12V panel and controller to 48v

jimchitown

Hello all, I currently have 10 homemade panels and a unisolar 144watt and a charge controller running a 12v system, about 1 kilowatt total but only 1/3 hooked up now. I would like to buy a new battery bank and run at 48v.  my panels put out about 21 volts max or so, and I am using a cheapo ebay controller to charge a 12v battery, with 12v loads. I have 2 good 48v inverters, and would like to move up to a 48v battery system so I can use the inverters, my question is how can I use the panels to charge the 48v battery? most of the charge controllers I see ( cheapo ebay )say they work best at 60v from the panels, others I have looked at are not specific. I was considering keeping the 12v side, and hooking up a step up boost regulator to charge the 48v from the 12v battery, which seems very inefficient. so, is there a 48v charge controller that can work with 21v panels? I am not excited about stringing the panels in series to up the voltage, not sure if that would be a good idea. I would be mostly interested in a charge controller I could just hook up to the 21 v panels, and charge the 48v battery, I thought MPPT chargers would step up if needed?  thanks for helping this noob...:) any recommendations?

Estragon

There are boost controllers available from Genasun and Solar Converters, possibly others.

Normally you would want to use a MPPT controller to buck a higher voltage down to battery voltage to minimize line losses and/or allow for the use of smaller wire from array to controller. Is there a particular reason you want to avoid putting panels in series for the higher voltage?

Depending on battery temp, 60v from array to 48v nominal might be a bit low. Veq on a cold bank could be well into the 60s volts. In winter, even Vabs can be over 60v on my bank.

I would be reluctant to charge with a boost converter between controller and bank. It could work, but I'd be a bit concerned about voltage sensing and control being so far removed from the controller to bank.

Photowhit

Not sure where you are reading 21 volts, I'm guessing at short circuit? just reading across the leads  with no load? This would be the VOC number, under load they would be around 17.5-18 volts. PWM charge controllers would generally want 4 of your panels inseries, assuming they are the same size panels. you could run t2 string of 4 panels to the charge controller.
Usually a spec sheet will include a maximum voltage, I'm guessing you still want a cheap alternative to MPPT type panels? Here's a cheap 48 volt charge controller as an example;

http://www.lightinthebox.com/50a-48v-pwm-solar-panel-battery-controller-lcd-charge-regulator_p4765796.html?currency=USD&litb_from=paid_adwords_shopping&utm_source=google_shopping&utm_medium=cpc&adword_mt=&adword_ct=153425486380&adword_kw=&adword_pos=1o1&adword_pl=&adword_net=g&adword_tar=&adw_src_id=2141113852_686903006_41037565171_pla-256709381999&gclid=Cj0KEQjwxPbHBRCdxJLF3qen3dYBEiQAMRyxS2a6ZHJoO5GwgR1VVhvRR_QgwyGGOe6YnqRWLWt6rVUaAkxt8P8HAQ

jimchitown

Thanks for the replies so far. Here is what I was thinking, and paranoid about:  I already have a 12v charge controller, very similar to the one in the link, and a couple of golf cart batteries on my panels. I thought I should just add the 48v bank, and hook a step up voltage regulator to the 12v battery outputting to the 48. I was thinking of using one of these, which I already use to charge a li-ion battery bank I built for travel:

 http://www.ebay.com/itm/900W-Digital-Control-LED-DC-DC-Boost-Module-Power-Supply-Step-up-Converter-O5A4/301994973187?_trksid=p2045573.c100507.m3226&_trkparms=aid=555017&algo=PL.CASSINI&ao=1&asc=41375&meid=41e569c3dd6543a1b0ad3705cf0342d6&pid=100507&rk=1&rkt=1&

not sure what the losses would be though.

I was concerned about such a high voltage, by running the panels in series. Also, is it ok to have such high voltage running through the panels?

I have 11 panels I made, from your standard ebay cells, with the 6tabs on back and 2 strips on the front, 1.8w  .5 volts if I recall, and yes, the open circuit voltage is 21 or so. I also have a 144w unisolar roll and stick on panel.  I would have to measure under the load tonight when I get a chance. 

so basically I am not sure of the voltage vs amperage vs wire size etc. I do not want to start a fire...

also, I have a diode on the end of each individual cable right before the battery. each panel has a single wire running 15ft from my roof...I am using thick stranded speaker cable right now, that is probably a no-no

hope this helps explain my situation, and again, I appreciate any advice...

Jim 

Photowhit

I think you are right to worry, most of the fires I know about were homemade panels. Commercially made panels usually have a series max voltage of 600 or 1000 volts.

I don't know how you would regulate the charging cycles from the DC boost module, unless you then ran the current through a charge controller.  I'm not tech savvy enough to know if that would work.

Estragon

The dc converter appears to output a constant voltage.

To properly charge the 48v nominal bank with 12v nominal pv you would want a 12v charge controller feeding a proportional voltage DC boost converter, which in turn feeds bank. The 12v CC will initially output full available current while output voltage rises to absorb voltage setpoint, eg 12.4v rising to 14.6 absorb, then hold voltage constant while current drops off as bank gets full. When bank is nearly full, the CC drops voltage to float, eg 13.4v. These voltages should all be temp compensated by the CC. The boost converter should multiply the various voltages by 4.

The thing is that the charge controller is seeing battery voltage only indirectly. You could adjust Cc setpoints to compensate and hope any variance stays constant I guess.

