Timeline for Is it a good idea to use a single heatsink for two IC's?
Current License: CC BY-SA 3.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 17, 2017 at 16:22 | comment | added | Shakti Phartiyal | How does the breakdown voltage affect heating ? | |
| May 17, 2017 at 15:13 | comment | added | D.A.S. | you must understand BDV of thin insulator and BDV from PLT to ensure safety. breakdown voltage, power line transients...and Rth thermal resistance 'C/W | |
| May 17, 2017 at 15:07 | comment | added | D.A.S. | uand good electrical insulation and thermal,conduction ... a tradeoff and check surface air speed not CFM . I used plenum and spoiler on intake to suck eddy current air over the hotspots. yours is the fin surface. not easy compromises until you understand. | |
| May 17, 2017 at 14:55 | comment | added | Shakti Phartiyal | @TonyStewart.EEsince'75 are you saying i should go ahead with the single heat sink and put on a fan with failure detection ? | |
| May 17, 2017 at 14:16 | comment | added | D.A.S. | Factors to consider are BDV of insulation, Rth of thermal conductor layer and air flow VELOCITY using a plenum for eddy current air flow at the fin surface not LAMINAR air flow in CFM of the fan. This last part is most critical for optimal MAX temp rise of the junction of 40'C max. Poor designs will be higher. Excellent designs will be 20'C rise in Tjcn . Fan speed may be regulated by T feedback of a dummy diode or thermistor, and fan fail detect is a must. This was my personal revelation in a contracted design for AVAYA that I did on a 185W 1U closed top unit rack design. Cost was very low. | |
| May 17, 2017 at 14:00 | answer | added | Trevor_G | timeline score: 6 | |
| May 17, 2017 at 13:54 | history | tweeted | twitter.com/StackElectronix/status/864841653563084801 | ||
| S May 17, 2017 at 13:46 | history | suggested | dlatikay | CC BY-SA 3.0 |
caps, punctuation
|
| May 17, 2017 at 13:44 | review | Suggested edits | |||
| S May 17, 2017 at 13:46 | |||||
| May 17, 2017 at 12:13 | vote | accept | Shakti Phartiyal | ||
| May 17, 2017 at 12:03 | comment | added | Shakti Phartiyal |
@RohatKılıç heatsink with heatsink with black anodized coating is a good suggestion I can look into.
|
|
| May 17, 2017 at 11:48 | answer | added | Joren Vaes | timeline score: 25 | |
| May 17, 2017 at 11:45 | comment | added | Rohat Kılıç | @Shakti Calculate temperature rise of both components (using power dissipation and Rth(j-a)) under the same conditions. If the difference is too high then yes, there will be a transfer. If you use a heatsink with black anodized coating then the transfer will be less because most of the heat will be absorbed by heatsink itself. | |
| May 17, 2017 at 11:42 | comment | added | Hearth | Heat being transferred between ICs is often a good thing, actually. Keeping two things at the same temperature makes them have more similar characteristics than they would otherwise, which helps with load sharing. | |
| May 17, 2017 at 11:42 | answer | added | Olin Lathrop | timeline score: 5 | |
| May 17, 2017 at 11:42 | comment | added | Andrew | Head transfer will depend on the relative temperatures of the two parts and the thermal properties of the heatsink. If one part is significantly hotter then yes, heat will be transferred. | |
| May 17, 2017 at 11:39 | comment | added | Shakti Phartiyal | @RohatKılıç Will there not be an issue of heat being transferred to the other IC ? | |
| May 17, 2017 at 11:35 | comment | added | Rohat Kılıç | If their pads can be connected to the same point (i.e. GND) then you can connect them to the same heatsink without a need of isolation. Or else you will need an isolator. | |
| May 17, 2017 at 11:29 | history | asked | Shakti Phartiyal | CC BY-SA 3.0 |