Fried Solar Panel?

Van Living Forum

Help Support Van Living Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Yawppy

Member
Joined
Sep 3, 2014
Messages
16
Reaction score
0
I had installed a Windy Nations 100W or 150W panel about a year ago, around 10 months. I never got around to putting in an inline fuse, and really did not connect to the solar charger properly as I found out a couple of weeks ago, lessons learned, and this time plan to do an inline fuse, of 10 or 15 amps.

Just finished a course in relation to job rehabilitation training in electronics, learned a bit through VA and have gone to see what is going on, so heres the low down:

I was happy with the panel, was working great, till a couple weeks ago. Had just purchased a 12v F/F, and was done, just a few more tweeks here and there, and was content, but as I was showing it to a friend, on a bright sunny day, noticed no power was coming in, went to check, and as I was checking the connections to the solar charger, one of the bat wires came loose from the connector. ( Lesson learned pliers and electric tape do not work when crimping)  So figured it be a quick fix, and with just finishing the VA course, knew I needed better connectors, and a proper crimper.

So, got the connectors and crimper etc.. went to do the fix, got a spark removing the bat cables, but no issue, otherwise, till I removed the connector from solar charger and solar panel, another spark/short before even unscrewing the connection... which seems to have shorted or fried the solar charger, and even the panel. I bought a new panel, cables, and charger, all work fine, now am wondering if the panel is shot, or if there is a fuse in the black box where the cables are. yet when I connect the original W.N panel to the new solar, dead no power coming in, ( again wish I have fused inline)

My question is about the original panel, do they have a fuse in that little black box, hoping so, or is it just fried beyond repair?

WINDY NATION PANEL (THINK ITS FRIED?)

SOLAR CHARGER

NEW 100W SOLAR PANEL
 
When you took that electronics course ,,,,,did you get a meter?
 
Yawppy said:
now am wondering if the panel is shot, or if there is a fuse in the black box where the cables are. yet when I connect the original W.N panel to the new solar, dead no power coming in, ( again wish I have fused inline)

My question is about the original panel, do they have a fuse in that little black box, hoping so, or is it just fried beyond repair?

There are bypass diodes in the black box on the back of your solar panel, no fuse.  The diodes could be blown, but I doubt it.  If blown they are replaceable.

The correct way to test a panel is to place it in full sun and bridge the two leads with a voltmeter: that will give you the Voc (open circuit voltage, should be listed on a sticker on the back of the panel or look it up).  Next bridge the two leads with an ammeter: this will give you the Isc (short circuit current, again look at the specs).

If they are close the panel is good.  If no voltage/current: it could be a number of things and you have to probe to find out what.  If Isc is low: it is probably a diode.  IMO you would have to work hard to fry the panel.

 -- Spiff
 
Sounds to me like maybe the panel died before you had trouble with the connections.

But as Spiff said, check the output with a digital meter.

The panel needs to be in full sunlight, no shade or shadows.

Check the voltage, meter set for 20 volt range, or the range closest to that. You should see something close to 17-20 volts.

If you do, unhook the test leads,  then change the meter over to 10A scale, move the positive test lead over to the 10A jack, and then short the panel wires to the test leads on the meter. You should probably see something at or about 5 to 6 amps.

If it fails either or both tests, you have something wrong with the panel, the diode, or the wiring.
 
never hook the panels up before the battery. never disconnect the battery with the panels hooked up. I have never seen a panel fry not saying it can't happen but I have never seen it. get a meter and start testing. highdesertranger
 
Do the tests as listed above. The panel has a warranty if you don't take it apart. I agree with the "controller is broken" opinions.
 
highdesertranger said:
never hook the panels up before the battery.  never disconnect the battery with the panels hooked up. 

THIS THIS THIS!!! ^^^^^^^
 
Some panels have a Voc or Volts open current above 20v. You may have to switch to a higher setting for that test.
 
The sequence of connecting panels to controller to battery. I have tried to research this. The problem is the danger of shock from solar panels. Not that any damage will be done to any component. If there is any other information I would like to read more.
 
