Van RV Proposed Solar System Check
MysteriousFoundation
Registered Users Posts: 22 ✭✭
Hello, everyone. I think I've read enough to design a system without being completely foolish, buy we'll see!
Background and Limitations
I have a large Sprinter van I'm converting to an RV. The roof has enough room for 700W of monocrystalline panels while leaving space for a large vent fan. I'm looking at these 175W panels. (Feel free to suggest better/cheaper panels with free shipping!) I've resigned myself to needing to tilt the panels to get enough power, so I'm using the "Optimal Year Round" tilt values from http://www.solarelectricityhandbook.com at a couple of different latitudes for planning. (And who knows, maybe I'll buy some linear actuators and tilt them to ideal angles every month remotely!)
Power Usage
Just in case these tables look like garbage, I'll attach PDF print-outs as well. (They're in the first reply.) I wish this forum would let us keep colors and borders in tables, but it actively removes them.
These are the most accurate numbers I could find for the various DC devices, and I measured the AC devices with a Kill-A-Watt. I tried to keep the hours at the maximum realistic usage. If I can find a good DC desk fan, that'd be nice, but barring that, I'll need to use an AC one. I sleep with a fan on 100% of the time--even in Winter--for the noise and breeze. As you can see, it's the single highest energy usage. Low vent fan most of the day, with perhaps medium when cooking. I'm assuming an 85% inverter efficiency, which is the most conservative number listed by Xantrex and other brands. (However, the end-to-end solar efficiency factor of 0.52 supposedly already accounts for this, but the difference of double-counting it is not particularly large.)
Battery Sizing
Here, I take the total Wh and multiply times 2 for a day of reserve power, and targeting 50% DOD. At 12V, this yields a desired battery bank size of 322 Ah. While there are some 6V deep cycle batteries about that size, I doubt I can beat the price of 4x Trojan T-105's making it 450 Ah. In my previous post I mentioned I have a source for them at $110 each plus core. That's too much battery, but as long as I don't use it, I should be fine, right? However, I will do a price check on 2x Trojan J305P-AC.
[Post continues in first reply]
I hit the character limit!
Background and Limitations
I have a large Sprinter van I'm converting to an RV. The roof has enough room for 700W of monocrystalline panels while leaving space for a large vent fan. I'm looking at these 175W panels. (Feel free to suggest better/cheaper panels with free shipping!) I've resigned myself to needing to tilt the panels to get enough power, so I'm using the "Optimal Year Round" tilt values from http://www.solarelectricityhandbook.com at a couple of different latitudes for planning. (And who knows, maybe I'll buy some linear actuators and tilt them to ideal angles every month remotely!)
Power Usage
Just in case these tables look like garbage, I'll attach PDF print-outs as well. (They're in the first reply.) I wish this forum would let us keep colors and borders in tables, but it actively removes them.
| Appliance | Qty | V | I | W | Efficiency | Hours | Total Wh |
| Dometic Fridge 40 qt (90F/41F) | 1 | 12 | 0.64 | 7.68 | 1 | 24 | 184 |
| Xantrex 1000W Inverter phantom draw | 1 | 12 | 0.35 | 4.2 | 1 | 12 | 50 |
| LED light (single) | 1 | 12 | 0.25 | 3 | 1 | 4 | 12 |
| LED lights (additional) | 3 | 12 | 0.25 | 9 | 1 | 2 | 54 |
| Maxxair Vent fan (med) | 1 | 12 | 1.2 | 14.4 | 1 | 4 | 58 |
| Maxxair Vent fan (low) | 1 | 12 | 0.6 | 7.2 | 1 | 16 | 115 |
| Tablet charging | 1 | 12 | 0.5 | 6 | 0.85 | 2 | 14 |
| Phone charging | 1 | 12 | 0.5 | 6 | 0.85 | 2 | 14 |
| Desk Fan (high) | 1 | 120 | 39 | 0.85 | 4 | 184 | |
| Desk Fan (low) | 1 | 120 | 23 | 0.85 | 8 | 216 | |
| Raspberry Pi | 1 | 120 | 2.5 | 0.85 | 4 | 12 | |
| Dell Monitor (25% Brightness) | 1 | 120 | 11 | 0.85 | 4 | 52 |
These are the most accurate numbers I could find for the various DC devices, and I measured the AC devices with a Kill-A-Watt. I tried to keep the hours at the maximum realistic usage. If I can find a good DC desk fan, that'd be nice, but barring that, I'll need to use an AC one. I sleep with a fan on 100% of the time--even in Winter--for the noise and breeze. As you can see, it's the single highest energy usage. Low vent fan most of the day, with perhaps medium when cooking. I'm assuming an 85% inverter efficiency, which is the most conservative number listed by Xantrex and other brands. (However, the end-to-end solar efficiency factor of 0.52 supposedly already accounts for this, but the difference of double-counting it is not particularly large.)
