Gags, Humor, and Solar Power
I recently managed to replace the misbehaving Outback SmartHarvest 20A MPPT solar charge controller with a proper Morningstar controller, and now all is right with the world. The Outback was exceptionally naughty right out of the box, and its replacement was no better. The USA office of Outback appeared to be manned only with sales types and customer service reps, with no technical staff on board. The sole solution seemed to be trying another unit. So I did, but opted for the Sunsaver. Since I could not in good conscience sell or even give the SmartHarvest away, it is now residing in the local landfill, where it belongs.
With Morningstar’s remote temperature sender attached to one of my batteries, the Sunsaver now operates in tandem with their TriStar MPPT 45A. The Tristar handles everything coming in from the 360W roof panels, while the Sunsaver serves to feed power in from the 200W ground panels via the external rear wall plug that comes standard with the Four Wheel camper. (A simple internal wiring harness change was needed to feed both of these units to the same battery pack for my installation.) Four Wheel uses the less expensive “PWM” version of the Sunsaver in production, since most buyers who use solar install one panel on the roof, and/or plug in a ground panel. As long as the total input doesn’t exceed 6.5 amps, you’re good to go and no harness changes are needed.
PWM solar charge controllers are normally the way to go for most installations, since they do the job at much less cost. As long as the panels are joined positive to positive and negative to negative (a parallel connection) and the wire gauge and lengths are able to handle the resulting amperage without too much loss from resistance, it’s the smart way to go.
My setup, however, would throw way too much amperage through the standard 10Ga roof wires and the 20-foot ground wires when connected in parallel, requiring heavy cables to avoid the big resistance losses that would result. Since I was not about to replace the roof or ground wiring with heavy cable, I chose MPPT controllers that would allow me to add panels without stressing the wiring much. This is done by connecting each of the two panel sets in series, which boosts voltage instead of amperage. They each mimic those old D-cell flashlights, where the batteries are stacked in series to power a higher voltage bulb. The key is that standard 10Ga wires can handle high voltage, but not high amperage. Boosting voltage does not aggravate losses in wire efficiency, and allows much longer runs of thinner wire without sabotaging panel power output. The TriStar is adjustable in its charging voltages, so it was an easy matter to match it with the charging routine of the Sunsaver. I turned off equalization on both controllers (a once-a-month charging routine to slow battery sulfation) to avoid competition and because I use a dedicated desulfator, the Battery Life Saver (BLS-12/24C, due to my pack’s capacity rating. The BLS-12N is good up to 200Ah).
So, I’ve got about 72VDC coming in from the roof, and 36V coming in from the ground panels, each leading to a dedicated controller. MPPT controllers are the only type that can deal with such high voltages, since only they can convert the voltage into charging amperage. The only heavy wiring needed is between the controller and the batteries, which is a universal need anyway. In particular, stringing two sizable ground panels together in parallel over 20-foot wires would be quite a wasted effort due to the losses from electrical resistance. The only ways around this are to either bulk up the wire gauge to heavy cable, or shorten the ground wire lengths to just a few feet, which negates the whole point of having ground panels that can be placed out in the sun some distance away. Adding more panel wattage to compensate simply aggravates the inefficiency problem. Better to keep the amperage to that of a single panel, and wire them together such that voltage is ramped up instead. This makes keeping lightweight 10Ga wiring practical.
Now, two MPPT controllers obviously cost a heap more than one, which points out the liability of high voltage series wiring. That is, shading one or more panels with this system can pretty much collapse the output of the combined panel array. So, parking in the cool shade and setting out the ground panels with everything connected to one controller doesn’t do much for you. What helps is having two independent charging systems, one for the ground panels and one for the roof panels. Park under a tree, and while the roof system becomes nearly useless there, the ground panel controller is still sawing away with whatever panel wattage is connected to it, out in full sun. In my case, that’s 200 watts. That may be too little power to fully recharge 420Ah of batteries, but it’s still a definite help.
