Decadent_Spectre
Well-Known Member
I thought you were building a reflex cabinet?
2-way horn system based on the mk3b2
- Thread starter aeroash
- Start date
I thought you were building a reflex cabinet?
Vineethkumar01
Well-Known Member
There is a front short horn for the 15PR400. The initial plan got changed slightly
The bass alignment is bass reflex/ported.
Decadent_Spectre
Well-Known Member
There is a front short horn for the 15PR400. The initial plan got changed slightly
The bass alignment is bass reflex/ported.
Have you made it like the Altec VOTT style then? Either way a short horn in front of a reflex has historically provided good subjective results. IMO a good decision. Enjoy.
Vineethkumar01
Well-Known Member
grindstone
Active Member
Thanks for all that. See if this imports as a start. If not, here's an eyeball-swagged input screen. You'll need CAD for a proper Vtc, Atc, L12-L34 etc. This gets you a better first cut at the "pipe organ" resonances (LS Wiz->Pwr->Combined 1). For some reason, my v55 HR has buggy filling--will reinstall old rev in a bit.
Attachments
Vineethkumar01
Well-Known Member
@grindstone: Thanks for these sims.
I am not very familiar with advanced modeling using Hornresp.
Can you explain a bit more details about modeling a vented enclosure with the front short horn in the way you have shown above and what the longitudinal model means?
I am not very familiar with advanced modeling using Hornresp.
Can you explain a bit more details about modeling a vented enclosure with the front short horn in the way you have shown above and what the longitudinal model means?
grindstone
Active Member
It's just a way to capture the lowest freq air column resonances (the ones hardest to damp) instead of treating the volume only as a pure compliance. Parallel walls => standing waves. Straight pipe closed at both ends = half wave resonator, but with two holes in it (driver & port).
Maybe look at the open cylinder harmonics three things down here and just mentally shift the waves 90 degrees to put the nodes and antinodes in the right spot for a both-ends-closed pipe Modeling this way is the attempt to capture that and get it in the impedance and the tuning and the response, etc. As-conceived, the reality will measure lower-tuned than the vcad 5 parameter reflex model--HR tuning modeled the height way will be closer. You can then stick some filling swag in there, see what it does, and export that impedance to use--not as good as reality, but better than simple reflex model unless the real box doesn't have that cross-batt, etc.
If the real port was in the middle of the back, then the way you had it would be a better model (assuming cross-batt still used or non-parallel top/bot). How you model depends on what you want to see (and what you *know* is happening), right?
Again, at this point, the boxes are done and it's all much ado about nothing except for reference/posterity, but this is why to model that way. Where you put the driver can null-out a harmonic. Where you put the port can go after a higher-frequency, etc. The cross-section is the plan area of the "triangle"(ish) main body of the air column. This stuff is a continuum/gray with regard to the distribution of harmonics. The details matter once extra shapes or chambers are involved--and it's almost never well-behaved. Tapering changes things as well. End-driven, end-tapped, tall + skinny is the worst/hardest. Squatty is feature when things get big
When height and driver & port offsets are chosen, you are freezing-in the distribution of the harmonics richocheting around inside your cabinet. The only tools left are lossy (or more acoustic filtering chamber games).
Adding on a fronthorn length and throat chamber length goes into the resonant circus, too, but here you sidestepped some ugly MR ringing by using the big throat (St > Sd). I didn't see that yesterday and just blindly assumed that all the wood gnawed-out was for a conical front chamber volume to lowpass everything acoustically--figured you guys already hashed that out and did it to eat HF from the woofer, etc. Strictly-speaking, I also messed-up the throat chamber which, in this specific model, should have been put in the RC entry as Vrc 2.5 and Lrc 2.92 with Vtc=Atc=0 so it went in the right place, but the big throat mostly negated the mistake.
Pragmatically, every tradeoff in a big 2-way is a painfully difficult choice and here the driver must be high and things are what they are and all will be fine. The symmetrical fronthorn offsets the driver down a ways to its own benefit. If this were all a brand new blank sheet design, you could iterate various port and driver locations, etc.
