Difference between revisions of "Minutes 11-19-2007"
| Line 6: | Line 6: | ||
----------------------------------------------------------------------- | ----------------------------------------------------------------------- | ||
| + | ==Fernando's Thermal Measurements== | ||
| − | + | Fernando reviewed thermal measurements that he had made of his | |
| − | + | fully loaded daughter board on the bench. Both sides of the | |
| − | + | board were in contact with an insulating layer. Due to strong | |
| − | + | variations in thermal emissivity of the system, he relied on | |
| − | + | measurements using thermistors. His numbers were as follows: | |
| − | + | <ul> | |
| − | + | <li>Cooling strap at daughter board end = 36.4<sup>o</sup>C | |
| − | + | <li>ASIC top = 35.8<sup>o</sup>C | |
| − | + | <li>+2.5~V regulator = 46.6<sup>o</sup>C | |
| − | + | <li>+1.25~V regulator = 39.9<sup>o</sup>C | |
| − | + | <li>Copper edging of daughter board = 35.9<sup>o</sup>C | |
| − | + | <li>PCB = 35.5<sup>o</sup>C | |
| − | + | </ul> | |
| − | + | He also measured the temperature along the copper strip that extends | |
| + | away from the daughter board. About 6~in away, the temperature was | ||
| + | measured to be 24.3<sup>o</sup>C. | ||
| − | + | ==Cooling System Design Requirements== | |
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | The daughter board card temperatures will ultimately need to | |
| − | + | be kept at a temperature of about 32<sup>o</sup>C (or about | |
| − | + | 90<sup>o</sup>F). This is the temperature for which the cooling | |
| − | + | system needs to be designed. The basically amounts to a cooling | |
| − | + | power of about 2~W per daughter board (or about 65~mW per ASIC | |
| − | + | channel x 3 chips per board x 8 channels per ASIC). The good news | |
| − | + | from Fernando's thermal measurements is that the PCB temperature is | |
| − | + | essentially the same as the temperature of the copper strip on the | |
| + | edge of the daughter board. This is due to the good thermal | ||
| + | emissivity of the card/ASIC chip heat sink. | ||
| − | + | ==Bill's Cooling System Design== | |
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | + | ||
| − | 5). Bill stated that when a board is available for testing, he would | + | Bill reviewed his current cooling system design. Each FDC package will |
| − | + | be fed by a single input cooling line that will wrap around each package | |
| + | in a single loop and then exit the solenoid. A thin copper or aluminum | ||
| + | tube will spiral around the package ~24 times (once for each active | ||
| + | layer). The planned size for the tubing will be 4~mm O.D. and 2.5~mm | ||
| + | I.D.. The daughter boards cards will each be connected to an L-bracket. | ||
| + | This bracket will be in attached to the cooling line via a two-piece | ||
| + | block (a small aluminum clamp) that connects around the cooling line. | ||
| + | Bill stated that when a board is available for testing, he would | ||
| + | like to have one for his own measurements. | ||
| − | + | ==Issues in the Design== | |
| − | + | ||
| − | + | ||
| + | <ul> | ||
| + | <li>What is the best coolant to employ (nominally we are thinking | ||
| + | about using fluorinert)? | ||
| + | <li>How can the design be made to minimize the material inside | ||
| + | the solenoid? | ||
| + | <li>How best to connect the aluminum cooling blocks to the | ||
| + | cooling tube? | ||
| + | <li>Determine the final heat load that the system needs to | ||
| + | handle? | ||
| + | <li>Can this system be expanded to handle the cooling requirements | ||
| + | of the CDC and BCAL? | ||
| + | </ul> | ||
| + | |||
| + | ==Presentation to the Hall D Group== | ||
| + | |||
| + | Bill will present a preliminary design to the Hall D group in December | ||
| + | so that folks can give feedback on the design and understand the | ||
| + | materials added in the inactive region of the FDC. | ||
Minutes prepared by Daniel. Send any comments or corrections along. | Minutes prepared by Daniel. Send any comments or corrections along. | ||
Revision as of 16:38, 19 November 2007
Special FDC Cooling System Meeting
Date: November 19, 2007
Participants: Daniel, Fernando, Bill
Contents
Fernando's Thermal Measurements
Fernando reviewed thermal measurements that he had made of his fully loaded daughter board on the bench. Both sides of the board were in contact with an insulating layer. Due to strong variations in thermal emissivity of the system, he relied on measurements using thermistors. His numbers were as follows:
- Cooling strap at daughter board end = 36.4oC
- ASIC top = 35.8oC
- +2.5~V regulator = 46.6oC
- +1.25~V regulator = 39.9oC
- Copper edging of daughter board = 35.9oC
- PCB = 35.5oC
He also measured the temperature along the copper strip that extends away from the daughter board. About 6~in away, the temperature was measured to be 24.3oC.
Cooling System Design Requirements
The daughter board card temperatures will ultimately need to be kept at a temperature of about 32oC (or about 90oF). This is the temperature for which the cooling system needs to be designed. The basically amounts to a cooling power of about 2~W per daughter board (or about 65~mW per ASIC channel x 3 chips per board x 8 channels per ASIC). The good news from Fernando's thermal measurements is that the PCB temperature is essentially the same as the temperature of the copper strip on the edge of the daughter board. This is due to the good thermal emissivity of the card/ASIC chip heat sink.
Bill's Cooling System Design
Bill reviewed his current cooling system design. Each FDC package will be fed by a single input cooling line that will wrap around each package in a single loop and then exit the solenoid. A thin copper or aluminum tube will spiral around the package ~24 times (once for each active layer). The planned size for the tubing will be 4~mm O.D. and 2.5~mm I.D.. The daughter boards cards will each be connected to an L-bracket. This bracket will be in attached to the cooling line via a two-piece block (a small aluminum clamp) that connects around the cooling line. Bill stated that when a board is available for testing, he would like to have one for his own measurements.
Issues in the Design
- What is the best coolant to employ (nominally we are thinking about using fluorinert)?
- How can the design be made to minimize the material inside the solenoid?
- How best to connect the aluminum cooling blocks to the cooling tube?
- Determine the final heat load that the system needs to handle?
- Can this system be expanded to handle the cooling requirements of the CDC and BCAL?
Presentation to the Hall D Group
Bill will present a preliminary design to the Hall D group in December so that folks can give feedback on the design and understand the materials added in the inactive region of the FDC.
Minutes prepared by Daniel. Send any comments or corrections along.
