I've got a 12 v Peltier cooler salvaged from a broken cool box and I want to build a one can beer cooler powered from my solar on my boat, I've looked at the ones off the shelf and the cooler is at the base, I'm thinking that as cold moves downwards with gravity that it should be at the top , what do you think ?

I'm going to use extremely thick PU foam insulation to reduce power consumption, at least 100 mm.

I also need ideas on a shut off circuit when it reaches chilled temperature of 6-9 deg C.

 

I've built a few TEC coolers, and I always put the TEC on the top, for just the reason you stated and also the heatsink is more efficient.

 

Cold air moves down because of gravity.

I suggest that conduction is an advantage here if you can design this so the Peltier touches the can. As for a thermal switch with a range of +/- 1.5C... You are asking for something more sophisticated than a Klixon brand temperature sensor. How hard are you willing to work, or how much are you willing to spend, to get a temperature that accurate? I think the Jeopardy answer to that is, "Why were beverage coolers full of ice invented?"

 

What's the voltage and current rating of your boat solar panel?

 

What's the voltage and current rating of your boat solar panel?

The 500 w solar feeds into a 12 v battery bank so there's more than enough AH capacity

 

Cold air moves down because of gravity.

I suggest that conduction is an advantage here if you can design this so the Peltier touches the can. As for a thermal switch with a range of +/- 1.5C... You are asking for something more sophisticated than a Klixon brand temperature sensor. How hard are you willing to work, or how much are you willing to spend, to get a temperature that accurate? I think the Jeopardy answer to that is, "Why were beverage coolers full of ice invented?"

Im thinking put the can in upside down and shape a convex aluminium block that fits into the concave base of the can, clip the lid down so its making full contact with the can .

 

I'm thinking that as cold moves downwards with gravity that it should be at the top , what do you think ?

True.
But I would think you want the elemen to be in contact with the can where there's liquid, so perhaps at the side of the can near the top.

I also need ideas on a shut off circuit when it reaches chilled temperature of 6-9 deg C.

That could be done with a solid-state temperature sensor (e.g. LM35), and a comparator (e.g. LM339/393) driving an N-MOSFET to control the Peltier cooler current.