The main column resonance is an extra "load" (or free damping if you prefer) when boxes get toward that size. All reflex boxes are a combination of the "normal" (simple) LC (inertance-compliance) resonance and the air column size resonances--it's just that most boxes are smaller so either the harmonics don't get excited as hard (ie are out of passband) or are more easily dampable (higher freq). In big boxes, these resonances can muck-up the low-mids (response squiggle 2XX-6XX). The cross-batt of fiberglass between driver and port will aid cleaning that up.
As in all matters of audio, you will see opinions advocating everything from the empty pure compliance to the nearly-filled TL-type. Acoustically resistive treatments are a wonderland for experimentation that, once-heard, become as personal as food preferences IMO. Critical damping is a fine goal but most people also like a little ringing/ripple to "fill-out" the sound of soundbodies.
The old ways of blind empiricism still work, too. Battery (1.5V) + momentary switch (woofer only!). Add layers of fabric over ports and repeat make/break. Listen both directions and add fabric until it clicks and doesn't "bong". Stapling fabric over the rear of the driver damps both the driver and the cab instead of just the port. Loading the port with straws, etc for some more friction, etc. Hanging curtains away from cab walls. All these things are tools. The filling/lining/damping is personal, it's box-specific, and it depends on what's driving it, too (source resistance).
Some acoustically-resistive things are simmable, but it gets back to needing to measure to verify the sims before they're trustable anyway. To add more detail, well, HR is out and other tools are required (each with their own issues).
I really only meant to show the standing wave influence, slap a hunk of filling in the middle and show all that trash getting knocked-down It'll be fine. If it's empty (unfilled), should have a couple squiggles 2XX,4XX,6XX Hz, should show in the impedance, too (empty). Congrats on the big boxes and well done. Wish I could hear 'em. Putting them in a corner will help lift the bottom to catch-up to the fronthorn some but maybe you have subs for low-lows, etc and don't need it or were going to tune higher, etc. At any rate, outstanding project and thanks for posting.
Maybe look at the open cylinder harmonics three things down here and just mentally shift the waves 90 degrees to put the nodes and antinodes in the right spot for a both-ends-closed pipe
Modeling this way is the attempt to capture that and get it in the impedance and the tuning and the response, etc. As-conceived, the reality will measure lower-tuned than the vcad 5 parameter reflex model--HR tuning modeled the height way will be closer. You can then stick some filling swag in there, see what it does, and export that impedance to use--not as good as reality, but better than simple reflex model unless the real box doesn't have that cross-batt, etc.If the real port was in the middle of the back, then the way you had it would be a better model (assuming cross-batt still used or non-parallel top/bot). How you model depends on what you want to see (and what you *know* is happening), right?
Again, at this point, the boxes are done and it's all much ado about nothing except for reference/posterity, but this is why to model that way. Where you put the driver can null-out a harmonic. Where you put the port can go after a higher-frequency, etc. The cross-section is the plan area of the "triangle"(ish) main body of the air column. This stuff is a continuum/gray with regard to the distribution of harmonics. The details matter once extra shapes or chambers are involved--and it's almost never well-behaved. Tapering changes things as well. End-driven, end-tapped, tall + skinny is the worst/hardest. Squatty is feature when things get big
When height and driver & port offsets are chosen, you are freezing-in the distribution of the harmonics richocheting around inside your cabinet. The only tools left are lossy (or more acoustic filtering chamber games).
Adding on a fronthorn length and throat chamber length goes into the resonant circus, too, but here you sidestepped some ugly MR ringing by using the big throat (St > Sd). I didn't see that yesterday and just blindly assumed that all the wood gnawed-out was for a conical front chamber volume to lowpass everything acoustically--figured you guys already hashed that out and did it to eat HF from the woofer, etc. Strictly-speaking, I also messed-up the throat chamber which, in this specific model, should have been put in the RC entry as Vrc 2.5 and Lrc 2.92 with Vtc=Atc=0 so it went in the right place, but the big throat mostly negated the mistake.
