PV Maintenance Protocols: Preventing Common Issues

[Jaryd]

I would like to explore the essential maintenance protocols, common failure points, and data-driven maintenance schedules based on our extensive experience with installations across Scotland.

System Performance Monitoring​

Implementing remote monitoring systems (such as SolarEdge, Enphase, and GoodWe) is crucial for tracking key performance indicators:

• String-level current measurements
• Module-level voltage variations
• Inverter efficiency rates
• Performance ratio (PR) tracking
• Temperature coefficient impacts

Technical Inspection Protocol​

Our annual inspection checklist includes:

• IR scanning for hot spots (optimal temperature delta < 10°C)
• IV curve testingparameters:
  • Voc measurements
  • Isc verification
  • Maximum power point tracking
• Torque testing of mechanical connections (typically 10-15 Nm for most mounting systems)
• Insulation resistance testing (minimum 1MΩ at 1000V DC)

Inverter Maintenance Specifications​

• DC/AC ratio optimization
• Internal temperature monitoring (optimal range 20-40°C)
• Capacitor degradation assessment
• Fan functionality verification
• Data logging system checks
• Firmware update protocols

Environmental Impact Analysis​

Scotland-specific considerations based on our extensive experience:

• Salt spray corrosion in coastal installations
• Wind load stress on mounting systems (up to 100mph in exposed areas)
• Micro-crack development from thermal cycling
• Biological soiling rates in different regions

Performance Degradation Analysis​

Based on data from over 500 installations and detailed performance tracking:

• Average annual degradation rate: 0.5-0.7%
• Micro-crack impact: 0.3-1.2% additional annual loss
• Soiling losses: 2-5% without cleaning
• Shading losses: up to 20% in worst cases

Mounting System Integrity​

Critical check points include:

• Roof attachment torque verification
• Rail displacement measurement
• Clamp pressure assessment
• Waterproofing integrity verification
• Grounding continuity testing

Documentation and Compliance​

• IEC 62446 testing requirements
• BS 7671 electrical safety standards
• MCS certification maintenance
• G99 compliance verification
• Data logging for warranty claims

Technical Troubleshooting Matrix​

Common issues and resolution protocols:

String failures​

• Voltage isolation testing
• Bypass diode verification
• Connection box thermal scanning

Inverter faults​

• Error code interpretation
• DC input validation
• Grid compliance testing

Mounting system issues​

• Structural integrity assessment
• Water ingress testing
• Thermal expansion compensation

I'd like to here from other installers about their experiences with these protocols, particularly in harsh coastal environments or with specific inverter brands. What additional tests have you found valuable during your maintenance?


As Jayrd from JME Green Energy, I've shared our findings from Scottish installations to spark a discussion about best practices in solar maintenance. I'm looking forward to hearing about other installers' experiences and approaches.

 

[svetz]

Welcome to the forums Jaryd!

Hope to see some photos of your installations in Scottland! I'm in the U.S. (Florida Keys) and a DIYer... so not quite what you're looking for....but...

Monitoring
In terms of monitoring, it is critical. I specifically chose Enphase for panel level diagnostics and shade optimization. But also for automation of other systems by using the data available from the API. Enphase has a local API which they don't officially support and they've made fairly hard to use, but I find I can get the data I want (mainly to get alerts if any panels stop producing or are producing less than expected based on available sunshine so I know to clean them). I do an inspection every time I'm on the roof, mainly to see if any hurricane force winds snapped any bolts, but I don't do any electrical checking as you do as I have remote monitoring. I also don't baby sit the system, that's what the computer does, if it sees a problem it emails me.

Bolts
It's interesting that you do torque tests of the mechanical connectors, I've never found any loose, but I'm not subject to freezing/thawing. Is that a big problem in your location?

Failures
I have RMAed a few microinverters, but with the monitoring information it was easy to see that they were resetting themselves abnormally during the day. Took a while to convince tech support, definitely recommend having all the data handy.

Salt
As my panels are close to salt water I too was originally worried about salt spray. As the panels met the salt spec (IEC 61701 certification) I was more concerned about the balance of systems with corrosion from dissimilar metals, but so far no issues.

Micro-cracks
How do you test for micro-cracks? Is that the thermal scanning you mentioned? If so, do you have any data you can share with us? There's a thread on that here (https://diysolarforum.com/threads/using-flir-to-find-microfractures-and-other-solar-problems.30/), but it lacks practical advice from someone knowledgable.

Degradation
You said you see an annual degradation rate of 0.5 to 0.7%. Do you find it steady, or do you see the first year as an outlier with the biggest single loss? I've found this hard/impossible to measure due to the local weather (e.g., the third year I got more power than prior years as we had less rain).

 

[Jaryd]

Thanks for the warm welcome.

Regarding bolt torque testing, yes, that is part of our maintenance routine, especially for systems over 5 years in service because of Scotland's freeze-thaw cycles and the strong coastal winds. The combination of thermal cycling and constant wind load will generally cause deterioration in mounting systems over time.

We primarily use IR scanning for micro-crack detection at peak production hours to look for temperature deviations that could indicate a problem. I can't give you numbers pertaining to specific clients, but we have found thermal imaging to be an effective preventive maintenance method when its use is combined with on-site routine inspection.

Interesting point about the salt spray, we do exactly as you do regarding coastal installations: IEC 61701 certification. Our coastal experience in Scotland shows that different metals have to be very carefully isolated to avoid galvanic corrosion for long reliability.

Commonly, we would see slightly higher degradation rates in the first year; after that, they tend to stabilise to more consistent values for subsequent years. Weather variables affecting the readings are significant. I totally agree with you on the other hand, Scotland's volatile weather pattern makes it more difficult to make comparisons of one year to another without normalising conditions due to weather.

Your approach with Enphase monitoring seems very effective. We also highly value panel level monitoring for speed in diagnosing issues remotely. Would be interested to hear more about how your system handles alerts during these events.

 

[svetz]

Would be interested to hear more about how your system handles alerts during these events.

It's nothing special. I use the microinverter API to get panel data, check for if values are outside of a tolerance and send myself an email if anything looks wrong.

If they need cleaning or if a microinverter is underperforming is a little more difficult.

For cleaning I have a EcoWitt Wittboy weather station that monitors solar input. From it's API I get the watts/m² at that instant (which turns out is actually a 15 minute average).

In the first attempt knowing the area and efficiency of the panel and the current watts I compared each panel against the expected the output capped at the microinverter capabilities. I figured anything under 5% needed cleaning when I got around to it and anything under 50% needed attention (e.g., bird leaving the remains of a fish). The first pass was a dismal failure as the panels varied more than 5%.

The second attempt I reasoned that the panels got about equally dirty, so I could take the average of all of them. Still didn't work. But then I noticed that if I graphed it over time it averaged to the right value.

The third attempt was sort of a mix looking at the average over the day and the still looking at individuals panels. The last isn't particularly useful, I have to know what's going on too (e.g., in winter I get variable leaf shading on the southmost panels so they have reduced power from the other panels).

Still isn't perfect. Being lazy when the number builds up to where I tell myself I should clean them I start thinking it'll probably rain soon.

If a non-shaded microinverter is constantly underperfoming Enphase has no issue RMAing it, but quite to my surprise a number of them are over-achievers (so far anyway, we'll have to see how they are in another decade ; -). Some of that is probably the Seabreeze coupled with my cool-roof keeping the microinverters cool.