Panel shows voltage on its own but 0 in series.

[diWhy]

I'm making up 7 panel stings. I've come across one that's doing something I'm totally baffled about.
I can measure voltage when it's not connected to anything but if I connect it in series with just one other panel, then it outputs zero voltage.
All I can think of is there is something in the small black box the cables are coming from that disables the output when the voltage goes higher than the panel can output on its own?

Stumped.

 

[FilterGuy]

Is it under load in both scenarios or are you measuring open circuit voltages?

Without more data it is hard to say what is happening. However, I would guess it has a bad cell and the bypass diode is turning on when in series.

 

[diWhy]

Hi, sorry, was not sure what else to add.

Yes, I'm only measuring open voltage at the moment, no load since I'm just setting up the strings right now.
When I bypass it in my string, the string is as it should be. As soon as I connect this panel to another in series, I get zero voltage across the two connectors, -/+/

 

[Mattb4]

I am guessing it has MC4 connectors. Did you check to ensure that it is not wired in reverse polarity? Your meter should show if the panel when measured by itself is giving you a positive or negative voltage. Standard is the male MC4 is positive and the Female MC4 is negative.

 

[diWhy]

Yes, MC4 connectors. I tested touching the internal connectors and also using small extension cables I've made for testing.
Polarity is correct or at least, leads to each MC4 are correct, positive to positive, negative to negative. If reversed, it shows a - sign on the meter of course.

 

[diWhy]

I've seen the little diode box open on some videos but is it even worth trying to fix if that's the problem?

 

[Hedges]

Two panels in series is zero volts, but each makes voltage on its own?
Could be connector isn't making contact.

(more common it high resistance contact.)

Are they fully mated?

 

[diWhy]

I tested all panels direct to panel connectors and with the short 2' extensions. On all panels I get voltage except this one panel when it's connected to another panel. I checked the connectors over and over too, moving them around while testing and they were fine.

 

[robbob2112]

Are the connectors original or replaced? If replaced is it possible they were reversed so when you hook it up to a correct one you get zero volts. Of course if that is true you will have a negative voltage reading on it using the correct end?

 

[diWhy]

Right, my mater shows when the polarities are reversed by adding a minus sign in front of the digits.
The penal has the original connectors but you know what, I'll go check yet again, just in case I missed that .

 

[diWhy]

No, polarity on the connectors check out too but that was a good call.

 

[diWhy]

I can't even find replacement panels anywhere so far.
Original is Mage / Powetec Plus, 250 / 6 PL US AC.
Replacement could be MAGE POWERTEC PLUS 250/6 MNS
Number of cells 60 (6x10)
Most important part is must fit. Originals are 65.15 L x 38.94 W x 1.54 H

I watched a video long ago about fixing the diodes in the junction box, has anyone ever done that?

 

[Hedges]

sure, people have.

If you can open the junction boxes, measure voltage and measure continuity (or open circuit) through the mated connectors.

 

[diWhy]

I can get at the panel from underneath and can test but the six diodes all acted the way way. I only have a test meter that can tell me if there is a short/current flow however so not sure how I could test the diodes while they are connected.

 

[hwy17]

The connector on an MC4 can slide back in the housing and fail to achieve insertion. It's not obvious unless you hold it side by side with another connector.

Really big flaw with the MC4 I couldn't believe it when I realized it. Maybe I'm crimping them wrong otherwise it's just the force of the gland grip to keep the connector far enough forward.

 

[diWhy]

The connector on an MC4 can slide back in the housing and fail to achieve insertion. It's not obvious unless you hold it side by side with another connector.

Really big flaw with the MC4 I couldn't believe it when I realized it. Maybe I'm crimping them wrong otherwise it's just the force of the gland grip to keep the connector far enough forward.

I agree but in this case, that's not the problem. The panel is putting out voltage even with the extensions I made, it's just not able to be connected in series.

 

[hwy17]

I agree but in this case, that's not the problem. The panel is putting out voltage even with the extensions I made, it's just not able to be connected in series.