Pragmatically, every tradeoff in a big 2-way is a painfully difficult choice and here the driver must be high and things are what they are and all will be fine. The symmetrical fronthorn offsets the driver down a ways to its own benefit. If this were all a brand new blank sheet design, you could iterate various port and driver locations, etc.
The main column resonance is an extra "load" (or free damping if you prefer) when boxes get toward that size. All reflex boxes are a combination of the "normal" (simple) LC (inertance-compliance) resonance and the air column size resonances--it's just that most boxes are smaller so either the harmonics don't get excited as hard (ie are out of passband) or are more easily dampable (higher freq). In big boxes, these resonances can muck-up the low-mids (response squiggle 2XX-6XX). The cross-batt of fiberglass between driver and port will aid cleaning that up.
As in all matters of audio, you will see opinions advocating everything from the empty pure compliance to the nearly-filled TL-type. Acoustically resistive treatments are a wonderland for experimentation that, once-heard, become as personal as food preferences IMO. Critical damping is a fine goal but most people also like a little ringing/ripple to "fill-out" the sound of soundbodies.
The old ways of blind empiricism still work, too. Battery (1.5V) + momentary switch (woofer only!). Add layers of fabric over ports and repeat make/break. Listen both directions and add fabric until it clicks and doesn't "bong". Stapling fabric over the rear of the driver damps both the driver and the cab instead of just the port. Loading the port with straws, etc for some more friction, etc. Hanging curtains away from cab walls. All these things are tools. The filling/lining/damping is personal, it's box-specific, and it depends on what's driving it, too (source resistance).
Some acoustically-resistive things are simmable, but it gets back to needing to measure to verify the sims before they're trustable anyway. To add more detail, well, HR is out and other tools are required (each with their own issues).
I really only meant to show the standing wave influence, slap a hunk of filling in the middle and show all that trash getting knocked-down
It'll be fine. If it's empty (unfilled), should have a couple squiggles 2XX,4XX,6XX Hz, should show in the impedance, too (empty). Congrats on the big boxes and well done. Wish I could hear 'em. Putting them in a corner will help lift the bottom to catch-up to the fronthorn some but maybe you have subs for low-lows, etc and don't need it or were going to tune higher, etc. At any rate, outstanding project and thanks for posting.
Vineethkumar01
Well-Known Member
@grindstone: Thanks a lot for the detailed discussion above. I got the general idea now.
This kind of richness of discussion/analysis/ideas is what we have been missing in this thread so far.
So thanks again for your posts above.
Yes, I agree that we didn't spend much time on iterating with the port/driver locations and box sizes in the software.
So it is not an optimized box from that point of view.
We went with a simplistic enclosure modeling in VituixCAD, a tolerable box size, and kind of a more aesthetically pleasing box design.
We thought that we can treat whatever comes up due to standing wave problems with damping, to a reasonable extent.
So damping is definitely going to be an integral part of this box.
Now that the boxes are done, let's see what all things show up in the measurements and if we can do something about it..
This kind of richness of discussion/analysis/ideas is what we have been missing in this thread so far.
So thanks again for your posts above.
Yes, I agree that we didn't spend much time on iterating with the port/driver locations and box sizes in the software.
So it is not an optimized box from that point of view.
We went with a simplistic enclosure modeling in VituixCAD, a tolerable box size, and kind of a more aesthetically pleasing box design.
We thought that we can treat whatever comes up due to standing wave problems with damping, to a reasonable extent.
So damping is definitely going to be an integral part of this box.
Now that the boxes are done, let's see what all things show up in the measurements and if we can do something about it..
