Numerous treatises and design guides for heatsinks exist and many sellers provide basic C/W ratings for their product range. The document cited below provides what seems to be better than most combination of practical information on a limited but useful cross section of medium to high power dissipation heatsinks from one manufacturer. While their intention is that you buy their products (and the pricing seems reasonable in at least some cases) the information is generic enough to act as a useful heatsink performance estimator in many applications. (If anyone else has other related reference material equal to or better than this do please share it.) Wakefield heatsink catalog excerpt - pages 45 - 53 * http://bit.ly/heatsinks01 *** [direct link at end of email for those who must]. Provides some excellent information on real world heatsink profiles and performance for a range of power dissipations and applications with forced and natural air flow. Dimensions, cross section diagrams and performance graphs of thermal resistance and watts at set temperature deltas for a range of power inputs Somewhat annoyingly lower rated units tend to be rated in watts per given delta temperature for natural convection but C/W for forced air. These are easily enough converted but make on the fly comparisons just a little harder. 9 pages of heatsink catalogue for commercially available products with ratings in the range ~=3D 0.2! - 12 C/W with natural convection (*1) and 0.045 to 2.5 C/W with forced air cooling *1 - From 527-24B 60C at 5W, to the frabjous & ginormous 510-14M (page 53) 0.21 C/W with natural convection _______________________ *Forced versus natural cooling rates - design effects:* Of interest is that the ratio of natural to unblown cooling rates may not be the same even for members in the same "family" at the same air flow rates. eg the above mentioned 510-14m achieves 0.21 / 0.059 C/W for natural / 100 cfm airflow while the 512-12m achieves 0.45 / 0.045 C/w for natural /100 cfm airflow Dimensions are 14m - 187 355 79 12m - 183 305 60 The dimensions are misleading - the smaller 12M heatsink has substantially fewer (about 70%) fins clustered together across only about 60% of the width while the larger 14m has more fins which are distributed across the whole width. The 12M optimises natural flow cooling - lots of area and wide airways. The 12m provides much higher air velocities at the same cfm. The result is that the ratio of cooling is 2+:1 in favour of the larger heatsink for natural flow but a useful 1.3:1 advantage for the smaller heatsink when fan blown. Larger forced:natural =3D 3.6:1 Smaller forced:natural =3D 10:1 (!) ______________________________________ Should anyone care: I came across this page while looking at solutions for cooling a load in the 10's of Watts range for which I'd much prefer not to need to use a fan and I wanted to minimise cost and size (or course). Something that can run for 10+ years (20 even better) sans maintenance and cleaning is desired. If cleaning ever was required then safety, accessibility and minimal cleaning effort and skill with no special tools is desirable. This document seemed to offer the average user a good cross section of examples with a better than usual insight into performance aspects. If anyone else has other related reference material equal to or better than this do please share it. ** =3D http://www.wakefield-vette.com/resource-center/downloads/brochures/extruded= -heat-sink-wakefield.pdf --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .