Felipe, this usually works. Midrange and tweeter in phase, woofer out of phase. The hole at woofer/midrange XO should disappear once this is sorted, assuming XO filter is reasonably good.
Thanks Lojzek, my fault because when was the filter 2 way I reversed the mid-woofer and forget to left in phase with tweeter & put the woofer out of phase.
no good... baffle step on sba gets the whole balance wrong, with way too much bass because of room modes and corner placement. Moreover the SS needs attenuation.
btw you cannot assess bass integeation and global balance by 1 m distance measurements: you need the room response too.
Felipe, the general agreement in designing XO filter is to pull speakers away from corners, walls and other nearby objects. You have plenty of room so move speakers in the center and then perform measurements. First design a filter as if the box is to stand in free space, microphone not too far from baffle and once you have a decent version you will experiment and listen for awhile to get the feeling. And secondly, use the simulation tools to start working on the XO filter design for it's so much quicker to arrive at a workable solution than exchanging real parts. The measurements of the units should be taken at a desired mic position, then its spl response imported to a software to extract the minimum phase and then loaded to a simulator. The simulator also needs the measures between the units in respect to microphone. Additionally there should be extra measurements of the mid and tw tied in parallel together to find out Z offset, and these offsets should be defined in simulator as well. Sounds complicated, but it's worth it. There is a white paper by J.Bagby on finding the acoustic offset. Oh, forgot to tell you, you'd need to measure impedance of each unit so until you get this sorted, I guess this won't do.
Last edited:
Lojzek, if the speakers would stay away from walls and corners aftermeasurement i mighragree. But as they will return to their original and special location, i think it is much better to design respecting their special boundary conditions.
Merlin, as commented your idea of compensating BS on the woofer is a bad one for 2 reasons:
1º Acoustically there is no baffle step as a driver located close to a boundary works as on infinite baffle.
2º Electrically your BSC stuff destroys the nice LR 12 you had before.
On the other hand as your hipass on the SS is way too sharp ( High Q) you get this big deep in the low mid and also boosts badly the bass.
If i were you i would cut SBA higher and softer the SS.
In fact the SS does not need any high pass, and a simple 2.5 way should work better than what you are messing with now...🙄
1º Acoustically there is no baffle step as a driver located close to a boundary works as on infinite baffle.
2º Electrically your BSC stuff destroys the nice LR 12 you had before.
On the other hand as your hipass on the SS is way too sharp ( High Q) you get this big deep in the low mid and also boosts badly the bass.
If i were you i would cut SBA higher and softer the SS.
In fact the SS does not need any high pass, and a simple 2.5 way should work better than what you are messing with now...🙄
Attachments
Last edited:
High passed SS & without baffle step SBA, woofer out of phase, mid & tweeter in phase in green, yellow with BS.
Attachments
Last edited:
High passed SS & without baffle step SBA, midwoofer out of phase, woofer & tweeter in phase in green, yellow woofer out of phase midwoofer and tweeter in phase.
Attachments
Last edited:
Thanks Lojzek I know is the way, I did with the 2 way xover, now I don't want to change the 2 way xover, I only want to add the 3rd way.
![WP_20170401_001[1].jpg](/community/data/attachments/566/566915-dc4ad3e6bacf137fd9ed29931fdd4f95.jpg?hash=3ErT5rrPE3)
w+t in phase, midwoofer out of phase, SS highpassed.
Attachments
Last edited:
midw+t in phase, woofer out of phase, SS highpassed.
Attachments
Last edited:
Measurement post 76 isn't good, this one is good in green, yellow is grph post 77 midw+t in phase, woofer out of phase, SS highpassed to compare.
Attachments
Last edited:
- Status
- Not open for further replies.