have i destroyed my panels? please help!

[barkarius]

welll...

i have 5 24v 94w solar panels, they are solar glass (2 panes with the cells sandwiched between)but i dont think that makes a difference. i dont beleive i knew enough about pv installations before installing them, but i was in a rush... after mounting the panels i wired them parallel to each other, then connected them to the charge controller, a prostar 30-m, which i had already connected to 12 2v 750ah batteries, which were connected to a 1500w 230v sine converter. i measured the output of the panels, i cant remember exactly what it was, but it wasnt right. i rechecked the wiring and found i had managed to cross the + and - in 2 connector blocks. i rectified this, and rechecked the voltage. unfortunately i didnt notice i had left one of the meters probes in the amp measurement socket, and i also hadnt noticed another block which was wired reverse... as i pushed in the probe, there was quite a big spark, and the positive probe melted about an inch.... never do anything in a rush...

so i disconnected the panels from the controller, rectified the other reversed wire, and tested the open voltage, now 23.2v. i then disconnected the panels from each other, and tested them 1 by 1, each giving more or less the same reading. apart from one. this panel gave no reading, upon opening the junction box i found the 4 diodes (blocking i think) had exploded. ok. i reconnected the other 4 panels to try and get something running into the batteries, although i think nothing is. the controller reports no amps coming from the panels. so after reading info from the net it seemed to me that the diodes are optional as long as i can keep any shading from the panels. so i have tried to solder the panels junction boxs without the diodes, to very little success. if i solder them in the same way as before replacing diodes with wire, i get 0.2v. oops. is it possible the cells are damaged? i cant see any physical signs. at first i thought it might be the controller as well, but the self test come back fine.

now my problem is these arent my panels. im just finishing building a house, and these were my last job. if i have destroyed these im in quite a lot of trouble! i should have been concentrating more, the problem is im starting another one while finishing this one, which is good for neither.... this is my first time installing solar panels, ive only ever done 230v installations in the past. (im in spain, no 110 here.)

any help would be greatly appreciated...

thanks, ben

[jacquesm]

ok, I suspect two things are broken, the diodes in the one panel are toast, that much is clear and I think the charge controller is messed up, or broken.

It would help if you posted a schematic of what you're trying to do here, but from what I can see this is the way it should have been connected:

+ of all panels together to inline fuse of about 20A (24 V panels, about 4A each, 5 of them) or a 5 A inline fuse for each panel, output of fuse(s) to + input of a DC rated breaker, load side of breaker to '+' input on charge controller. - of all panels together to - block of charge controller.

• output of charge controller to DC disconnect load side
  
• input of inverter to DC disconnect load side
  
• of battery to DC discconect input side
  
  

If you're using a shunt:

 '-' block of charge controller to high side of shunt (if any)

 '-' battery - to low side of shunt

 '-' inverter wire to low side of shunt

If you're not using a shunt:

 '-' block of charge controller, battery - and inverter - bonded together

Before blowing up another charge controller I suggest you make a drawing of the above circuit (I don't have any drawing package handy, sorry) to make sure you understand it thoroughly. If you're extremely lucky all you're down right now is a bunch of diodes (they could be shorted internally), test with only one panel to make sure you get it right and include that inline fuse (that would have saved you this time around if you had installed it in the first place).

When switching this on VERIFY ALL VOLTAGES before engaging breakers the first time.

The correct way to switch them on is FIRST the main breaker, wait for the inverter to come up, then the solar breaker. The reason for this is that the battery voltage is stable, but the solar voltage is not and if you don't draw them down with the battery they could be way high. The disadvantage of this way of hooking things up is that if the main breaker goes you have an 'undefined' situation on your hands, but the alternative (hooking the solar controller directly in to the battery) would give you serious trouble in the case of a malfunction of the controller.

A good - but expensive - alternative would be to use two DC disconnects for the solar, one before the controller to protect the array, and one after it to protect the wiring from the batteries. That however is not to code...

[bob golding]

are you using the same meter? if it is a digital one it might still be showing a reading, but it doesnt mean it is the  correct reading. i have  done this more than once with cheap meters.

bob golding

[John II]

Hello Ben: I don't have a lot of info, but I'd borrow or purchase another meter and measure the output of each PV panel separately. If they are each 24v divided by 94 watts = 3.9 amps.