Running at low voltage you need much bigger wire. 1/4 voltage means 4x current. Wire and (very importantly) breakers/fuses should be sized for current. Each parallel connection should be protected. 1000w at 12v is 83a. Will need big wire and a short run from pv to Cc.

I have no idea what voltage the homemade panels will take. As photowhit says, commercial panels are 600+v.

BB.

I will humbly suggest that you do not run your home made solar panels in series (or even parallel) for mounting on your roof (or over any sort of flammable materials).

http://forum.solar-electric.com/discussion/comment/59862

12 volts and below--It is very difficult to sustain an arc... 24 volts and other, it is very easy to sustain an arc (DC arc-welding works very well at 24 volts and above).

There are secondary issues with many (some, many, all?) EBay solar cells--They may be rejected lots from a solar panel mfg... One of the failures is the cell can overheat under normal operation.

http://tauscience.com/iris/

And other possible failures:

http://www.pveducation.org/pvcdrom/modules/degradation-and-failure-modes

-Bill

bsolar

youve got 10 panels and worried about peanuts, sad you invested in 48v inverters, i'd hock them on ebay and get some 12v inverters .. i have my panels individually wired with 12awg to a tie-in box near my battery/controller station and the longer runs are over 30' .. i see virtually NO loss from panels on the end vs closer panels with a 12' run, in fact, the efficiency of the actual solar cells used in the panels is WAY more important than losing your cookies worrying over a high voltage setup or fancy chargers .. you want to make up for some loss (if you can actually measure it) do ya? add ONE more panel .. you'll beat any gain you might get with all that fancy wiring and chargers .. trust me on that, if for no other reason than i can prove it to you in a video and a multimeter if you want me to ...

mike95490

sad you invested in 48v inverters, i'd hock them on ebay and get some 12v inverters

I'm pretty sure this is not very good advice.

I currently have 10 homemade panels

This isn't thrilling either.

In a perfect world, to charge a 48V bank, you would wire a PV array for about 80-110Vdc and let a MPPT controller near the batteries, downconvert that voltage to amps, and then run the heavy cables a short distance.

mcgivor

I'm with @mike95490 keep the inverters and get some real panels designed to work at higher voltage, along with a compatible charge controller , the home made panels  are good for educational purposes, good on you for venturing down that path, most wouldn't consider that. Build a real 48v system with all the right components, you'll never regret doing it, trying to hack your way using incompatible components will result not only in frustration, but disappointment, or worse,  assuming of course it is within your budget to do so.

Estragon

It seems to me that decisions about string and bank voltages involve factors we don't know.

How far will the array be from the batteries? 20' might not make much difference. 200' probably will.

Aside from charging a travel battery, we know nothing about loads or function. As Mcgivor speculated, this could be an educational project. Maybe it's to keep a few lights on during outages at a weekend cabin. Maybe it's a full-on off-grid house with pumps, fridges, etc. Without knowing more, our balance of system advice really isn't worth much.

I do agree that using homemade panels in a good sized production system isn't wise.

BB.

Just to give you an idea of what DC current does for arcing vs AC:
https://www.youtube.com/watch?v=Zez2r1RPpWY
-Bill

jonr

It's interesting how the arc self-extinguishes after awhile.  Maybe he is using rectified and filtered AC and the filter capacitors run down to the point where it's pulsed DC.   If so, then the DC arc is initially ~308V vs the AC arc at 220V.  Yep, the heating elements glow much brighter on DC.

mike95490

That's what "touch safe" fuse holders in combiner boxes will do too.  Never open them when powered.  I use the DC rated breakers, cost about the same for a breaker, as a holder and fuse,

BB.

Yes, it looks looks a 4 pack of capacitors in the upper right (and one diode so 1/2 wave rectifier).

So, it is a bit of a not equal 220 vac vs ~(sqrt 2 × 220vac peak) 311 volt pulsed dc current.

I did not check the video that closely. Thank you.

- Bill

jonr

I'm curious just how long it takes to get an arc to extinguish.   I could imagine that in some cases, brief interruptions in DC could make it safer.  For example, a 1 ms dropout every 100 ms would have negligible effect on total current flow while possibly preventing sustained arcs.  The issue came up when looking at driving a home water heater with PWM DC and trying to use the existing thermostat.  Ie, what pulse width/duty cycle would be safe?

Estragon

@jonr - Just guessing, but it seems to me the answer would be "it depends". Surely factors like humidity, temp/air density, increasing gap as conductors melt etc. would be factors?

jimchitown

thanks for the help and interest everyone! it is greatly appreciated. I will get some pics and check the voltage tomorrow when I m home during the day. I do have one question about the voltage of the panels though. what is the short circuit voltage? do you mean while it is connected to the charge controller or with the leads from the panel shorted? that should give 0 volts no? also, I am curious why the original unisolar panel I bought had the positive and negative connected? I have 6 panels I built just sitting in storage until I move or hook them all up where I am, and wonder if they should be stored shorted. I will provide more details, including how I built my panels and what equipment I have now, and what I planned on doing. the links provided above are very interesting, especially the one showing panel hotspots and the flaws in cell manufacturing. I bought a kilowatt of cells a few years ago and have been happy so far. I bought them from ML solar.  more info to come tomorrow or Sunday the latest, again I appreciate everyone's advice and help...thanks.

BB.

Some non- crystalline cells (thin film) should not set in sunlight without a load (or shorted outputs). The open circuit voltage can damage the panels.

From what little I have read. Those panels are used on GT type systems.

Does that apply to UniSolar panels, I have no idea.

- Bill