By: Tillmann (IP: 146.140.7.*)
Written on: 09-08-2005 14:59
Hello,
I'm not really sure I understood what you mean.
With a TEC, of course you need some thermal interface material (e.g. pad) on BOTH sides for good performance. And a thinner pad will be better than a thicker pad. So there is nothing unusual or astonishing about your result, it is exactly what was to be expected.
bye,
Till |
By: nikolai (IP: 63.110.69.*)
Written on: 09-08-2005 18:37
Thank for your responce. I see my question required some editional enformation. For some technical reasons I have to use the pad with low stiffness (it should be as thick as posible), but keep the thermal resistance as low as posible (from this point of view the pad should be as thin as posible). I started from the combination - one 0.5mm pad on hot side of TEC and thermal grease (good finish of the surface to be cooled) on the cold one. The thermal resistance of the grease layer is negletably low in respect to the thermal resistnce of the 0.5 mm pad. Other words: the interface cosists of the 0.5mm pad - TEC - thin layer of the grease with almost zero thermal resistance. The thermal performance of that system was acceptable, but the stiffness - not. Another combination of the structure of the interface: 0.5mm pad - TEC-thin layer of the grease (leftover from previouse tests - I did not clean surface) - the 0.5mm pad. I was surprised that editional thermal resistor on opposite side of the TEC (editional 0.5mm pad with relatively high thermal resistance in respect with the resistance of the grease) doesn't effect the TEC performance!? (The thermal performance of the interface with two 0.5mm pads in series on hot side - TEC - thin layer of the grease on cold side, did not surprise me - it degradated as it should be).
That let me find a simple solution of my problem: I got the low stiffness of the TEC system - it is determined mostly by the stiffness of 1mm pad ( two 0.5mm pads in series), as well as the low total equalent thermal resistance of the two 0.5mm pads in series, which is equal to the thermal resistance of one 0.5mm pad.
I am curious is that effect (roughly speaking- the effect of simmetry of the thermal resistance of interfaces on both sides of TEC) ocasional or fundamental. |
By: nikolai (IP: 63.110.69.*)
Written on: 09-08-2005 18:47
Thank for your responce. I see my question required some additional enformation. For some technical reasons I have to use the pad with low stiffness (it should be as thick as posible), but keep the thermal resistance as low as posible (from this point of view the pad should be as thin as posible). I started from the combination - one 0.5mm pad on hot side of TEC and thermal grease (good finish of the surface to be cooled) on the cold one. The thermal resistance of the grease layer is negletably low in respect to the thermal resistnce of the 0.5 mm pad. Other words: the interface cosists of the 0.5mm pad - TEC - thin layer of the grease with almost zero thermal resistance. The thermal performance of that system was acceptable, but the stiffness - not. Another combination of the structure of the interface: 0.5mm pad - TEC-thin layer of the grease (leftover from previouse tests - I did not clean surface) - the 0.5mm pad. I was surprised that additional thermal resistor on opposite side of the TEC (additional 0.5mm pad with relatively high thermal resistance in respect with the resistance of the grease) doesn't effect the TEC performance!? (The thermal performance of the interface with two 0.5mm pads in series on hot side - TEC - thin layer of the grease on cold side, did not surprise me - it degradated as it should be).
That let me find a simple solution of my problem: I got the low stiffness of the TEC system - it is determined mostly by the stiffness of 1mm pad ( two 0.5mm pads in series), as well as the low total equalent thermal resistance of the two 0.5mm pads in series, which is equal to the thermal resistance of one 0.5mm pad.
I am curious is that effect (roughly speaking- the effect of simmetry of the thermal resistance of interfaces on both sides of TEC) ocasional or fundamental. |
By: nikolai (IP: 63.110.69.*)
Written on: 09-08-2005 18:48
Thank for your responce. I see my question required some additional enformation. For some technical reasons I have to use the pad with low stiffness (it should be as thick as posible), but keep the thermal resistance as low as posible (from this point of view the pad should be as thin as posible). I started from the combination - one 0.5mm pad on hot side of TEC and thermal grease (good finish of the surface to be cooled) on the cold one. The thermal resistance of the grease layer is negletably low in respect to the thermal resistnce of the 0.5 mm pad. Other words: the interface cosists of the 0.5mm pad - TEC - thin layer of the grease with almost zero thermal resistance. The thermal performance of that system was acceptable, but the stiffness - not. Another combination of the structure of the interface: 0.5mm pad - TEC-thin layer of the grease (leftover from previouse tests - I did not clean surface) - the 0.5mm pad. I was surprised that additional thermal resistor on opposite side of the TEC (additional 0.5mm pad with relatively high thermal resistance in respect with the resistance of the grease) doesn't effect the TEC performance!? (The thermal performance of the interface with two 0.5mm pads in series on hot side - TEC - thin layer of the grease on cold side, did not surprise me - it degradated as it should be).
That let me find a simple solution of my problem: I got the low stiffness of the TEC system - it is determined mostly by the stiffness of 1mm pad ( two 0.5mm pads in series), as well as the low total equalent thermal resistance of the two 0.5mm pads in series, which is equal to the thermal resistance of one 0.5mm pad.
I am curious is that effect (roughly speaking- the effect of simmetry of the thermal resistance of interfaces on both sides of TEC) ocasional or fundamental. |
|