You are doing a great job...
arunkvivek
Well-Known Member
Can't wait to see the finished ones... massive efforts for DIY
While the bass bins were puttied and left to dry I snuck in and finished the new exponential adaptor. The inner shell is printed in resin and enveloped in mahogany. Ideally, I would've loved to have the entire adaptor milled in hardwood, but my fabricator expressed limitations. Nothing changes functionality-wise, it should do for now.
Vineethkumar01
Well-Known Member
A whole lot of measurements/data incoming
Set up for taking polar measurements
Bass horn raw polar measurements (5ms gated measurements from 1.5m away from 0 degrees to 90 degrees taken in 10 degree steps)
Faital pro HF146 on Mk3b2 raw polar measurements (5ms gated measurements from 1.5m away from 0 degrees to 90 degrees taken in 10 degree steps)
Reference axis taken as 20degrees off axis response
Nearfield responses of the 15PR400 driver & port (taken 5mm away from woofer dustcap)
Testing damping inside the bass box
Impedance measurement of bass reflex box
Impedance measurement of HF146 on mk3b2
A potential Crossover
Corresponding full spectrum polars (designed for listening 20 degrees off axis)
Normalized directivity
Maybe redundant, but here is a reverse null test
That's it for now. Box finishing, listening based adjestments of crossover/EQ etc to follow
Set up for taking polar measurements
Bass horn raw polar measurements (5ms gated measurements from 1.5m away from 0 degrees to 90 degrees taken in 10 degree steps)
Faital pro HF146 on Mk3b2 raw polar measurements (5ms gated measurements from 1.5m away from 0 degrees to 90 degrees taken in 10 degree steps)
Reference axis taken as 20degrees off axis response
Nearfield responses of the 15PR400 driver & port (taken 5mm away from woofer dustcap)
Testing damping inside the bass box
Impedance measurement of bass reflex box
Impedance measurement of HF146 on mk3b2
A potential Crossover
Corresponding full spectrum polars (designed for listening 20 degrees off axis)
Normalized directivity
Maybe redundant, but here is a reverse null test
That's it for now. Box finishing, listening based adjestments of crossover/EQ etc to follow
Putty.
Superb speaker connectors! well built, sturdy, smooth rotation.
This one needs rubber wheels in order to save my back. The black door is removable and is sealed with a gasket. The driver is mounted from the inside.
Yelamanchili manohar
Well-Known Member
@aeroash and @Vineethkumar01 ...what a gorgeous build this has been 
. The attention to detail and workman ship have been top notch. I must say, there was a lot of love poured into this one
The result is one beautiful speaker I could look at all day. Though the graphs and curves are beyond me to interpret, I bet it sounds great and looks a stunner.
Extremely well done. You both have set the bar very very high for any future DIY builds on our forum. After this, I know i can never post pics of my own builds. But atleast i save some money on a new camera 
.
@grindstone ...thanks for being around here when we need you the most. Very happy to have someone with your knowledge and experience, and a willingness to share
Cant wait to see what comes next...All the very best
. The attention to detail and workman ship have been top notch. I must say, there was a lot of love poured into this one The result is one beautiful speaker I could look at all day. Though the graphs and curves are beyond me to interpret, I bet it sounds great and looks a stunner.
Extremely well done. You both have set the bar very very high for any future DIY builds on our forum
. After this, I know i can never post pics of my own builds. But atleast i save some money on a new camera .@grindstone ...thanks for being around here when we need you the most
. Very happy to have someone with your knowledge and experience, and a willingness to share Cant wait to see what comes next...All the very best
mahaaprasaad
Well-Known Member
Splendid work..!!! Gorgeous look with that wooden background and floor;
Vineethkumar01
Well-Known Member
Current crossover details:
Your Effort deserves a drive from Bangalore to Mangalore to just see these in person, listening would be a bonus, Fantastic work
keep going
Last edited:
You're more than welcome.Your Effort deserves a drive from Bangalore to Mangalore to just see these in person, listening would be a bonus, Fantastic work
keep going
Order your Rega Turntables & Amplifiers from HiFiMART.com - India's reputed online dealer.