Take the two wires that come from your PV panel and plug them into your tester and measure the amperage. It should be getting at least 3.9 amps in a bright noon day sun.

Put jumper clips to bypass the diodes if you suspect they are not working to do the test on each PV if necessary.

If I had to take a guess in the dark, I think it's possible that your tester is damaged, your diodes in your PV's may need replacing, and It's highly possible that your charge controller may not be functioning properly. At least those are a few things I would look for.

John II

[Opera House]

Those burnt up diodes are likely in parallel with sections of the panel.  They are there in case a leaf sticks on one section or in case you do what you did.  Unfortunantly, this is the reason they suggest that you fuse each panel seperately.  Those diodes are supposed to the brunt of the current till the fuse blows.  If you are lucky, the diodes shorted out.  If they opened, all the current went to the panel which is really just a bunch of flat diodes in series.  Enough current could have fused the interconnects.  As said before, thst the panel by itself.  Panels are basically current sources and it is ok to test them by using a meter in the current mode on the terminals.  Make sure the meter is still correct.

[laskey]

Even thought only one set of diode has failied in a visible way all the other sets have probably failed too... Often a blown diode looks like a good diode.

If you haven fried the diodes, and there are no fuses to blow, then you've fried the panels (but that's hard to do so I don't suspect it yet.)

Check your wiring again too.  Make sure it's right.

Cya,

Chris

[richhagen]

First, don't panick yet.  At this point I would disconnect every piece of equipment involved and test everything individually to see what works and what doesn't.  (including the meter)  I'm not clear on exactly how everything was interconnected when the short occurred, but I would test all of the diodes in all of the panels.  If they are bad, (shorted, open, conduct in reverse, or forward voltage drop is not consistent with specs for that type of diode) replace them.  It seems likely that you will find some bad ones, fortunately they usually fail to short and this would likely have protected your panels.  If the surge was to large or two long, they could have failed open though and this could have caused excessive voltage across the p/n junctions of the cells and caused damage to the panels themselves (this would be bad). Use a meter that you know works to test these diodes.  Once done with the diodes, I would move on to the panels.  Test the output from each individually.  (if you didn't test and replace any faulty diodes first, you wouldn't know if a dead panel was from the diode or the cells)  

I would build some type of a combiner box to minimize the probability of this type of incident recurring. (it is a hazard of rushing to install new equipment and overlooking something in the haze of excitement)  The minimal lost power is probably worth the added insurance against ruining expensive equipment.

see http://www.fieldlines.com/story/2004/12/12/232315/76

When done with that, I would move on to the controller.  I would feed the controller a known input and then test to see if it performs properly.  If you lack any other power supply capable of feeding a small amount of dc in the appropriate current range, then you could use the tested and reconnected panels.  If the output voltage is in the proper range to charge your batteries, and the measured current matches what should be there based upon the input current and voltage, then all is probably OK.  I would measure the voltages at each stage of charging if possible to make sure that all are within tolerances.    Best of luck, Rich Hagen

[barkarius]

thanks for the advice everyone!

i managed to find a shop that sells the diodes (miracle) and i have replaced the broken ones, and some ones i wasnt sure about just to be safe. the panels are making voltage again, but little (i think?). the most voltage being made in full sun is never more than 23-24 volts. it should be a lot higher right? 30v or thereabouts?

in the junction boxs are 6 metal strips from the cells, i matched them up and measured the voltage, the first 2 make around 6v, the next 2 make 23-24v, and the final 2 make another 6v. the middle 2 make exactly the same amount of voltage as i measure at the outputs. should the first and second 2 be increasing voltage? i experimented by wiring them + to - to try and increase the voltage, but when i did this the voltage died completely. when i disconnect the first and second two it makes no difference. anyone have any ideas?? if anyones interested i can post pics of the junction box on the panel.

i think the controller might be ok, am i correct in thinking if the solar current isnt higher than the battery current the controller cant charge the batteries? the batteries seem to be holding around the same as the panels, though they arent being used much.

i have made a terminal box as suggested, thanks for the idea! hopefully this wont happen again.

merry xmas

ben

[juiced]

While this has nothing to do with your panels not working, im just curious if you followed basic angeling procedures?