Troubleshooting solar setup... or troubleshooting my troubleshooting!!

[thomm225]

Originally Posted by CatNewBee

it is not unit is not uncommon with solar panels, cells are huge and tiny fractures can cut part of the area of a cell, the remaining part still outputs some current. Cells are in series, so if one cell gets weak, the whole string is limited to the current of that cell. Also hotspots due to partial shading and resulting reverse currents can fry a cell.

Thanks for training me on series vs parallel. I think I had it first time in 1975.

Understand your point though but all 4 panels?

I probably shouldn't get involved in these discussions because it goes so frustratingly slow and the person trying to make the repairs makes the same mistakes newly hired techs do with inconsistent testing, uncalibrated or old test equipment, and trying to over analyze

OP is doing a fine job it's just that when you do it for a living you learn short cuts and you know usual problems

But troubleshooting is fun so I throw out a few hints............like keep it simple.

My panel test straight to a battery would quickly let the OP know if the panels were charging the battery at about the same rate. Do the test twice with #2 and another panel plus have the battery drained to the same point for each. Record charge time to 14 volts. Similar solar/sun conditions

Good Luck......and use sound logic

[thomm225]

Originally Posted by CatNewBee

it is not unit is not uncommon with solar panels, cells are huge and tiny fractures can cut part of the area of a cell, the remaining part still outputs some current. Cells are in series, so if one cell gets weak, the whole string is limited to the current of that cell. Also hotspots due to partial shading and resulting reverse currents can fry a cell.

The last thing I actually worked on a couple weeks ago had a similar schematic to the link below. (problem was no display) It's an old CRT Monitor and not many of our young guys troubleshoot this sort of thing but they are great with computers, drives, software , admin, etc

http://www.jstookey.com/arcade/WG_25...mplete_lg_.jpg

Point is you better know a bit about series and parallel before beginning to troubleshoot this and quite a few other things.

I have actually also repaired a few CPU's ……...this in the 1980's when CPU's were several large circuit boards

We used to select the drive or paper tape reader by switches on the front using BCD

The OP's problem deals with 4 solar panels, one controller, batteries, wire and a few connections.

[basssears]

Originally Posted by noelex 77

It would be sensible to repeat the Isc measurement with another multimeter. The testing results do not make sense. Voc and and Isc are not the same as the measurements with the controller which is at Vmp and Imp, but they should correlate well.

OK, yet another update. I retested using both the inline meter I used before for Isc and my clamp on meter, and the values are very close to the same:


Short circuit current is supposed to be 2.84a:


So that would seem to indicate that everything other than #2 should be doing fine, but the outputs are still lame (and oscillating up and down, not sure if that means anything, but my other panel setup on my van doesn't do that).

Renogy has agreed to have me send them panels #2 and #4 but don't seem interested in testing #1 & #3.

I borrowed my fire department's FLIR camera and took a thermal image of each panel, know idea how useful it is as a tool but thought it would be an interesting experiment.

Panel #1:


Panel #2:


Panel #3:


Panel #4:

[basssears]

And Renogy's reply to me:


I do sincerely apologize for the miscommunication on my end, I have reviewed this e-mail thread searching for potential problems or errors that I might have missed and unfortunately, I did miss something very important. When you drilled into your solar panels to mount them in your configuration, you did void the warranty of these panels as any type of modification to the panels such as drilling or removing the MC-4 connectors will result in voiding the warranty. When you remove our MC-4 connectors we can no longer warranty the panels production as we cannot ensure that the modification was done correctly and is not resulting in any kind of wattage reduction. Please be advised that we will not be able to get these panels in for testing, as they have no warranty.


Guess you're just supposed to lay these gently on the deck and never let your boat move... no wonder everyone just runs a Honda genny...

[thomm225]

Originally Posted by basssears

And Renogy's reply to me:


I do sincerely apologize for the miscommunication on my end, I have reviewed this e-mail thread searching for potential problems or errors that I might have missed and unfortunately, I did miss something very important. When you drilled into your solar panels to mount them in your configuration, you did void the warranty of these panels as any type of modification to the panels such as drilling or removing the MC-4 connectors will result in voiding the warranty. When you remove our MC-4 connectors we can no longer warranty the panels production as we cannot ensure that the modification was done correctly and is not resulting in any kind of wattage reduction. Please be advised that we will not be able to get these panels in for testing, as they have no warranty.


Guess you're just supposed to lay these gently on the deck and never let your boat move... no wonder everyone just runs a Honda genny...

