Fan On Demand

Miller isn't about to Ok such a retrofit without doing a careful thermal design review. That's expensive, and not worth their time for an older machine. So you're on your own, and any liability due to your retrofit is on your shoulders.

The problem is that even if you thermally bond sensors to the heatsink and transformer, other parts of the circuit may get enough conducted, convected, or radiated heat to be damaged before the sensors call for the fan to run. In other words, even though you might protect the diodes and transformer, you might inadvertently cook the control board because it expects a certain continuous flow of cooling air.

Me, I'd let the fan run. But do what you like. It is your welder, and your nickel.

Gary

You could always wire in a switch that you could control manually when you feel it needs some cooling

Be very cautious in changing a design. Liability laws might be re-flecked on to you.

There is something called thermal resistance. Theta J(parameter). It is used in semiconductor cooling and other heat sink concepts.

The thermal resistance (just like ohms law) can be high enough that if the sinking (fan) isn't there, the core would overheat long before an outside sensor can see it.

Martin

After years of Miller Bobcats & Bluestars and shop stick machines my very first MIG setup was a Century (rue that day) but I put a sensor in the line for fan on demand and it worked great! I think I used a point of 125F for contact. But since anything Century makes is crap .... in our shop with 5 machines "if it ain't BLUE it won't do!

Check the duty cycle of the machine to see how long you may weld at a higher current. The machine should automatically cut off before you can do any damage to the rectifiers.

Harry