Let me try to make sense of this. If all the water is going through the grain bed, and said water is cooler than the set temp on the controller, this cool water will make its way through the grain bed and show up in the bottom of the kettle where the temp probe is. The temp probe would measure the cooler temperature and provide the feedback that would tell the controller to turn the element on. But the crux is that temp will never be higher than the set point in the controller :/
It is an interesting problem, what with the water being recirculated to the top of the grain bed can never be warmer than the set temp on the controller because of the location of the probe. The water on top is bound to cool below the set point whether one is overflowing or not.
Would it make a difference, and has anyone tried, setting their temp a few degrees warmer than desired mash temp then measured the bed temp. Seems that with this set up one would need a degree or so of "superheat" , if you will, to hit the desired bed temp. Feels very similar to a HERMS set up.
So, considering all this, maybe you're on to something. If you make sure the temp on top and bottom is as close to the same temp as possible, that would hopefully keep the temps in the middle as close to the set temp as possible, assuming the gain bed is flowing at a reasonable rate. I like your idea if turning off your pump every so often to make sure the grain bed is flowing. That seems to be the best of both worlds.
I also have to remind myself that this is a simple analog controller with a pretty wide band on hysteresis. You just won't be able to hold temps to within 1/10 of a degree, so maybe this whole conversation is more academic in nature than practical.
Thats exactly what I do............raise the controller temp a few temps so my grain bed temp probe is at my desired mash temp. If I want 152F, I might need 153F or 154F on the controller to hit desired temp.
I use pipe insulation on all my external pipes leading up the recirculating pipe to minimize heat loss when it exists the return pipe. Total of $1.00 in insulation material.
I can say the the better the grain bed flow the closer the controller temp can be to your desired temp in the grain bed.
Jamie