They'd go out of business if they allowed every backyard mech/tech to drill into their panels or play around in the J box and expect a refund if it didn't work later.

Haven't you been working on this problem for almost a month?

Should have taken 30 minutes or so.

We're talking 4 panels, a controller, batteries, wire, and some connections......

It appears you may only have one problem possibly with panel #2. (depending on if you checks are really sound) I'd still check the wiring and if you do another elaborate test beat on the panel and move the wires around and see if you get a change.... IN CURRENT. I know how much you guys like to rely on your current checks

Good Luck, and thanks for posting. My Victron MPPT 75/15 arrived today, and it looks like a quality product!

[basssears]

Originally Posted by thomm225

They'd go out of business if they allowed every backyard mech/tech to drill into their panels or play around in the J box and expect a refund if it didn't work later.

The "if you removed the MC4 the warranty is void" seemed a little harsh (and not in their warranty details) but I see your point.

Originally Posted by thomm225

I'd still check the wiring and if you do another elaborate test beat on the panel and move the wires around and see if you get a change....

I did try to wiggle whatever I could, but now that I know I have nothing to lose I'll probably marry the two better panels and see what I have. I did also go straight from the J box and didn't see any change.

Originally Posted by thomm225

My Victron MPPT 75/15 arrived today, and it looks like a quality product!

I certainly don't have any complaints about the Victron, although my only other experience is an older Tracer which works fine but isn't the most user friendly.


Thanks for your input.

[Captain Bill]

Do you have a copy of the warranty? If so does it say you can't drill holes in the frame for mounting. If it doesn't specify what voids the warranty they can't make it a condition later. Look up the magnuson-moss warranty act.

[basssears]

Originally Posted by Captain Bill

Do you have a copy of the warranty? If so does it say you can't drill holes in the frame for mounting. If it doesn't specify what voids the warranty they can't make it a condition later. Look up the magnuson-moss warranty act.

Nope, nothing in the warranty that specific, very general:
https://www.renogy.com/template/file...s/Warranty.pdf


Just not sure I'm ready to battle this over $80/panel (even though I'm pissed at my hack self for getting in this position)...

[thomm225]

You should still test panel #2 and see if it will charge a battery at a similar rate as another panel.

Use wire with no splices etc and make sure the connections are good. (different than what is attached to it now)

Plus do a good visual inspection on the panels looking for cracks, holes, dents, etc...…

I just keep hoping your problem ends up being like mine which is usually wiring. Once, my negative wire had been almost completely cut in half by getting caught under the lazarette lid. (but still hard to see the cut) The panel would kinda work if the wire was in the exact right position

I found three cuts in my negative wire and had to cut it all the way back to like 6" from the J Box...….where I spliced it with a wire nut. That nut has already been crushed though by the panel probably when I was stilling on it or standing on it to mess with my outboard. Either steering it or pull it up out of the water

I have other splices and to get my panel to charge sometime I have to move the wires at the splice with my fingers!! I like the logic in troubleshooting not necessarily the repairs. I need to replace the wire totally and have gotten some new wire but haven't gone the next step

[atoll]

Originally Posted by curtis742

Simple test, no meter required. Place panel in direct sunlight, take positive wire in right hand and negative wire in left hand. Touch both leads simultaneously to tongue. If pain is felt that panel works. Repeat with remaining panels one at a time. If a panel produces less or no pain it is defective
Check diode with ohm meter.

better yet, get a small child to test for you,their senses are much more acute,and have not been dulled by whiskey,cigarettes,spicy food etc

[noelex 77]

Those Isc readings look more realistic. Apart from panel number 2, the Isc and the Voc readings are exactly what I would expect from normal healthy panels. The only unusual aspect is the variation in Voc, which I would have expected to be a closer match on identical panels, but even the lowest reading of 21.03v is not abnormally low.

Normally Voc and Isc testing is enough to show a panel is healthy. Voc demonstrates the panel’s ability to deliver voltage and Isc shows a panel’s ability to deliver current.

It should be remembered that by definition these testing points occur when the panel is putting out zero power. So they do not directly test a panel’s ability to deliver any power at all.

However, I am at loss to think of a fault that could occur in the panel that would produce a normal Voc and Isc, but result in such poor performance at Vmp and Imp and therefore not allow the panel to produce the power that it should. Obviously such a fault exists or there is some fault with the controllers (both of them), the way they are set up, or the way they are connected (for example are they set for a 24v rather than a 12v battery).

