OH, one thing I did want to mention, if memory serves, if you make a shorter tweeter than your mids, the falloff at a distance is not the same. You need to be sure you measure the overall levels at the listening location, or have some way to adjust it.
That is, at 10' the tweeter may be a lot lower than the mid-woofs compared to 3'
That is, at 10' the tweeter may be a lot lower than the mid-woofs compared to 3'
My original intention is to align the AMT's plane to the midrange voice coils vertically. In a MMMMT setup it's easy, in a MMTMM setup.. well, not anymore because when I design the midrange-tweeter box there has to be a dip in the middle for the tweeter where the dip's walls might act as some kind of horn maybe. Okay, perhaps 30° deviation from the flat front plate but still.. (and I'm not sure if it's 30°at all.. if I wanted to keep driver distance small, I would have to count with a steep "wall" around the tweeter which makes the horn-effect even stronger).
A small, horizontal waveguide on the AMT may help it match the woofers dispersion pattern and give you a more uniform (horizontal) wave launch. Plus boost the sensitivity some.
I'd be looking at packaging of the tweeter at ear height and the arranging the woofers accordingly. That may well end up being either an MTMMM or TMMMM or MMTMM, but not likely a MMMTM (respectively from top to bottom)
I'd be looking at packaging of the tweeter at ear height and the arranging the woofers accordingly. That may well end up being either an MTMMM or TMMMM or MMTMM, but not likely a MMMTM (respectively from top to bottom)
I was thinking as simple as a conical shape with a bit of hand sanding of the transitions -- if you could get close to an oblate spheroid shape, that'd probably be best. The angle of the conical section would be driven by the dispersion of the midrange, or where you'd deem best, maybe something like a 120 degree angle or so.
What? I thought 1/2 wavelength or less was the rule of thumb to avoid lobing?
Isn't this easy to fix by just running the tweeter inverted?
What? I thought 1/2 wavelength or less was the rule of thumb to avoid lobing?
"Arguably the worst option is to use 1/2 wavelength spacing. Because that will create a null. (Because the drivers are out-of-phase at the xover point, due to them being 180 degrees out-of-phase.)"
To avoid a null, you have to compensate by manipulating the phase. That can be done by:
1) flipping the polarity
2) using asymmetric xover slopes
3) using delay
4) all of the above
Any reason why one shouldn't go this route?
What about if the center-to-center spacing is over or below 1/2 wavelength? Say 1/4 or 3/4 of a wavelength.
1 Wavelength CTC spacing is always worse then 1/2 wavelength; the lobing you try to avoid will happen at half the frequency and more lobing for the higher frequencies is created; as close as possible results in the least lobbing.
An AMT/planar/ribbon has a cylindrical wavefront, little vertical off axis response and thus ctc spacing is less important.
I would put the tweeter in the middle except if I like to be able to put a second speaker up side down on top (tweeters close to each other) of it.
An AMT/planar/ribbon has a cylindrical wavefront, little vertical off axis response and thus ctc spacing is less important.
I would put the tweeter in the middle except if I like to be able to put a second speaker up side down on top (tweeters close to each other) of it.
I'll post a proper response once I'm done with work.
The answer is complex
I posted an answer here, in the 'original' thread about this setup : Optimal MTM Geometry
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.