[mike31154]

Yes on the checkvalve. It's between the air pump and the carboy input tube which won't prevent any air leakage at the stopper. Both tubes are very tight in the stopper but as I mentioned earlier, I suspect there's some air escaping around the glue plug in the middle of the stopper. So between cycles I can see the water level in the line to the tank drop but this just means it takes 30 seconds after the air pump starts to repressurize the carboy and get the water back over the hump into the tank. I just noticed that once this started happening, the flow from the tube into the tank was more of a drip than a flow for the 5 minute on cycle. Once I raised the carboy it improved the flow which should make sense since there's less static head to overcome. Once I get a new stopper and refill the carboy, I'll see what happens and post my findings.

Thanks again for the idea. One of the things I like about it is if for some reason the carboy runs dry the air pump will just do what it does, pump air. If you let an ATO with a water pump or power head run dry, you run the risk of burning out the pump, tripping a breaker or even starting a fire from an overheated pump. Still, I'm thinking when I set up my new 90 with a basement sump, I'll go with a mechanical gravity feed with the humidifier float switch to control the flow. Even simpler and uses absolutely no power. Just have to mount the feed container above the sump.