At 04:57 PM 8/20/2009, Jinx wrote: > > a bunch of power resistors - 2 rows of them. > >Can you put alternate resistors in a row on opposite sides of >the board and/or stagger the rows so that alternate pads are >mid-component wrt the other row ? Or bend the resistor >leads so that the resistors are in rows (if that's what you want) >but their pads are not Thanks for the suggestion. I don't think that resistors on both sides of the PCB is going to work for me - they run hot and I need them above the PCB so that they are in the airflow stream to be cooled. But - it turns out that I goofed when I started my layouts. Each board need to have 9 channels, with 3 resistors each. I placed enough resistors on the PCB to do 12 channels (the original version of this project). Turns out that I now have LOTS of space - minimum clearance is now 0.200". And I was able to make the PCB even more narrow - the status LEDs are easily visible. Thanks to all for the suggestions. I particularly like the idea of milling slots between the pads - I think that would have been my next move. dwayne PS - this project was first done back in 1992. Its a monitoring system that monitors the health of individual rectifier diodes in a large DC power supply. This version monitors 18 diodes that are all connected in parallel. Each diode is individually fused with 2- 1000A fuses in parallel (2000 Amp carrying capability). There are 12 banks of these diode arrays in each power supply (2- three phase bridge rectifiers). Yeah - these are LARGE power supplies! We've done other versions with 26 diodes per array and 24 diodes per array. Regardless of how many diodes there are in each array, there are 12 arrays per rectifier. The problem is that the monitoring cards have been moved from inside the rectifier cubicles (lots of airflow) to another enclosure located some feet away from the rectifier. They didn't put ANY cooling in those cabinets. Each input on each card dissipates about 5W (spread out over 3- 5W resistors). This version has 18 inputs per card (18 x 5W), there are either 12 cards or 24 cards in this enclosure. That's a LOT of heat to get rid of. Surprisingly enough, the boards have survived without problems under those conditions for around 5 years. Now they are starting to fail. The boards have run so hot for so long that the organic material (epoxy) has disappeared from the PCB material in the immediate vicinity of the power resistors and all that remains is the fiberglass weave. The boards are failing because of mechanical issues - the resistors are coming loose from their connections. I've got a couple of options. If the customer wants to salvage these boards, I'm going to move the power resistors to this daughter board that is mounted above the main PCB. That will allow them to be cooled by both convection as well as forced-air cooling. The other option is to re-design the original PCB to move the large, bulky capacitors that are currently on the top side of the PCB to the bottom of the PCB and re-locate the power resistors to the space where the capacitors used to be. Regardless of which method the customer chooses, the enclosures that house these cards are getting air handling systems added to them. I just wish they had done that back when we warned them about the cards needing adequate cooling way back then. dwayne -- Dwayne Reid Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax www.trinity-electronics.com Custom Electronics Design and Manufacturing -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist