Date: Thu, 23 Nov 2000 10:51:57 -0000 From: David Atkinson To: 'Vikram Dhillon' Subject: Peltiers - heat sinking Hi Vik, I hope you got back to Sheffield alright - and in less than 6 hours! You asked me to describe the Peltier controller system to you so I thought I'd drop you a few lines. When buying a Peltier the manufacturers state the device's DTMax, and QMax (the maximum heat load the Peltier can pump). DTMax is the maximum temperature difference that the Peltier can provide. This can only be achieved when there is no heat load. Conversely, QMax is the heat load that will reduce the device's temperature difference to zero. Therefore, for a given Peltier the greater the heat load (i.e. closer to QMax), the lower the temperature difference that it can produce. FOr Ultracam the heat load will be about 3Watts (in dry nitrogen) or about 1Watt if operated in a vacuum (due to no convective heat load). The Peltier we are using is a 3 stage Melcor device which has a spec of Qmax=15.6W and DTMax=97 degrees C (aside this is actually the manufacturer's spec for a hot side temperature of 27degrees C - while our hot side will be closer to 16degrees C so we won't get quite as large a DTmax). We will be using an Alpha - Omega Peltier controller to regulate the current supply to the Peltier. THis is a closed loop PID controller which can supply a large current supply (we need about 6Amps) and utilises a temperature sensor to feedback the Peltier cold side temperature. Well actually the temperature sensor will be mounted into a copper 'coldfinger' block which is the interface between the Peltier cold side and CCD. The Peltier's hot side will be mounted directly on top of a copper heat exchanger (base of head) which is cooled with a recirculation chiller - Neslab Merlin 33. The Merlin has to be able to sink the heat that the Peltier is pumping plus the heat the Peltier is generating itself (which far outweighs the heat it is pumping) in each of the three heads (in addition to various other heat loads - SDSU etc). The Merlin 33 recirculation chiller can provide 750W of heat sinking at 10 degrees C - which is ample. THis means that the coolant to the copper heat exchanger will be maintained at 10degrees C (provide the total heat load never goes above 750W). There is a temperature differential across the heat exchanger, which means that the Peltier's hot side will actually sit at about 16 degrees C rather than 10 degrees C. I have attached a report which I wrote in August, when I was looking into Peltiers. A few things have changed since the report was written (i.e. there will be a interface coldfinger between the Peltier's cold side and CCD, we may be operating in a vacuum, and we have gone for a beefier ricirculation chiller - i.e Merlin 33 not Merlin 25), but it should give you an idea of some of the issues I'm trying to describe here. Cheers Dave ************************************* David Atkinson Electronic Engineering Group UK Astronomy Technology Centre Royal Observatory Edinburgh Blackford Hill, Edinburgh EH9 3HJ ************************************* email: da@roe.ac.uk http://www.roe.ac.uk Tel: +44 (0)131 668 8100 +44 (0)131 668 8229 (Direct) Fax: +44 (0)131 668 1130 *************************************