a few Led lights on you will be at a bit higher state of charge than the chart shows but that is not a bad thing. Do get a good digital voltmeter.

[Sauntress]

The existing panel (and it is one panel, not two) is, to judge by the dimensions a Unisolar MBC-262 which would have been installed in the early 1990's. The owner's manual, which I still have says
Rated Power (Wp) 11,
Voltage Typ Max Pwr (V) 15.60
Current Typ Max Pwr (A) .70
There is more, but presumably that is enough for information purposes, I hope. I have printed and read through, several times, your (Adelle) long and learned explanation and realise that the more I think I understand, the less I actually understand. Thus I am in the hands of the experts. I also printed the Article by Richard Perez "Lead-Acid Battery State of Charge vs. Voltage which will make nice bed-time reading.
And yes you have answered my question. But I confess I am confused. Both you and Mitiempo now seem to be saying that if I convert all to LED (which I will certainly do) I could almost get by on the existing Unisolar panel. Or could do if I added a 40 watt panel. Is that right?

I will get hold of a digital voltmeter
The batteries are "maintenance free rechargeable sealed Lead-Acid.