But connecting in series involves other connectors not involved in the single panel + extension test right? So those connectors would need to be checked, even if they are factory.

 

[diWhy]

I made all of the extensions and tested every one of them. In fact, the rest of the string works, it's just this one panel I cannot connect in series. I even changed one of the working extensions to make sure. Believe me, I tested the connections many many times .

The real question is, how can I fix these diodes because I'm going to have a hard time trying to find replacement panels. Been looking all day already.

 

[robbob2112]

I can get at the panel from underneath and can test but the six diodes all acted the way way. I only have a test meter that can tell me if there is a short/current flow however so not sure how I could test the diodes while they are connected.

If you ohm out the diodes they will read infinity with the leads one direction and nearly but not a short the other. Use the highest setting for ohms on the meter.

 

[diWhy]

If you ohm out the diodes they will read infinity with the leads one direction and nearly but not a short the other. Use the highest setting for ohms on the meter.

Thank you.
I didn't know this could be done in circuit. I might need a new meter to get this done but I'll try again with the one I have.

 

[SeaGal]

Bit late to this party. Can you clarify that this is still all open circuit testing - as you put in post #3?

If so, are you saying that a voltmeter connected to the dodgy panel reads a voltage on its own, but as soon as you connect it to other panels, despite those other panels not being connected to anything else, it shows no voltage?

Also, regarding the connecting in series.... does this strange behaviour also happen if the panel-in-question is at one end of the open string, in the middle AND at the other end of the string?

Are the panels just raw panels, or do you have any optimisers or similar attached to them.

All sounds very puzzling to me.

 

[robbob2112]

Thank you.
I didn't know this could be done in circuit. I might need a new meter to get this done but I'll try again with the one I have.

by in circuit I mean with the diodes still attached - but the panel disconnected from the others - both leads.

And another thought - Have you checked to make sure that neither of these leads is electrically connected to the frame? If you have the frames connected together via ground wire and/or clamped together and there are shorts from the panels to the frames in various places it could cause an issue.

 

[diWhy]

>Can you clarify that this is still all open circuit testing - as you put in post #3?

No load, nothing is connected to the inverter yet so only the panels and the strings are being tested.

>are you saying that a voltmeter connected to the dodgy panel reads a voltage on its own, but as soon as you connect it to
>other panels, despite those other panels not being connected to anything else, it shows no voltage?

Correct. If I test the panel on its own, it shows good voltage. As soon as I connect it to another panel, in a string of two, there is no more voltage across the two panels.

>Also, regarding the connecting in series.... does this strange behaviour also happen if the panel-in-question is at one end of the open
>string, in the middle AND at the other end of the string?

Correct. this is how we confirmed this panel is not working right. I tried connecting another panel on its negative and get nothing. I tried connecting to on its positive, I get nothing. As soon as another panel is connected, there is no more voltage flow.

>Are the panels just raw panels, or do you have any optimisers or similar attached to them.

Nothing, just the panel/s.

 

[littleharbor2]

>Can you clarify that this is still all open circuit testing - as you put in post #3?

No load, nothing is connected to the inverter yet so only the panels and the strings are being tested.

>are you saying that a voltmeter connected to the dodgy panel reads a voltage on its own, but as soon as you connect it to
>other panels, despite those other panels not being connected to anything else, it shows no voltage?

Correct. If I test the panel on its own, it shows good voltage. As soon as I connect it to another panel, in a string of two, there is no more voltage across the two panels.

>Also, regarding the connecting in series.... does this strange behaviour also happen if the panel-in-question is at one end of the open
>string, in the middle AND at the other end of the string?

Correct. this is how we confirmed this panel is not working right. I tried connecting another panel on its negative and get nothing. I tried connecting to on its positive, I get nothing. As soon as another panel is connected, there is no more voltage flow.

>Are the panels just raw panels, or do you have any optimisers or similar attached to them.

Nothing, just the panel/s.

Crazy. Sure sounds like a polarity issue

 

[diWhy]

I checked polarity too and that is correct as well. Hate mystery problems. I was sure it was one of the diodes gone bad and wondered how to fix that.