Yes, I'd definitely test the tweeter with just a 5-10uF series capacitor. That will illustrate what the driver acts like without attenuation or compensation. It's a good sanity check.
As for test points, what you should see is approximately 11Ω across the tweeter output lugs, with the compression driver disconnected. Not much else can be checked with an ohmmeter, because capacitors look open and coils look like shorts when presented with DC.
compaddict Messages: 84 Registered: April 2014 Location: Auburn, CA USA
I was pretty good at tuning carburetors and mechanical/vacuum advance distributers..
Then came EFI and I never looked back.
MiniDSP 4X10 and an Outlaw 7075 (7x75w) on the way.
Might keep the two mono 200w for the 2226h.. Don't know yet.
Sorry for not troubleshooting the XOs.
I'm with you on the carburetors and distributors. Back in the 1970s, I used to tell my hotrod buddies that we could do a lot with microcontrollers. And now, of course, everyone does. But just for fun, my own personal ride has a Holley and and distributor. Even has points, but I cheat a little bit 'cause I use the points to drive a power transistor, and that's what energizes the coil. Never pits the points that way.
The woofer-to-mid crossover is a pretty simple first-order low-pass on the woofer. I use a 5.6mH coil, so crossover is around 250Hz. The midhorn has no electrical high-pass, so the two overlap in the upper-midbass / lower-midrange which helps smooth the top octave of the modal region. However, the mid-to-tweeter crossover isn't as simple. The transfer function is pretty complex. Best to look over the "Crossover" and "Simulations and Measurements" sections of the Pi Speakers FAQ.
In particular, study the information in the link called "Crossover optimization for DI-matched two-way speakers." Even though that document is about a two-way design, it's the exact same process I use to dial-in the upper crossover in a three-way design. So if you can duplicate that process, you can duplicate the crossover. I essentially design a digital crossover using the Spice models in WTPro, and then implement the crossover with passive parts.
The woofer coil should be 5.0mH to 6.0mH. The span is only about a 15% difference and is within tolerance. This is especially true in first-order filters, and even more so in this case with so much overlap between sound sources. We're looking for blending between woofer and midhorn, and any value between 5.0mH and 6.0mH is fine.