Single Driver polarity measurement changes

[Adam_M]

HI!
I'm working on a design, and I struggled with phase alignment between the woofer and mid. To troubleshoot, my plan was to back out, measure each driver individually and see what was going on. Along the way, I measured the mid (with no other drivers connected) out of phase (with the amp), and it didn't look anything like what I expected. When I wired it back up in phase, it was more or less what I thought it should be. Measurement axis and distance don't cause any unexpected changes in the response. How can a driver be out of phase with itself?

[Chris Roemer]

You're not showing us phase (angles), you're showing us FR.
If your mic position doesn't change, your FR should be the same no matter which way the driver leads are connected (if I'm understanding what you did correctly)?
Something else must be going on? ?

[itschris]

Since it looks like you had a crossover filter in the in phase measurement, any chance you reversed the polarity in front of the crossover rather than just connecting the driver in reverse?

[KnuckleDragger]

Since it looks like you had a crossover filter in the in phase measurement, any chance you reversed the polarity in front of the crossover rather than just connecting the driver in reverse?

Hmm ?
If the voltage to the speaker is AC voltage, then why would polarity to the crossover matter ? It would still just switch the phase of the system, correct ? What am I missing ?

[Adam_M]

Chris, you are correct, I didn't show phase - was simply checking by throwing the mid out of phase.

A fresh set of eyes and I think I see the problem. I had the XO wired in such a way that when the mid is in phase, everything works as expected, but when I swap the polarity on the mid, the circuit topology changed. Basically, I inadvertently wired up 2 separate ground points and one of the grounds was connected to the + lead of the mid, while the second was connected to the - lead of the mid with the mid polarity reversed. It might be time to ditch all of the alligator clips and get a couple terminal strips. Spaghetti alligator lead/clip wiring is hard to trace, especially when you've been staring at it for hours.

With the mid in phase, topology is as expected. With the mid out of phase, its a strange topology that produces mathematically predictable, but not expected results. I need to either move 2 leads so the ground point moves consistently or tie the grounds together in a different way. To be confirmed when I have time in the evening to fire the measurement rig back up.

Basically, as they call it in the IT biz, user error. Hopefully that's all it is.

[rpb]

Change the polarity at the driver, or in other words, change it in the wiring after the x-over.

[Adam_M]

RPB, that's what I was doing, but I was using the lead coming out of the mid as a grounding point and didn't move the second wire attached to it. Moving the second wire with the first, or just clipping the second wire to another ground point solved the issue.

[xmax]

You will also see some anomalous behavior leaving sections of the filter hanging
off one of another when only trying to measure part of a filter and of course things
changing with no load. During this phase I build a HP and LP section on the $2
pcb (bandpass too) with extra leads soldered to the woofer LP with 3/16 spades
to easily connect or disconnect each section. Sometimes I will leave a component
lead or 2 unsoldered to connect alligator clips to change out or add parts quickly,
this at least give me a solid foundation to work from, nothing like working hard
for an hour just to find a connection was wrong or something. It is also very
easy to unsolder and change parts when needed (quicker than
battling 17 clips and wire!) I also solder on screw terminals for quick part
changes on the bigger PCBs from Madi.

Part # 260-130

[rpb]

RPB, that's what I was doing, but I was using the lead coming out of the mid as a grounding point and didn't move the second wire attached to it. Moving the second wire with the first, or just clipping the second wire to another ground point solved the issue.

Most of the boxes I'm working on have banana plugs for the speaker. I wire those to the woofer. The wires for the tweeter exit through a couple of drilled holes on the back of the box. When I want to change polarity, I usually just switch it at the banana plugs. Sometimes, I have jumpers attached on both wires for the tweeter where they exit the box, and I switch there instead. Anyway, you found the problem. I was just trying to be helpful.