When you derived the attenuation values using spice for your Pi crossover, you calculated values specifically for either driver option. And so when you came up with values for the Econowave crossover, you must have looked at the Selenium driver specifically in Spice. Is that correct?
In other words, substituting another driver is not acceptable without running the numbers?
I thought some aspects of compensation in a compression driver were generic.
Actually,
most aspects of compression drivers are generic. In fact, most aspects of CD horns or waveguides are too. I'm not talking about differences between horns that employ diffraction verses those that are radiused smoothly. That's a different matter entirely. I'm talking about power response, and what sets the overall shape of the response curve in a compression driver mounted on a constant directivity horn.
The bottom line is this: If you are using a CD horn or waveguide, there is only one transfer function you need to worry about. They all share basically the same one. That's why there is one curve for CD equalization.
It is the conjugate of the power response of the driver.
When I first started doing this, I sort of assumed there would be more differences between drivers and horns. Different horns have different coverage angles, flare shapes and features. Different drivers have different diaphragm materials, motor structures, voice coil impedances and phase plugs. About the only thing that's the same is the exit aperature size.
But what I found was that the biggest difference is the voltage sensitivity. The driver's impedance plays a part and the horn's coverage angle plays a part. Once you know that, you're pretty much set.
There are many reasons for this. Each variable has a set of conditions that it describes. But they tend to act together. For example, if a horn provides constant directivity, then the SPL within the coverage angle tracks the power response of the driver. So the narower the angle, the higher the SPL within the pattern, but it is always the same shape as the response on a plane wave tube. So equalize that response and you have the correct response anywhere in the pattern, for any horn that provides reasonably uniform directivity.
There are a lot of things in the compression driver that can be variables. Change the amount of compression and bandwidth, efficiency and sensitivity changes. Different diaphragm materials and thicknesses mean potentially different mass rolloff and different breakup modes. Their phase plugs' ability to equalize path lengths and reduce front chamber volume can provide extended top end, but some designs rolloff sooner. Voice coil inductance sets another rolloff pole, basically marking the end of useable response.
But in the end, you'll notice all modern 1" exit compresion drivers act somewhat similarly, with mass rolloff starting around 4kHz. The biggest difference is behavior in breakup - some are well damped, others not. Most start around 8kHz, and some get real peaky above that. It can give some additional high end, but it can also make them sound shrill. Can't do much with that except choose compression drivers carefully. You don't want a compression driver with a lot of spikes in its response curve, because you can't do much with them. Beyond that, since mass rolloff happens around the same place in each driver, this means top-octave augmentation should start the same for each driver. About the only difference to design for is the impedance, which sets the voltage sensitivity.
That's why the R1/R2/C1 values I've used for years in my crossovers are the exact same values you should use in your crossovers, even though the horns and drivers are different. This is true of just about any driver on any CD horn.