Similarly, parking the rig under overcast skies calls for maximum charging power, since the roof’s 360 watts of power will not be adequate to recharge that much battery capacity after a long night of raiding the 12V fridge and watching Ernest Saves Christmas several times. In that case, deploying the ground panels will display a total of 560 watts of panel potential. Naturally, both overcast and the inability to aim the Intrepid’s roof-mounted panels toward the sun takes its toll on power output, but you get the point. However miserable or good the charging conditions, two independent charging systems allow for some flexibility in maximizing whatever charge is available, and making those MPPT systems allows for retaining practical wiring gauges and lengths. It’s all good, if you have a big battery pack and can handle the initial cost.
Now if you’ve suffered through all this, here is your reward, though some may not think it so. It’s a short clip from Ernest Saves Christmas, with the late Jim Varney in his “Ernest P. Worrell” persona. There are three other very different but carefully crafted characters he does in the same movie. The “Ernest” character makes one assume that Varney was a one-trick pony based on goofy slapstick, but in truth, he studied Shakespeare at the Barter Theatre in Abingdon, Virginia and was considerable in his adaptability to varied roles. By age 17, he was performing professionally in nightclubs and coffee houses. The Ernest P. Worrell character simply caught on (along with his “Auntie Nelda” character), and he made more breakthrough commercials than you can shake a stick at before beginning a TV series and then moving on to some awful but fun movies. Enjoy.
Always enjoy reading your posts, with their interesting detail.
So I take it your two controllers are “playing well” together? I’m about to make a similar choice and haven’t decided whether to do two or one. A friend did a similar setup to yours but with Victron controllers and they did not seem to play well together (only one would go to absorption stage, etc.).
I’d be interested in your thoughts, since you have a similar setup. I noted where you said that if the roof panels were in shade, and the ground panels in sun, but you only had one controller then the shady panels would kind of wipe out the sunny panels. But would they really if the two sets were connected in parallel (but each set was in series)? Or maybe you can’t do that because you have such a large voltage difference between your roof set and ground set…
I have a set of two 100 watt ground panels, that run at 17.8 volts Vmp (so 35.6 volts since they are in series). They are currently humming along happily on a Morningstar Sunsaver MPPT 15 (I nearly went for the Outback that you did, whew!).
I’m going to be mounting a set of two 135 watt roof panels, also in series. They are at 17.5 volts Vmp (35 volts in series).
So I’ve narrowed it down to two separate Sunsaver 15 MPPT controllers (with, unfortunately, two separate RM-1 displays), or one Morningstar Prostar MPPT 40. The MPPT 40 is appealing because you can program it right from the display on the face (I don’t run Windows, so have to hook up with a friend to make changes to the Sunsaver in MS View). A slight negative is that that super duper display is only on the face of the unit, which of course will be mounted well out of normal sight. But I recently found out that the RM-1 remote display will run with the MPPT 40 (even though it has less functionality).
My thinking is that if the two strings (ground and roof) are in parallel then it won’t matter if roof is in shade and ground in sun. Unless I’m wrong 😀
I could also potentially save a little money and get the Prostar MPPT 25, since it doesn’t hurt the controller to “over amp” and there are few times that would ever occur (and by definition those would be “fat” times and I’m set up to survive in lean times). But that’s a minor point compared to deciding on the two Sunsavers vs. the one Prostar MPPT.
The other possible choice is two Blue Sky 1524’s (15 amp units) with their “makes two controllers perform as one” IPN remote. But I’d be close on voltage (okay down to like 25º, but nothing close to the huge cushion the Morningstars give with their higher voltage ratings). Blue Sky’s controllers seem oddly low in their voltage limitations (maybe because of the “conventional wisdom” that RV’s should always wire in parallel…?). Programming from the display screen, which is a plus, and a shunt based meter built into the display (but I already have a separate one of those).
By the way, I did some tests with my to-be-mounted-on-roof pair of panels. I laid them flat on the ground (as I don’t plan to tilt them on the roof – have the ground panels for that), in full sun, and then proceeded to do some tests. First off, was how much would I “lose” by using a 15 amp controller. Turns out, basically nothing. At noon in early August the most the Sunsaver put out was 14 amps, in full sun. I didn’t get any “over amp” warning messages. Next was would series be a mistake? I could run them in parallel without too much trouble since they aren’t going to be off on a long tether like my ground panels, but still nicer to use thinner wire.