Battery Sizing
| Grand Total Wh | 965.3 |
| Days | 2.0 |
| DOD | 0.50 |
| Effective Wh | 3861.1 |
| Batt Amp Hours | 321.8 |
Here, I take the total Wh and multiply times 2 for a day of reserve power, and targeting 50% DOD. At 12V, this yields a desired battery bank size of 322 Ah. While there are some 6V deep cycle batteries about that size, I doubt I can beat the price of 4x Trojan T-105's making it 450 Ah. In my previous post I mentioned I have a source for them at $110 each plus core. That's too much battery, but as long as I don't use it, I should be fine, right? However, I will do a price check on 2x Trojan J305P-AC.
[Post continues in first reply]
I hit the character limit!
0
Comments
Here's where things get a little bleak. I used the solar irradiance charts to calculate an average year to highlight the worst case scenarios. I think October is the latest I'd want to stick around in Rochester NY, and NC is the furthest North I'd want to be in November, so here is a full year of values, switching locations in those months, and again in March:
The insolation values are for South-facing, "Optimal Year Round" tilt for the two latitudes. I calculated the watt hours by taking 700W * 0.52 * Insolation. Days of power is the watt hours divided by the grand total power usage above (965.3 Wh). Like I said earlier, I've been told that 0.52 factor already takes into account inverter efficiency. It doesn't make a huge impact to double count it, though. If I change the efficiency to 1, my power usage drops to 891 Wh, and the worst "Days power" above (Dec) rises from 1.3 to 1.4. Battery bank size drops to 297. All in all not a huge change, so I might as well keep the more conservative estimate.
You can see that I often can't pay back two days of usage in one day. This concerns me a little. If I have good day, bad day, good day, bad day... my batteries might not fully charge. This is also ignoring device usage while the batteries are charging. That scenario is not very realistic, though. I'll much more likely face a few days in a row of rain and I'll be stuck using the generator regardless.
Please let me know if my ratios of batteries to power usage to panels are OK or if I'm off. Thanks!
Solar Controller
Assuming all of the above looks OK, what controller do you all recommend? I see the MorningStar ones mentioned on here a lot. Specifically the TS-MPPT-XX ones. I'm still confused on how to size the controller.
PDF Printouts attached below
https://www.solarblvd.com/products/solar-cynergy-180-watt-12v-monocrystalline-solar-panel2018/
I'd say it's likely they are made in the same factory. These fom solarblvd appear to come with just a junction box with MC-4 connectors. stating that 1000mm cables are optional. This would be just fine for me avoiding an extra connection.
I would go with a fanless controller for an RV, Morningstar makes a good product, Schneider would be another option. These are 12 volt nominal panels so you could use a PWM if you wanted.
I would check into O2cool fans, I saw one still listed as 12 volts on Amazon. Here is a 'sure' 12 volt version;
https://www.amazon.com/RoadPro-RP8000-Indoor-Outdoor-Power/dp/B00KJGRLJI/ref=sr_1_9?ie=UTF8&qid=1531194971&sr=8-9&keywords=o2cool+10
The standard 10" are now 9 volt, I have 2 running now in my home, on for air circulation in the back bedroom with air conditioning and 1 in the window in the un-air condition area of the house. Even on wall warts they draw 10Watts or less.