I have a PWM controller.  If the battery voltage is below 14.whatever It is in bulk mode, the transistor switch is on.  When the battery gets up to the constant voltage level, absorption, boost, or whatever you call it, the transistor switch gets turned off and on repeatedly to keep the voltage at the constant voltage level.  If the controller is connected to the system including the roof fan, phone charger, and led lights and the battery lead gets disconnected the panel full 20 volts gets applied to all the stuff.  If there is enough stuff connected to keep the circuit loaded the voltage doesn't go too high.  Otherwise, the voltage goes too high.  The fancy ten speed controller for the fan can poof with enough volts.  Once the loads are all dispatched to the great beyond and no longer keeping the voltage down the controller itself can go poof.

If the engine is running and the alternator is alternating and all the  vehicle electronics are turned on the same thing can happen when the battery is disconnected.  The alternator output spikes and with no battery load all kinds of bad things happen.  Usually alternator diodes roast and lots of electronics need to be replaced.  

In both cases the systems could be designed to operate without the battery.  It would take more circuitry to do the voltage regulation to have it work with the battery missing.  It could be done but it would make it more expensive, less profit, and of very little additional value to the consumer. 

An MPPT controller does what a PWM controller plus it reduces the voltage, increasing the current.  That requires a circuit into which the current can flow.  Without that current flow the voltage will rise.  It will have similar but more expensive results. 

This spring, February 2017 as I recall, there was a thread on the reddit vandwellers forum about a solar failure.  It was in England so there wasn't much sun during winter.  As the spring brought more sun the 300 watt panel blew the 7.5 amp fuse between the controller and the battery.  It destroyed the MPPT controller.  It really can happen.  Yes, the fuse size was terribly wrong.  It was the fault of the idiot who installed it.
 
Weight said:
The sequence of connecting panels to controller to battery. I have tried to research this. The problem is the danger of shock from solar panels. Not that any damage will be done to any component. If there is any other information I would like to read more.

From my Morningstar TriStar manual:
• Never allow the solar array to be connected to the TriStar with the battery
disconnected. This can be a dangerous condition with high open-circuit solar voltages present at the terminals. 
https://www.morningstarcorp.com/wp-content/uploads/2014/02/TS.IOM_.Operators_Manual.04.EN_1.pdf

I think this is written for high voltage stationary systems.  With the small 12V, 24V or 36V mobile systems this should not be a problem, but IMO it is still good practice.

As to danger of shock, from IEC 60950-1:
0.2.1 Electric shock
Steady state voltages up to 42.4 V peak, or 60 VDC are not generally regarded as hazardous under dry conditions for an area of contact equivalent to a human hand. Bare parts that have to be touched or handled in the normal course of events should be at earth potential or properly insulated. 

Morningstar recommends a fuse between the solar panels and the controller (60 amp for my 45A controller).  The battery side should always be fused as close to the battery as practical.  I use ANL fuses by MaineSail's recommendation and they are an easy way to disconnect panels and battery.

 -- Spiff
 
Spiff. What you posted reinforces my understanding that the problem is not equipment damage. What I found for my 12 volt system is it is necessary to fuse both ends of the cable from the controller to the positive battery. My panels are in parallel, and each requires a fuse to prevent the other panels feeding back into one panel if a failure/short in one panel. The internal panel circuit may not handle the 18+ amps from three remaining panels. I have a DPST switch to disconnect my controller from both the panels and batteries with one action. I do not have any accessories that would be fed from the solar if the battery is disconnected. My solar controller is connected to the battery. My accessories are only from the battery. Disconnect the battery from the solar, no accessories to be fried by the solar. The controller will not pass current above the set point charging voltage, 14.8. If it does it is not a good controller. I got a nasty shock from a 24 volt battery through an injury on my finger. Not life threatening, but I will not do it again.
 
I am wondering what happened when they stuck the meter leads up the panels MC4's.
 

Latest posts

Top