You have done enough testing for a lifetime, but I cannot remember you measuring the current output when a panel was directly hooked up to the battery. This would eliminate the remote possibility of some of the potential problems with the controllers that have indicated above.

Maybe it's time to use the hammer as you suggested in post #27, but have the beer first, it will hurt less .

[thomm225]

Originally Posted by noelex 77

You have done enough testing for a lifetime, but I cannot remember you measuring the current output when a panel was directly hooked up to the battery. This would eliminate the remote possibility of some of the potential problems with the controllers that have indicated above.

This is sort of what I've been saying all along.

Forget all the elaborate tests and simply hook the panel (#2)
to a battery that has been drained to a certain voltage that you determine.

Let it charge to 14 volts or so max. Maybe 13.8. You decide. Check VOC before hookup if you like with meter

If you must check the current go for it but it must be done at the same level of battery charge are it could be different

I would just monitor voltage and time on #2 and a panel you believe is good. if they equal or are anywhere close, the panels are good

Ps. Use good wire and have good connections

Btw, when an experienced tech comes after another tech to work on a problem the first tech couldn't fix, the experienced tech will not trust anything the first tech did. He starts completely from scratch...…..this is my point with the tests

The simpler the test the better. A panel. Two wires. One battery. Two tests. monitor voltage and time. Compare results

[Paul Elliott]

Originally Posted by basssears

So that would seem to indicate that everything other than #2 should be doing fine, but the outputs are still lame (and oscillating up and down, not sure if that means anything, but my other panel setup on my van doesn't do that).

A panel by itself won't have oscillating outputs -- that's the controller operating, perhaps correctly. You provided Isc and Voc data, and the relatively small difference between the clamp meter and the direct-connection ammeter can be ignored for now; they're in the same ballpark and that's all we need. Panel #2 looks bad.

Your power measurement (Watts) doesn't tell us much, since we don't know the battery voltage while you are measuring this. I suspect that the controller is doing what it should, and keeping the battery from being overcharged. That, or it's broken.

If you do make the direct-to-battery connection test, don't worry about discharging the battery to any particular level. Just measure battery voltage and panel current. If the battery is somewhere between 10 and 16V (let's hope it is) the panel current will be only slightly less than Isc if everything is right. You won't damage the battery by overcharging if your test only lasts a few minutes.

But honestly, a properly performed Isc and Voc test should tell all you need to know about your panel.

[basssears]

Originally Posted by Paul Elliott

Your power measurement (Watts) doesn't tell us much, since we don't know the battery voltage while you are measuring this. I suspect that the controller is doing what it should, and keeping the battery from being overcharged. That, or it's broken.

Battery voltage was 12.1 or lower, sometimes 11.9 or so, as I had the high beams and driving lights on to make sure I had a load... the battery definitely wanted more power than the panel/controller was giving it, so I don't think it's that.

Originally Posted by Paul Elliott

If you do make the direct-to-battery connection test, don't worry about discharging the battery to any particular level. Just measure battery voltage and panel current. If the battery is somewhere between 10 and 16V (let's hope it is) the panel current will be only slightly less than Isc if everything is right. You won't damage the battery by overcharging if your test only lasts a few minutes.

I still plan on trying this (nothing to lose as I voided my warranty with my mounting setup), only problem with the straight voltage test is even if the panel brings the battery up in voltage I still don't know how much the panel is really putting out... I mean, a 5 watt panel would eventually bring my battery up, just like a 50 watt panel putting out 5 watts. And I unfortunately don't (currently) have a "known good" panel to use as a baseline time wise.

But we shall see.

[thomm225]

I cannot believe you destroyed 4 panels by drilling into the frames.

Did the panels ever deliver the proper wattage after you drilled the holes?

The post above about charging for a few minutes is incorrect. Also, if the battery is fully charged there will be less current flow than if the battery is low

Example. I allowed my batteries, two 12 volt deep cycle in parallel to be discharged down to 12.4 volts throughout a whole day with equipment on in the boat and my 65 watt panel facing the cockpit sole

I was changing the location of my panel and had to disconnect anyway so I did the test. After reconnect, I turned the panel to the sun at 5 pm and checked VOC which was 17.8, and reconnected. Voltage now is 12.4

After about and hour the voltage was back up to 13.1 when I left and climbing

Maybe you could compare it to one of the camper panels if they are near the same size. In other words do the same test with a known good. It's not rocket science