Here is a bit of discussion of the Endless breeze and o2cool.
https://www.cheaprvliving.com/heat-cold/stay-warm-winter-cool-summer/
It has a timer built in and 4 fan speeds and a cycle through option (that nobody likes...lol) Works pretty well, and I like that it shuts off, I'll use it when crashing in my Scion Xb on trips to see mom. Here are posted energy levels from the company.
Here is the posted amp draw, by the company in Amazon [email protected]
One speed: 0.17A two speed: 0.35A three speed:0.6 A four speed: 0.8A
Puts out quite a good bit of air. I usually leave it on 2nd speed and set the timer for over night. I have just let it run and my 10,000(claimed) power bank shows about 60% charge left after sleeping (likely 7 hours). Lots of variables there!
Not cheap... Though I got mine when they had a 25 or 30% off one item sale.
https://www.amazon.com/OPOLAR-Included-Operated-Operation-Adjustable/dp/B077SDZL88/
Regarding your device useage, the domestic refrigerator consumption seems too low, 12V, 0.64A, 7.68W, or is this a chest refrigerator like SunDanzer?
900W 3 × 300W No name brand Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah FLA 24V nominal as a backup system.
5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergencies and welding.
I have a similar compressor, but much more heavily insulated box, which keeps consumption down to around the 180wh/day range listed.
If the unit is absorption type, the consumption may be just be for electric lights etc, with cooling provided by LPG. These can be run on electric alone, but they're really inefficient and will take way more than 180wh/day running electric only.
Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
Yes, it's a chest or "cooler" type design. https://www.amazon.com/gp/product/B01ENA4ZH6
The numbers came straight from the manufacturer. The amperage I listed is for the fridge set at 41°F in a 90°F environment, maintaining already cold items. It's has a compressor, and it only runs as needed. It's 0.16 amps when idle, and 3 or so when actively cooling. So, it must average out to 0.64 amps. I can't find those numbers right now, but I'll keep looking.
Here's one anecdote from the Amazon review: "90+ degree weather and in a hot SUV parked in the sun and it never used more than 28 amp/hrs over 24 hrs." That's 82% more usage than what I listed. But another anecdote said it's 0.47 Ah/h (once cool) in his living room.
You have to read a lot of reviews to find ones that list the amp-hour usage, because most people just list the amps when running. It obviously won't have a 100% duty cycle, so that's not useful information except for deciding what size fuse to use.
I finally found it, and I was off by a small amount.
https://www.dometic.com/en-us/us/products/food-and-beverage/coolers/electric-coolers/dometic-cfx-40w-_-152176
Now, obviously, that amperage doesn't include opening and closing it, or adding new room temperature food, but it's a good baseline. There are already so many conservative estimates in these numbers, I feel OK leaving this one optimistic. Especially if I take chilled food from a store and put it in the fridge immediately. And since it's top loading, you don't even lose the cold air when you open it. I also plan on buying the official insulated cover. I'm hoping that the quoted numbers don't already include using that.
Fans
If a USB fan can move enough air, that would be a large savings in power. I can always use my phone to generate white noise if the fan's not loud enough. (For my purposes, all these "Very quiet!" fans are not a good thing!
Removing the desk fan from my math and adding the Oploar 8" on high for 12 hours brought my total energy usage down to 709 Wh/day. Almost 27% less! I really hope a fan like that works out. For only $30 it's worth trying out. It would raise my worst sun example (Dec in Greensboro) from 1.3 days of power up to 1.77 days.
Panels
Controller
I'm still looking for specific advice on what controller size I need, not just brand advice.
How many amps does the controller need to be rated for to be safe and effective with the four panels I linked in the original post?
Overall
Ignoring specific usage, does my math seem OK? Would you be comfortable having only 700 W of panels given this usage and these batteries? (With or without the AC desk fan.)
The math looks modest, but man.... the stories I read on reddit, Amazon reviews, or YouTube videos, has people powering a fridge, two laptops, phones, a Maxxair fan, and whatever else, and they're like, "We have 300W of panels and a 150 Ah battery." -_- It makes me feel like I'm really over-engineering things, but I obviously don't know what their real usage is.
