Presently I experiment with a small island net: Two used solar modules 24 V/70 W in series and two 12 V/26 Ah lead-gel batteries in patrallel. I want to operate my ham radio station from that.
Over the summer my BlueSolar MPPT 75/10 worked very well. But recently, since it got colder, I quite often get #33 errors. These errors come while the input voltage is hardly higher than 70 V. After the charger has turned off I get an open-circuit voltage in the 77-79 V range measured by the BlueSolar MPPT 75/10. My DVM shows a few 100 mV more.
Using a scope I found positive and negative voltage spikes with some 5 V peak amplitude at the PV inputs every 30 µs. It appears that the over-voltage protection is triggered by these spikes of the voltage converter.
As a first measure I have added a 1µF/100 V capacitor to the PV input. Take that literally: That RF-capable capacitor (no electrolytic) is directly at the connector.
On VHF/UHF I do not "hear" the solar charger. But I expect quite some EMC problems when I switch to shortwave. I will add some protection circuitry to avoid this.
But the most annoying problem still remains: The #33 error appears long before the input voltage reaches the specified 75 V.
