Ditto on all the issues with getting high deltaT and high power output out of a peltier, but what you can get is easily implemented and reliable control of environmental temp against a dynamic endo/exo themic reaction. That is not easily done using ice blocks and a bilge pump.
Here is what I have played with for low temperature control...
CO2 canister(donated by one of the faithful) which bleeds into a copper spiraled tube(ice maker hookup from home depot) which is inserted through two T connections (also from h-depot) the connections are dark grey plastic and compress down on the tube easily without leaks. The opposite compression nut is removed and replace with a threaded pvc fitting that allows the connection of an external sheath of some type. I often use a midium diameter tygon tube over a 1/2" threaded schedule 40 PVC 1" flange and a ring clamp to hold it tight.
#############################
/---- %%%%%%% %%%%%%% ----\
/ ****===============**** ****==============**** \
----------------------------------------------------------------------------
----------------------------------------------------------------------------
\ ****=====| |======**** ****=====| |=====**** /
\---- | | %%%%%%% %%%%%%% | | ----/
#############################
##### outer sheath - ring clamp not shown
%%%%% flang or other adapter
***** threaded ends of T compression joints
/----
\---- compression nut
-----
----- inner copper or glass tube
====
=||= T connector
What is being constructed (cheap and OTC) is a coaxial heat exchanger, that will cool a liquid flowing one direction on the outer tube using expanding CO2 gas from the canister on the inner tube flowing in the opposite direction. Expanding co2 gas is way fucking cold as shit mucho bigtime! I don't remember what terminal temp for CO2 is, but it hasn't frozen acetone at any flow rate that I have tested.
I use an 12v oil pump for boat maint. to pump the acetone, but I don't know how long it will last given the noise it is making after about 12 hrs of use. I am using a 12v solanoid valve from the high performance autoparts store intended for swithing nitrous on/off (120$) to blow bursts of co2 into the exchanger to implement gross temp control. Lets say to the nearest 10-15deg C, but you won't get much better than that from such a crude control method. Secondly, I have four speeds that I can drive the circulating pump, but only two are currently used: minimal maintnance flow and full cooling.
Where the peltier comes in is at the junction between the reaction unit and the cooling setup. By placing a peltier between the device to be cooled and the cooling system I can adjust with a very fine degree of control the amount of cooling delivered. I can boost the cooling by running the peltier in the forward mode, or reduce the cooling by running the peltier in the reverse or heating mode. If I am running the peltier in the reverse mode, I will also slow the coolant flow to minimal maintnance flow. Peltiers can adjust the temp and direction very quickly, too quickly in fact. If you don't implement a degree of deliberate histeresis to govern the temp swings, you can destroy the device through thermal stress. Buy some extras and attach them using thermal lube and screw clamps instead of thermal glue witch requires a chisle for part replacement.
You will also need two temp sensors, one on the peltier and one in the reaction chamber. By calculating the rate change and change of rate change factors you can quickly identify the dynamic "intentions" of the system. The peltier sensor change values will freewheel as the coolant modulates and the reaction chamger sensor change values will freewheel as the reaction sinks or sources energy.
So just input it all into your computer and fuck around for a few months and it should all work out!
I have no life!
And on the eight day, God created Meth...
... and hasn't done much of anything usefull since!
