Anyone else had issues with MPPT controllers showing wildly different wattage to actual panel output?

Picked up a Victron SmartSolar 100/30 a few months back to pair with two 200W panels (wired in series, so roughly 40V Voc) on my static shed setup. On a decent sunny day the app is showing maybe 180–210W peak, which seemed reasonable enough. But lately I've been noticing the reported wattage jumping all over the place — we're talking swings of 60–80W within seconds on what looks like a fairly clear sky.

I initially assumed it was just cloud edge effect doing its thing, but I logged a full day last Saturday and the numbers genuinely don't add up. At one point the app showed 94W while my clamp meter on the battery cable was reading current that should correspond to closer to 140W at my battery voltage (around 26.8V, 24V system). That's a fair gap, not just rounding error.

Wondering whether it's a dodgy shunt reading, a firmware issue, or something else entirely. I'm running firmware 1.61 on the controller. I've also got a BMV-712 on the battery side and the two rarely agree with each other, which isn't exactly confidence inspiring.

Has anyone else seen this kind of discrepancy with Victron kit, or is there a setting I'm likely missing? Happy to share the VRM logs if that helps anyone diagnose it.

@AlanPearce the "roughly 40V Voc" concern is worth looking at more carefully. Voc is measured in cold conditions — on a freezing morning those panels could push 45-48V combined, but that's not your efficiency issue.

Real-world losses stack up fast:

• Panel degradation from manufacturer specs (they're optimistic)
• Temperature derating — panels lose ~0.3-0.4% per °C above 25°C
• Wiring losses
• STC vs actual irradiance — UK skies rarely hit the 1000W/m² test standard

My setup with two Renogy 200W panels through a SmartSolar 75/15 typically peaks around 280W on a brilliant summer day, not 400W. That's completely normal.

Check your Victron Connect app's history tab — if the yield curve looks smooth and the controller isn't throwing faults, it's probably just physics rather than a fault.

@AlanPearce worth checking your cable sizing and connection quality before assuming it's a panel or controller issue. Voltage drop across undersized DC cabling is a silent efficiency killer — even 0.5V drop each way on a low-resistance run can translate to meaningful wattage losses that the MPPT never recovers. What gauge are you running from panels to controller, and what's the total cable length? I've seen installs on my shepherd's hut build where 4mm² cable over an 8-metre run was dropping enough to explain a 15-20% apparent shortfall. Also, @HamishMitchell makes a valid point about Voc — but equally worth confirming your panels are genuinely clean and unshaded, even partial shading on a series string absolutely hammers output disproportionately. The Victron app's "yield today" graph is useful here; if you're seeing the curve plateau suspiciously early in the afternoon, shading is likely the culprit.

Really common issue and I went through the same head-scratching on my static caravan setup. A few things worth checking that nobody's mentioned yet:

• What's your battery state of charge when you're seeing those low figures? If your bank is near full, the MPPT will deliberately throttle output — it's working correctly, just not harvesting what the panels could produce
• Are the panels actually identical models? Even slight mismatches in series strings can drag performance down significantly
• Have you checked for any partial shading — even a small shadow across part of one panel hits series strings disproportionately hard

On my van conversion I found a single shadow from a roof vent was costing me nearly 30% on cloudy days. Took me ages to spot it. The Victron app history graphs are genuinely useful for comparing peak output across different days once you know what you're looking for.

Something I noticed on my cabin setup — the Victron app's "panel power" figure is calculated from voltage × current at that moment, but it's never going to match the panel's rated output unless conditions are perfect (25°C cell temp, 1000W/m² irradiance). Real-world cell temps in summer can push 50-60°C even in the UK, which knocks efficiency down noticeably.

Also worth checking if you've got any partial shading hitting one panel — with series wiring that'll drag the whole string down hard. Even a shadow from an aerial or chimney for an hour can skew your daily figures significantly.

What's your actual battery voltage sitting at mid-afternoon? If it's already near absorption voltage the controller will be throttling back deliberately.

Great points already made here. One thing I'd add — have you double-checked that your panels are actually hitting their rated Voc in series? Cheap or mislabelled panels (especially from certain marketplace sellers) can fall noticeably short of their spec sheet. Worth grabbing a multimeter on a clear morning and measuring directly at the array before it even reaches the controller.

Also, battery state of charge plays a bigger role than people realise. If your batteries are nearly full, the MPPT deliberately backs off to avoid overcharging, so the wattage shown will be well below what the panels could theoretically produce. It's not a fault — it's working as intended. @LesCrane's point about the calculated figure is spot on too. What's your battery bank capacity and chemistry, @AlanPearce?

Good points from everyone already. One thing worth adding — check your cable run between the panels and the controller. Even a modest voltage drop across undersized or lengthy DC cables can make a real dent in what actually reaches the MPPT. I'd grab a multimeter and measure voltage directly at the controller's PV input terminals while it's generating, then compare that to what you're reading at the panels themselves. A significant difference there tells you the cables are robbing you of power before it even gets a chance to be converted. On a 40V series string it's easier to spot than on lower voltage setups. What gauge cable did you use and how far is the run from panels to controller?