Hello!
First speaker attempt - narrow and tall to flank tv and replace soundbar.
Is my crossover design in VituixCad heading in the right direction? (Should the graph look like that?)
If yes, then I'm going to start hunting for or producing with a mic the .frd file for the HiVi M6N 6" because I like the way the HiVi aluminum/magnesium cones look.
Drivers:
Tweeter: Dayton DC25T-8
Mid: HiVi (Swan) M4N 4"
Woofer: Dayton GF180-8 6.5"
Thanks for any input!
First speaker attempt - narrow and tall to flank tv and replace soundbar.
Is my crossover design in VituixCad heading in the right direction? (Should the graph look like that?)
If yes, then I'm going to start hunting for or producing with a mic the .frd file for the HiVi M6N 6" because I like the way the HiVi aluminum/magnesium cones look.
Drivers:
Tweeter: Dayton DC25T-8
Mid: HiVi (Swan) M4N 4"
Woofer: Dayton GF180-8 6.5"
Thanks for any input!
D
Deleted member 375592
You may not have accounted for the sound of mids decaying 3dB per doubling of distance while the tweeter and the woofer will decay 6dB per doubling of distance - yet.
Without a phase response it's hard to tell what is happening around the XO points between both driver pairs. Can you post the "6 pack" from VCad?
Also - if you are bothering with a 3 way, why not go for a larger woofer?
Where did you get your FRD and ZMA data? If manufacturer traces, have you applied baffle diffraction (including step loss) before using?
What acoustic offsets (Driver Z) have you allowed for on midrange and woofer?
Did you check the "minimum phase" checkboxes for each driver (if using manufacturer supplied or manipulated data?)
Also - if you are bothering with a 3 way, why not go for a larger woofer?
Where did you get your FRD and ZMA data? If manufacturer traces, have you applied baffle diffraction (including step loss) before using?
What acoustic offsets (Driver Z) have you allowed for on midrange and woofer?
Did you check the "minimum phase" checkboxes for each driver (if using manufacturer supplied or manipulated data?)
Hey hey! Thanks for the responses!
My post was intended to see if I understood what things should generally look like in the graph when designing crossovers.
Replies:
-Not a bigger woofer to keep the 46" or shorter tower as narrow as I can to appease my wife.
-Frd and Zma data from Dayton for tweeter/woofer. Mid data from a user on Parts Express Tech talk. I don't know what baffle diffraction or step is
-Driver Z offset I will need to learn what it means. The speakers will be next to the tv if it's about that
-i haven't checked minimum phase. I'll do that and post the whole graphic tomorrow.
-The 1.2khz dip is there when loading the frd data fresh
I've got a little reading to do!
Again- thanks for taking your time to respond.
My post was intended to see if I understood what things should generally look like in the graph when designing crossovers.
Replies:
-Not a bigger woofer to keep the 46" or shorter tower as narrow as I can to appease my wife.
-Frd and Zma data from Dayton for tweeter/woofer. Mid data from a user on Parts Express Tech talk. I don't know what baffle diffraction or step is
-Driver Z offset I will need to learn what it means. The speakers will be next to the tv if it's about that
-i haven't checked minimum phase. I'll do that and post the whole graphic tomorrow.
-The 1.2khz dip is there when loading the frd data fresh
I've got a little reading to do!
Again- thanks for taking your time to respond.
The 1,2 dip is most likely not the biggest issue, peaks are so much more noticeable/annoying, if you don't mind the dip, then don't tinker with it, that will open a bunch of cans of worms. That is likely because of the width of the baffle. If that's not an issue, let it go.
The dip between 5 and 10kHz is a phase problem. The ear is sensitive at 5-7kHz, so a dip there is also better than a peak but a phase problem in that range is clearly audible. Check if that's a problem. If not, don't touch that either. 😉 If it is a problem, that's because of the low pass filter of the mid and the peak of the mid driver at 5k.
Overall, that's within +/- 3dB and with that, a lot better than ~90% of the soundbars perform. It can likely be improved but you have to decide if it's worth the effort.
The dip between 5 and 10kHz is a phase problem. The ear is sensitive at 5-7kHz, so a dip there is also better than a peak but a phase problem in that range is clearly audible. Check if that's a problem. If not, don't touch that either. 😉 If it is a problem, that's because of the low pass filter of the mid and the peak of the mid driver at 5k.
Overall, that's within +/- 3dB and with that, a lot better than ~90% of the soundbars perform. It can likely be improved but you have to decide if it's worth the effort.
Dips and peaks, yes, but there's no real evidence that we're sensitive to phase.
Can you post all 6 graphs from Vcad?
Also- check the "Reference angle" checkbox on the main frequency response graph (normally black) - so we can see the summed response as well as the individual filtered driver responses (you have as blue green and red) eg:
What is your minimum impedance like ? (and where in the frequency spectrum?)
Also- check the "Reference angle" checkbox on the main frequency response graph (normally black) - so we can see the summed response as well as the individual filtered driver responses (you have as blue green and red) eg:
What is your minimum impedance like ? (and where in the frequency spectrum?)
The human ear is not sensitive to the absolute phase but it is very sensitive to phase change and since that is already a quite short wavelength, the movement of the head is in that frequency range enough not only to hear it but to be very annoying.
You're welcome 🙂
It doesn't matter if it's perfect, it's enough if it's fine for your use and your expectations but if it can be improved with low effort and low cost, then it's definitely worth doing just that.
I think the issue is that you are concentrated on finding something me being wrong instead of looking for what the actual problem is. I don't know why you're so fixated on that and I don't really care but on your warpath you are about to hijack and destroy multiple threads. That's not how a moderator should act.
The reason why an interference in that range is annoying is because it is the range which is most important for the recognition of the location of a sound.
I surely assumed you are aware phase differences result in interferences. I'm sorry I assumed wrong.
It seems more like you are having a personal problem with me. I'm open to discuss this in a PM.
BTW, I have not been knighted yet.
Thank you to everyone contributing, the directed learning is awesome.
@ICG - after reading about baffle step, I'm sure you are right that the .frd created by someone at home for the mid (HiVi M4N) has the baffle step around 1.2khz from their setup.
@Dave Bullet - after reading about what the reference angle is, I've played with things to try and flatten it out. I had it wrong thinking the individual driver curves mattered - it's their combination that does. Also set to minimum phase for all drivers. Also learned about Z offset - that I'll need to measure in the cabinet and then apply. Thanks for all the help with subjects for me to learn about.
I don't know what "minimum impedence" means for this - but they are all 8ohm speakers.
Attached is the "6 pack" of the crossover design.
Thanks for any input.
@ICG - after reading about baffle step, I'm sure you are right that the .frd created by someone at home for the mid (HiVi M4N) has the baffle step around 1.2khz from their setup.
@Dave Bullet - after reading about what the reference angle is, I've played with things to try and flatten it out. I had it wrong thinking the individual driver curves mattered - it's their combination that does. Also set to minimum phase for all drivers. Also learned about Z offset - that I'll need to measure in the cabinet and then apply. Thanks for all the help with subjects for me to learn about.
I don't know what "minimum impedence" means for this - but they are all 8ohm speakers.
Attached is the "6 pack" of the crossover design.
Thanks for any input.