I would hate to not buy enough and find out I'm destroying my batteries while having no room for more panels and having to make difficult cuts in usage or run my generator all the time, but I'd hate it just as much if I spend $2500 and I find out I'm only using 10% of my batteries per day when I could have just spent less money.
I'm just looking for some reassurance that I'm at least in the ballpark, and I'm not missing something obvious. Especially regarding current and charge rate, etc. Things that aren't necessarily obvious when looking at raw numbers.
FWIW, I have 2x6v GC ([email protected]) batteries on my sailboat (no solar). With a Danfoss compressor based top load fridge, and normal lights, laptop, etc. (but no fan), that will get to ~50% SOC in 1-2 days. With 700w of solar, I could just about get to fully charged from 50% on a sunny day with well aimed panels and little/no shading. You don't need to get to fully charged every day. If you get there every week or so, and avoid leaving the batteries at a low SOC for long, that's ok.
If I was doing it, I'd probably start with a single string of GC batteries and see how it goes. If a 2 string configuration makes sense based on actual experience, you can easily add a second string in the first month or two, or if something really goes sideways, you've only lost a single string.
Your math looks about right to me. You may have to run the genny occasionally, but we (mostly) all do.
Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
My question is: is that taken into account at all with the 0.52 end-to-end loss factor? I don't want to be too conservative. Or does that already assume you're using MPPT?
How many amps does the controller need to be rated for to be safe and effective with the four panels I linked in the original post?
- 0.81 for warm to hot solar panels (freezing or subfreezing panel temperatures would be approaching 1.0 for MPPT controllers)... Also includes a few percent losses for dirty panels
- 0.95 for MPPT efficiency (0.81*0.95=0.77 happens to work out for PWM system efficiency for other reasons)
- 0.80 for Flooded Cell Lead Acid battery (use 0.90 for AGM or higher)
- 0.85 for "average" loading AC inverter efficiency (note that very light loading can have much lower efficiency/Tare losses)
Multiply all the numbers together:- 0.81 * 0.95 * 0.80 * 0.85 = 0.52
The above numbers are for well designed systems running "nominal" configurations. If you do something "different", the numbers can change.Note that panel Wattage is "Marketing Specifications" not NOTC or PTC derated solar panel ratings (NOTC and PTC approach the ~0.81% derating factor--They are a bit less conservative, as I recall). Some folks like to use 75% panel+controller derating (instead of 0.77)--A bit more conservative.
If you run your system outside of "nominal" operation specifications, your deratings will be different (run FLA at 90-100% SoC, efficiency falls; run 3,000 Watt inverter with 10 Watt load, effiency will be way down, etc.).
They are relatively conservative (i.e., most people will see "at least" the predicted performance). If you were designing a system that needed 100% of predicted loads (say a refrigeration system) and no variable loading (optional loads, varying loads, etc.), then the base power needs should be increased by 1/0.75 to 1/0.65 (1.33 to 1.54x larger solar array suggested)-- You do not want to use 100% of predicted harvest every day. Some days will be better, some will be worse.
-Bill
For a PWM controller charging a 12 volt battery bank, 40 Amp minimum. If you "believe" the NEC, they want to derate by 1/0.85 * 1/0.85 or 1.38 x 4 x 9.8a = 54 Amp PWM controller minimum.
For MPPT controllers, we suggest a maximum "cost effective" over paneling of:
- 4 * 175 Watt * 0.77 panel+controller derating * 1/14.5 volts charging = 37 amp controller minimumB
As always, there are details... Check the manuals, and some vendors have a website that helps to configure the array to controller (maximum Voc-array voltage on cold days, Maximum Imp/Isc array input current, etc.).Check the manuals/specifications before purchasing your hardware.
-Bill
Fuses vs Breakers
If I go with the 4x Trojan T-105 in parallel, I was looking to get two in-line fuse holders and some 50A (or maybe larger) fuses to go at the ends of each string in case one string goes bad or gets disconnected. That way the other string will blow a fuse instead of taking the whole load. Then I thought about circuit breakers instead. It turns out that in-line 50A circuit breakers are about $13 on Amazon. That's as cheap as some ANL fuse holders and a couple of fuses. But are they trustworthy? That particular model I linked is re-branded by at least five different companies.
