Is there a rule of thumb? I've found that the angle about 7 - 8 degree generally works well at least with using first order filter with the mid situated just below the tweeter on the baffle. For my case, the measurement was done at between the mid and the tweeter.
I supposed if you measure on the mid axis, then the angle might be different.
I supposed if you measure on the mid axis, then the angle might be different.
There has to be a point (typically below this) where the distance to each driver measures the same, and you can derive the angle from that.
That's a good suggestion. I may try that next time.
So far I've been measuring the tweeter and mid separately, then measure them together by connecting the + and - of the two driver together. Then I would sum up the tweeter and mid in the simulation to see if it matches when both the tweeter and mid measured together with the mic.
But what you suggested seems like a very good alternative.
Reverse null could be challenging as the responses need to be properly set.
Another way is to measure each driver with the timing fixed. Manually delay one driver until phase slope approaches the minimum phase derived from the response. Apply the same delay to the other driver and if phase doesn't line up with its mimimum phase case, move the mic and try again.
Another way is to measure each driver with the timing fixed. Manually delay one driver until phase slope approaches the minimum phase derived from the response. Apply the same delay to the other driver and if phase doesn't line up with its mimimum phase case, move the mic and try again.
Should one assume that a reverse null is a part of every crossover? I feel it is misleading.
Why not measure at the listening axis? In a practical point if view, you can use taller box instead of tilting back, all you need to know is the distances from the drivers to the listening spot. A passive electrical solution is to use different slopes, but this can be tricky with low order slopes.
I strongly agree that the reverse null should include the crossover in place. Different crossover topologies and slopes introduce their own inherent relatives phase shifts (delay) to each section. The overall offset requires including these as well as the physical acoustic centres offset. I am not a fan of tilt-back because of the uneven response and lobing issues when a tweeter is off-axis. Much better is a baffle step but a dummy rig is useful to measure the required offset by reverse null.
Of course it is andy2. Lobing/nulls between drivers radiation pattern can be caused by physical misalignment (position) or phase misalignment (xo). A phase misalignment can be induced by an electrical correction or physical position at xo frequency. Both scenarios result in similar delay/timing issues, ie. main lobe is tilted off axis.
Am I misunderstanding what are you trying to correct?
Am I misunderstanding what are you trying to correct?
I think we may have two separate the
1. driver acoustic center
2. the phase alignment
It's possible they are related and in some way they are. I've pointed out that the measuring axis (mid or halfway between mid and tweeter) may affect the tilting angle.
The listening space is 3-dimensional and the axis you're listening is probably another variable.
If using first order filter, once the driver acoustic centers of the tweeter and mid are fixed, it's hard to adjust because you're being constrained by the first order filter slope. On the other hand, if you're using asymmetric xover, then you could adjust the phase even if the driver acoustic center is not aligned (such as using a combination of 2nd and 3rd order) and I think this might be what you're trying to point out.
wouldn't you measure the drivers on the tweeter axis and determine the delay needed to align the acoustic centers. Then use Geometry (A2+B2=C2 where the intersection of A &B is the mic and C is the slope of the baffle) to figure out what degree angle is needed to put the tweeter at that distance behind the woofer?
Seems pretty straight forward - or maybe I'm too logical?
Seems pretty straight forward - or maybe I'm too logical?
Ok, so you want to align the drivers physically first by tilting, and consider their response as the second order of business.
Are the suggested methods OK for you to do this?
I was trying to make a case that aligning the driver acoustic center and aligning the driver phase response are mutually exclusive, but it's not as clear cut, therefore those two things my be related. And I said that the listening space is 3-dimensional, and by tilting the baffle, you modify the driver phase at a given listening axis.
Acoustically, the other question to ask, is that is there a difference in using a step baffle vs. a tilted baffle?
Anyway, it seems to be more complicated and not as clear-cut as I had thought.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.