I laid them out and then set up various shade scenarios and watched my meters. Interestingly, branches making shadows on one panel only seemed to reduce things by the amount you would expect (i.e. I didn’t see evidence of the non-shaded panel being “wiped out”). It was only when one panel was majority shaded AND the second panel also had a fair bit of shade that I saw things really go south (3 amps or less).
Just one experiment one day though. But results were positive enough for me to run the roof panels in series, knowing that I *could* change them in future if necessary (wiring is not buried in roof).
Ha, now my comment is as long as one of your posts 😀 (But then I like long, detail-filled posts.)
As Chairman of the Short Novel Club, I welcome you! Your lapel pin and membership card are in the mail. 🙂
My controllers are both toiling away happily together, the only difference between them now being that the more sophisticated TriStar uses a dedicated “Battery Sense” wire to read the battery’s true voltage instead of relying on the main power cable which can lie, due to resistance. The Sunsaver lacks that ability. Otherwise, the charge routines are set to be identical. Morningstar controllers are specifically designed to work together in gangs fed from separate arrays, so if you’re going to lay out the money, you may as well buy what you know is going to work, whatever that is. Although they can be networked to act as one controller, they also do just fine when working together as-is, feeding the same battery pack. Best results are obtained when each controller’s charge routines are set to match, to minimize times when one controller will bail out and go to float, leaving the other controller alone at the oars.
You are right, connecting two series panel arrays together in parallel is not universally advisable. I was verbally advised this by a Morningstar engineer, because of the huge voltage differences between my arrays. Since you can safely add in identical-voltage individual panels in parallel, maybe it’s quite possible to do the same with two identical arrays where the individual panels are in series – and each array has the same voltage. I’ll let you find that out. 🙂 If I were to insist on one controller for my own (lopsided) system, I’d stick with my TriStar 45 MPPT and rewire the camper’s solar wiring to make the two combined arrays one big series setup (assuming that the controller is rated to handle that much panel voltage – I suspect it is not). With the ground panels stowed, the rear wall solar input would always have to wear a plug that maintains the series connection in their absence. You could do this too, but then you couldn’t complain if the combined array output dropped under partial shading, which your tests indicate don’t apply to your panels. I had no interest in finding out last year, considering that I had to scramble with modding to prep for departure, and that whatever I decided on, I’d have to live with for 7-8 months straight.
You’re already aware that the Sunsaver 15A is rated for 200W of panel power, so a couple of 135W panels technically exceeds that. For the benefit of other readers, it can accept way, way above that without harm, but the output will never exceed its rated maximum. There are positive economic reasons for doing this deliberately, outlined in Morningstar’s MPPT Technology Primer here. A dual-controller wiring diagram is here, and a diagram for wiring a battery bank to even out the “wear” among them all is here. Nice tests you ran, which points out that in actual use, the average user be lucky to even approach the specs. I have a dim memory of some newer panel types dealing with partial shading much better than in the past, so it could be that your panels may allow you options that others don’t provide. At any rate, it’s gratifying to see someone out there who tries, tests, and measures.
I too don’t have any Windows platforms, so getting fancy with programming is not an option. Plus, I’m cheap, so once my system is set up, the most I will do is basically check the charging phase the controller reaches at what time of day, especially in overcast or after heavy discharge. Since I’m packing an excess of battery capacity, the worst discharge I will normally see is 25% in one night…so I don’t really geek out on my pack except to occasionally make sure that it is reaching float for awhile before the end of the day. Other than that, I just use it without thinking much about it. I have other things to take my attention. Recovery in overcast has not been great, but that has been without any ground panels added, so far. That may move up to mediocre now, but so far I haven’t hit a string of weather that put much of a dent in my power usage habits.
I suspect that Blue Sky’s unit is designed with low input voltage limits to save on cost by aiming it at a market that doesn’t need quite as much adaptability. 45VDC input may not stand a chance with my system, but it would do for RV folks who assume that solar panels are solar panels except for wattage, and that you can just string on more like Christmas tree lights. It appears to be a consequence-free environment.