Mine's so accurate I've stopped checking the BBC — if the Cerbo's showing under 20W by 9am, I'm not bothering to put the chairs out.
Running 400W of Renogy panels on a south-facing roof and the pattern is uncanny. Sunny day = shoots past 18V input before I've had my first brew. Grim day = limps along at 8W like it's personally offended by the clouds.
Anyone else reading their solar data like tea leaves, or is it just me slowly going mad in a layby somewhere on the A38?
Ha, @BenJackson, I do exactly the same thing! My Victron's become the most reliable forecaster in the house — beats the Met Office hands down for what actually matters day-to-day.
I've got 600W on a slight south-west tilt and I've noticed the ramp rate in the morning is almost as telling as the absolute figures. On a clear day it climbs sharply and steadily; on a diffuse overcast day it sort of plateaus early and never really gets going. Once you've watched it long enough you develop an instinct for whether that plateau is "thin cloud that'll burn off" or "that's your lot, mate."
My wife thinks I'm obsessed but I've saved us from a soggy barbecue more than once, so she can't argue with the results! 😄
Genuinely hadn't thought about using it this way but now I can't un-see it. My SmartSolar 100/30 has been surprisingly consistent — overcast days it flatlines before 8am and sunny days it's already climbing past 50W.
What I'm curious about though: does anyone find the rate of morning ramp-up more useful than the absolute figure? I'm wondering if a steep early curve predicts a better afternoon than a plateau at, say, 30W by 10am.
Asking partly because I'm trying to time my EV charging sessions around peak generation and I need a more reliable trigger than just eyeballing the app. Has anyone scripted anything in Node-RED off the Cerbo data to automate decisions like this?
Really curious how consistent this is across different setups — does panel orientation make a big difference to the reliability as a forecast?
My van build has 200W on a roof that's obviously never perfectly aimed, so I imagine the readings are messier than @BenJackson's fixed south-facing array. Been thinking about adding a garden office with proper fixed panels and this thread's actually making me wonder whether a well-oriented setup would give cleaner "weather signal" data.
Has anyone tried logging the 9am figures over a few weeks and comparing against actual Met Office records? Wondering if there's a useful correlation threshold — like, under X watts almost always means overcast all day vs just a slow morning start.
@StuKnight — orientation makes a massive difference in my experience. I'm running 200W fixed on the motorhome roof, which is rarely perfectly south-facing once I've parked up. On overcast days the directional bias almost disappears — diffuse light hits everything fairly evenly — so the MPPT output becomes a cleaner signal for cloud cover rather than direct irradiance.
What I've noticed is the rate of change in the first hour after sunrise is actually more informative than the absolute wattage. A clear day climbs steeply and consistently; broken cloud gives you that erratic spiking pattern in the VRM history graphs. Took me a few months of cross-referencing with Windy.app to calibrate my gut feeling, but now it's second nature.
Fixed roof installations probably give the most reliable baseline for this kind of informal forecasting precisely because the geometry is constant.
@StuKnight my garden office array is tilted at 35° south-facing and the morning ramp-up curve is so consistent I could set my watch to it — anything flatter and you're basically reading tea leaves until noon.
Log in to join the discussion.
Log In to Reply
