MPPT Charge controller problem?

[inetdog]

But surely that doesn't account for why it reaches 100% charge very quickly when a tiny bit of sun comes out.

I would guess that the 100% indication is based entirely on the voltage at the meter connection. That may or may not be close to the terminal voltage across the battery bank. And the battery voltage is only weakly indicative of the battery SOC when there is current flowing (either charging or load current.)
In your case the internal resistance of the battery or the resistance of the charging circuit is causing the meter reading to depend far too much on the current from the CC.
It may also be a sign that your battery has a low effective capacity, having lost life to sulfaction, plate corrosion or other failure modes.

[jony101]

that's normal behavior I see that with my mppt too. As soon as the cloud clears out of the sun, the mppt recalculates and on mine it will shootup the voltage and amps and then settle back down. It only happens as the cloud clears the sun, at all other times it works fine. On sunny days with no clouds it never happens.

On mine when the voltage shoots up it kills the load out and i have to reset it manually, and i notice i have to reset it everyday. I have used the same mppt with about 4 different batteries, and it behaves the same with all of them.

I suspect some of the more expensive mppt have circuitry that can handle this better.

On my mppt (ecoworthy 20 amp) it has a way to check how much better performance I get using mppt over pwm. When i press the button to check the performance, the controller recalculates the mppt, the voltage and amps shoots up just like it does when the cloud clears the sun. Thats why I suspect the controller is recalculating the mppt when sun output changes.