Do any of you use DC circuit breakers instead of fuses? Are there reputable brands you'd recommend? How much should I be looking to spend?
|| 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 ,
Wires and Voltage Drops
The panels I'm looking at have 3' #12 wire leads. With four panels that's 24' of wire all looped around on my roof. I'll need to add probably 18' more to get it down to the battery bank because of the width of the array. I can buy 15 feet of extension cables in #8 or #10. There's a $15 difference. Totally negligible considering the cost of the system, but money is money! Is saving 0.18% of a voltage drop over 18 feet worth $15? That's 0.126V out of 70V. At 9.65A that's 1.22W! I can't lose that!
.... No, I'm just going to buy the #10. Or, heck, I could save $5 more and buy the #12 and still have only a 2% drop over all 42'.
This is all assuming that I should be using the panel short circuit amperage in this math. Should I use more to account for the "unusual solar circumstances" I've read about? Or do you only worry about that for safety and controller size, etc? I've read you should add 25% to the amperage for that calculation. If I use the 12.06A in the math, my voltage drop jumps to 2.52%. That's a 21W loss in total.
I don't even know why I posted all of this. We're talking a few dollars here for a fatter wire. But please tell me if I'm way off and I should care more about fractions of a percent of a drop. I think because I'm planning on running the panels in series that it matters so little. With lower voltages wire size would matter so much more.
MorningStar Calculator
It turns out their array size calculator will happily let you enter a positive Voc temperature coefficient! I only noticed it because the resultant "Minimum Vmp (average high temp.)" was higher than the "Maximum Vmp (record low temp.)" Thanks, MorningStar!
I get that it's operator error, but it should validate the sign of the number. Maybe I'll drop them a line.
With the correct coefficient, my array is no longer "optimal" for the MS-MPPT-45. However, it's not a big deal. It only becomes suboptimal at -9°F or lower where the voltage rises to where controller can no longer harvest the maximum power point. It says it won't harm the controller. That's not something I'm too worried about.
2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric, 700 ah @24 volt AGM battery bank. Plenty of Baja Sea of Cortez sunshine.
I don't think Midnite breakers need derating (the 25% thing), but others might. Be sure to get DC breakers rated for your record low temp string Voc.
Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
|| 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 ,
- One string of 4
- Two strings of 2
I can't wire them all in parallel. At least not according to the MorningStar calculator.I had planned on wiring them all in series because a higher voltage results in a lower percent voltage drop. I hadn't considered the shading issue. I was already going to make sure there was no shading from vent fans, roof rack, etc. When I'm in a park or national forest, I'm likely screwed either way, unless I can find full sun.
Thanks for the tip! I supposed I'll plan my wiring assuming I'm going to use 2x2. With a couple of MC4 Y junctions handy I can always switch between the two configurations.
|| 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 ,
2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric, 700 ah @24 volt AGM battery bank. Plenty of Baja Sea of Cortez sunshine.
As for the heat, I'll just wrap the van in space blankets! Joking aside, I'm going to have plenty of insulation inside. Hopefully it will be enough.
Thanks! That's good to know. I was planning on having a shutoff switch on the solar power.
I just redid some of the math for a 2x2 array. When dropping the voltage to two 35V circuits, each 12' circuit of #12 wire that comes on the panels will have a 2.63% voltage drop. Then, even with #8 wire, the added ~18' of wire will have a drop of 0.78% adding up to a total drop of 3.41%. That's no ideal.
I suppose I'll just run them in series unless I find I have a lot of problems with partial shading.
MorningStar TS-MPPT-45
Here's another random question. It looks like this charge controller doesn't utilize a shunt for measuring the current going into the battery, but it has all sorts of metering features available if you buy the TriStar meters, which I had planned on doing. How does it accomplish the metering accurately? Also, what happens if you run devices at the same time you are charging? Does that throw off the meter's estimation of your SoC?
|| 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 ,