Ya... 'Bra Boy is an old buddy of mine, I ultimately sold him the Santel MK91 when I bought my 10". He started doing experiments on dew prevention long ago and likewise I have finally switched to fans in the dewcap and don't use heaters anymore. The improvement is quite significant on a big scope, but probably not noticeable on scopes under 150mm.
Battery voltmeter typically shows 13.4V, which agrees with my multimeter. The battery is a DryPower Deep-Cycle 50Amp-hour LiFePo so enough to run the rig from 5 nights or more without a noticeable drop in output. ASIAir reports 12.8. The ASIAir draws 0.3A. The cables I made are 2m long, 18AWG copper (heavy duty speaker wire) and can carry more than 15A. The battery box has a circuit breaker (10A), voltmeter and 100A switch, no other electronics.
Part of the issue is how much power is being drawn by the camera cooler - the night I had lots of crashes was hot (30C) and the cooler was at 90% or more and struggling to hold 0C. The two previously nights when ASIAir worked perfectly were not hot and it was running at something like 20%. In addition the ASIAIr was powering the EAF from a USB port.
So it doesn't surprise me when the load on the ASIAir is safely below the spec limit (3A) it works OK, but exceeding that limit on a 12V output implies it may crash.