I ordered the HF driver and have a question about the correct RMS power to use when modeling HP filters with boost.

Don't ya lose some power when ya add boost in a HP filter? If I add 3dB boost in a 2nd order Butterworth, do I need to reduce power in the model by 1/2?

Apologies for taking so long to get back...

To model a rumble filter with WinISD, you want to use a Highpass/SOS user specified. It's been a while since I've read up on filters, but by definition, I *think* a Butterworh has to be Q=.7071, which will result in no peaking...at least, that's how WinISD treats them. You can compare the two by selecting the "Transfer function magnitude (EQ/Filter)" graph, which will let you see just the response of the selected EQ curve. If you enter a HP/Butterworth, no matter what you do to Q, you won't get a peak or change in the slope.

You don't lose power with EQ. The amp is capable of what it's capable of. In theory, you could limit potential output when the source signal contains content within the EQ curve and outside simultaneously. If the driver can reach its excursion limits in the EQ'd range, the out of band excursion will be limited to less than Xmax, reducing max potential output at those frequencies. But, that's a highly theoretical condition.

In the most simple of explanations, if the source signal calls for 100 watts at 25hz and you have 3dB of boost, the amp will deliver 200 watts for that 25Hz note, assuming it's capable. However, you are correct that WinISD isn't sophisticated enough to normalize the output model based on EQ and user specified input power. If you set input power at 100 watts with 3dB of boost somewhere, WinISD will show you output and excursion models that assume the amp will deliver 200 watts when needed. You can do something of a visual normalization by overlaying otherwise identical models with and without the boost. I like to adjust the input power of the EQ'd model down to compensate for the boost...say 35 watts with 3dB of boost for a 100 watt amp. Obviously, this won't be 100% accurate since boost slopes down from a peak, resulting in the model boosting that 35watts by less the further from the peak, but it kind of shows you the worst case scenario for excursion limited output at your center frequency.

-Brent