58 minutes ago, birdibird said:

What do you guys say about this story and the cable thickness?

It's rubbish. 6 mm² is good for about 40 A. 35 m is longer than average perhaps but nothing excessive.

58 minutes ago, birdibird said:

Can that cause the inverter break?

No. The internal resistance if panels is way higher than that of 6 mm² cable, even 100 m of it.

20 hours ago, Coulomb said:

It's rubbish. 6 mm² is good for about 40 A. 35 m is longer than average perhaps but nothing excessive.

No. The internal resistance if panels is way higher than that of 6 mm² cable, even 100 m of it.

Yeah, that's my take on it too. He is just talking bull. I am not going to use that seller anymore.

Still is it a mystery why this happened. Can you electorate a bit more on your idea about how the panels could have been in contact with earth?

2 hours ago, birdibird said:

Can you electorate a bit more on your idea about how the panels could have been in contact with earth?

Well, I saw bare PV conductors in the damaged Y splitter. Those, or broken MC4 connectors, could have contacted the panel frames or metal roof, which should be earthed.

For almost 50% of the inverter's output 50 Hz sine wave, neutral (which is at earth potential) is connected to the positive end of the DC bus. This is so that the negative (active more negative than neutral) half of the 50 Hz sine wave can be generated. That means that during this time, the negative end of the bus is hundreds of volts negative with respect to earth. On high PV voltage models, bus- is connected to PV-. If any of the PV- conductors contacted earthed metalwork, a huge current would flow from the bus capacitors through this fault circuit. This current typically flows through current sensors, as it's often more convenient to measure the neutral current rather than the live current, since it's near earth potential. So this may be how fault code 51 (output over-current) is registered.

On 2023/04/15 at 3:14 PM, Coulomb said:

Well, I saw bare PV conductors in the damaged Y splitter. Those, or broken MC4 connectors, could have contacted the panel frames or metal roof, which should be earthed.

For almost 50% of the inverter's output 50 Hz sine wave, neutral (which is at earth potential) is connected to the positive end of the DC bus. This is so that the negative (active more negative than neutral) half of the 50 Hz sine wave can be generated. That means that during this time, the negative end of the bus is hundreds of volts negative with respect to earth. On high PV voltage models, bus- is connected to PV-. If any of the PV- conductors contacted earthed metalwork, a huge current would flow from the bus capacitors through this fault circuit. This current typically flows through current sensors, as it's often more convenient to measure the neutral current rather than the live current, since it's near earth potential. So this may be how fault code 51 (output over-current) is registered.

I understand.

The only thing is that there is only a bare concrete roof there. They chose not to use frames or SolarBlocs (solarbloc.es, which I often recommend on flat roofs), the panels are lying flat on the roof with the cables underneath. When the red Sahara storm with rain came, I guess some of the connectors were lying in a tiny pools of irony water, as the roof is not perfectly flat. Now it has been dry since December here in Spain (!), and only recently F51 and sometimes F59 and now F09 showed up. I can't phantom how some earth connection could have happened, you?

8 hours ago, birdibird said:

the panels are lying flat on the roof with the cables underneath.

Oh. That's not good. Panels absolutely need air under them for cooling. Overheating can cause cracking of the contacts between the copper strips and the cells. In extremely rare cases, this is actually repairable, though my repair exercise ultimately failed. A link in case anyone is curious about solar panel repair; I note that this is possible only on all-back-contact panels, which at the time was only Sunpower brand of solar panels.

8 hours ago, birdibird said:

I can't phantom how some earth connection could have happened, you?

With this extra information, no, I can't see it either. Though if the panels have been overheating, as seems likely, all sorts of unusual effects may occur.

But perhaps by "flat on the roof" you actually mean parallel to the roof, yet with at least a 50 mm gap between the back of the panel and the roof surface. 100 mm or more would be much better. If that's the case, then overheating is unlikely, although since the roof is flat, it's possible that hot air could accumulate under the panels. In the usual mounting situation, convection will occur, keeping the panels to "only" 50-75°C when in full sun. I imagine that summer in Spain could get quite hot at times.

On 2023/04/18 at 2:24 AM, Coulomb said:

Oh. That's not good. Panels absolutely need air under them for cooling. Overheating can cause cracking of the contacts between the copper strips and the cells. In extremely rare cases, this is actually repairable, though my repair exercise ultimately failed. A link in case anyone is curious about solar panel repair; I note that this is possible only on all-back-contact panels, which at the time was only Sunpower brand of solar panels.

With this extra information, no, I can't see it either. Though if the panels have been overheating, as seems likely, all sorts of unusual effects may occur.

But perhaps by "flat on the roof" you actually mean parallel to the roof, yet with at least a 50 mm gap between the back of the panel and the roof surface. 100 mm or more would be much better. If that's the case, then overheating is unlikely, although since the roof is flat, it's possible that hot air could accumulate under the panels. In the usual mounting situation, convection will occur, keeping the panels to "only" 50-75°C when in full sun. I imagine that summer in Spain could get quite hot at times.

Yes, summers in Spain can get quite hot indeed 😃 🌞

Multiple people here have their panels lying on the roof, and I never heard anyone else having problems with overheating, cables etc.
It keeps being a mystery why F51 happened....

On 2018/10/06 at 5:29 PM, Chris Knipe said:

Jaco,

Not wanting to argue with you, but I've got the service manuals here too.  Not even the service manuals, have any reference to the Error 51.  Taking it in to the supplier (out of warrantee too, might I add), they aren't going to do jack because the manuals don't tell them what to do.  The fault is also VERY intermittent, and we all know the inverter is just going to be tested, after which it's going to be "oh there's nothing wrong with it." We all know how this works, especially in South Africa... If I have to buy a new unit, then everything is just frankly going to be placed on hold until I have time and money to upgrade the entire system and go off-grid (Eskom is making that SO easy these days).

13 & 14 that you posted above, is an Error condition.  51 that I am talking about, is a Fault condition.  I hear what you are saying however.  Surges (and especially Overloads), comes from somewhere.  They don't magically invent themselves. Something is causing, and triggering this.

No, I'm sure no one would deliberately exceed Voc, and sorry if it came out that way.  However, there must be someone that has previously (whether deliberately, or by accident) exceeded Voc?  And what happened in those cases?

As you say yes, I can (and more than likely will) re-wire things myself and see whether the problem persists or not.  One of the reasons one comes to a forum (and the benefits thereof), is to find out whether someone ELSE experienced something similar or not, and whether there is meaningful inputs that can be delivered.  I guess, that is obviously not the case.  No worries then...   You and I aren't going to get anywhere TBH as you've pretty much up your mind it's the inverter, and I've pretty much made up my mind it isn't.  So unless there is someone else with some inputs, that's it then I suppose.  No point in going around in circles.