Top hat inlet outlet, just making sure I have this right

[nikbrown]

I'm going to be ordering this dust collecter:
http://www.grizzly.com/products/3HP-Double-Canister-Dust-Collector-/G0562Z

It has a 7" inlet but I want to standardize on 6" ducting. So after reading through a bunch of posts I would think I should go with 7" ducting between the separator and the dust collector and the inlet on the top hat should be 6" ducting. Right?

[jgt1942]

if you can stay with 7" it would be better. Anytime you reduce the size of the pipe you reduce the amount of air that can flow through it. I've seen some sites that provide the formula (sorry I don't recall). Also you want to avoid sharp 90 degree turns and use smooth pipe as much as possible. PVC is a great choice over metal but most likely you will not find 7" PVC. Make your main run with 7" pipe and reduce down to 6". Also look at the Bill Pentz site (http://billpentz.com/woodworking/cyclone/index.cfm) for more great info. I'm in the process of building two Thein TopHats for my dust units.

[retired2]

Anytime you reduce the size of the pipe you reduce the amount of air that can flow through it.

That is the most likely outcome if you replace an entire system with a smaller pipe size.  However, it is not that simple and your statement is not universally true.  Depending on where you are on the fan's performance curve, and the amount of pipe being replaced, it is quite possible the FPM will simply increase so that total CFM remains the same.

To know for certain whether CFM would drop, you would need to add up the SP losses for the two configurations, and then look at the fan's performance curve to see how much, if any, CFM will drop.

But of course, few, if any, DC manufacturers provide a fan curve.  That would reveal too much information, and buyers would then know which DC is truly the best.  Most manufacturers just say their fan is rated at xxxx CFM, which tells you nothing, and if you are talking about cheap systems you can almost be certain that the number is exaggerated, and the fan curve falls off a cliff the minute any line losses are added.  With one of these systems I would have to say your statement is probably true.