Cool. Might have to do something similar for the Beyma 15P1200Nd drivers I have lying around.
I know you made a point of ensuring vent velocity stays below 15m/s. Is this essential? What would you consider to be the upper limit for PA subs?
I've done a little reading around, but not found much solid.
Chris
I know you made a point of ensuring vent velocity stays below 15m/s. Is this essential? What would you consider to be the upper limit for PA subs?
I've done a little reading around, but not found much solid.
Chris
I've always wondered in the Tuba 24 comparison charts on the BFM forum about the seemingly pathetic graph for the Yamaha SW118V 95-96dB only to 60Hz followed by roll off when 1 of my 12" maintains that efficiency to 40Hz in half the size.
Tests of DJ's Dream simulation a vented box I designed a while back (190L, 41.5Hz tuning) cab (8 cubic feet external with 12mm ply) predict 98-100dB sensitivity all the way to 40Hz tuning. In general a box the size of a Yamaha SW118V if tuned low (30Hz) should see about 96dB all the way to there!
Interestingly the guy on this video 6 x RCF LF18G400 Bassreflex, 2 x JM-sat212 - YouTube built subwoofers very similar to the DJ's Dream design, though I did not send him any plans and am in no way affiliated with the build process. Even the woofer RCF LF18G400 is the best response I've seen yet when I entered the T/S parameters in hornresp.
In general, overly high QTS drivers give better efficiency in the middle of the passband but less at port resonance and upper bass, example 102dB at 58 Hz and only 96dB at 40Hz tuning. They also run out of excursion quicker because of the peak displacement in the middle of the passband.
I've found for a tuning around 35-40Hz that an Fs/Qts ratio of between 100-120 respectively gets the best SPL response in Bass Reflex. The enclosure size would then be dependent on combinations of BL, Mmd and Vas of which I don't know off the top of my head but trial in hornresp.
Anyway, sorry if I went off topic. 15m/s is a decently slow vent speed. It should sound good. I don't know what the upper limit is, though I try to keep below 30m/s, although ports like this should smooth the entry and exit.
Tests of DJ's Dream simulation a vented box I designed a while back (190L, 41.5Hz tuning) cab (8 cubic feet external with 12mm ply) predict 98-100dB sensitivity all the way to 40Hz tuning. In general a box the size of a Yamaha SW118V if tuned low (30Hz) should see about 96dB all the way to there!
Interestingly the guy on this video 6 x RCF LF18G400 Bassreflex, 2 x JM-sat212 - YouTube built subwoofers very similar to the DJ's Dream design, though I did not send him any plans and am in no way affiliated with the build process. Even the woofer RCF LF18G400 is the best response I've seen yet when I entered the T/S parameters in hornresp.
In general, overly high QTS drivers give better efficiency in the middle of the passband but less at port resonance and upper bass, example 102dB at 58 Hz and only 96dB at 40Hz tuning. They also run out of excursion quicker because of the peak displacement in the middle of the passband.
I've found for a tuning around 35-40Hz that an Fs/Qts ratio of between 100-120 respectively gets the best SPL response in Bass Reflex. The enclosure size would then be dependent on combinations of BL, Mmd and Vas of which I don't know off the top of my head but trial in hornresp.
Anyway, sorry if I went off topic. 15m/s is a decently slow vent speed. It should sound good. I don't know what the upper limit is, though I try to keep below 30m/s, although ports like this should smooth the entry and exit.
To a flared port in hornresp, instead of doing a ported rear chamber, have no rear chamber and use the throat chamber (in cm^3 NOT dm^3) and construct the port using the horn segments. Use the PAR flare for linear area expansion if it's a port like sine143's design. Then do the combined response, set it to zero if the port is on the same face as the driver.
As you have reversed from simulating a ported rear chamber direct radiator, to a ported big throat chamber and no rear chamber, your path length difference will be positive when the port is nearer than the driver, so if you have a design with a flared port on the back and the driver on the front facing you, and this design is 60cm deep, the path length difference will now be -60cm instead of 60cm.
I would have expected the Yamaha sensitivity to be around 101dB over the narrow band.. although lower tuning is probably responsible too.
Anyway nice little box with a decent little hump at tuning. I would have expected the JBL subwoofer to be better than it was. BTW I'll have to pinch that port design for my bandpass. First I need a job though!
Anyway nice little box with a decent little hump at tuning. I would have expected the JBL subwoofer to be better than it was. BTW I'll have to pinch that port design for my bandpass. First I need a job though!
Off topic question: Which version of hornresp has the low and high pass filter settings and how they affect excursion? I usually go 2-5Hz below tuning with 24dB/octave filters as a safe guess but I like that you can check out of passband excursion.
Personally I'd try to keep it significantly below passband peak because it's easier to boost gain below passband and exceed Xmax without hitting voltage limit.
Thankfully I don't have to limit my bass reflex subs as the IPR 1600 is sized just right for them within high pass limits such that the DDT light comes on at about the hardest I'd push them.
Personally I'd try to keep it significantly below passband peak because it's easier to boost gain below passband and exceed Xmax without hitting voltage limit.
Thankfully I don't have to limit my bass reflex subs as the IPR 1600 is sized just right for them within high pass limits such that the DDT light comes on at about the hardest I'd push them.
The Filter Wizard tool was introduced in Version 32.00.
Suggest you download a copy of the latest release, Version 36.30, to take advantage of the numerous new features added since then.
Last edited:
Hey Sine, I'm interested in how this works. I've been comparing the Dayton PA385S in a few enclosures (Quarter Wave Vs Helmholtz) and since I have limited knowledge of modeling/modifying existing designs of the quarter wave based "Karflex" designs discussed in another thread, I have been playing with a simple reflex iteration in hornresp.
When trying the double flared port as you've used, I actually saw that flaring the ends RAISES Fb for a given overall port length. How can I use this principal to maintain a lower tuning with less port length and lower particle velocity?
That leads me to my second question: how do you simulate port velocity in HR! I only see the option when using a TH style system.
Lastly, is there any merit in trying a port method like BFM's Omni full range cabs? I don't see how those can play so low with the driver firing into that short of a port.
Thanks!
Sent from my iPhone using Tapatalk
Hey Chips. dont worry, took me forever to find port velocity. believe forum member GM pointed it out
calculate. go to power response
go to output, change to "port", or if you have modelled like I have (using horn segments in par), change it to horn, to provide you with only the output of the port. this will turn velocity option black so you can actually use it.
so.
heres my reasoning between the flared port being "shorter" for a particular tuning.
vs a straight port the same size as the smallest part of the flared port, the straight port will be shorter, HOWEVER, particle velocity will be above desired 15 m/s.
If I repeat this, with a straight port the same size as the mouth of the flared port, the straight walled port will be LONGER than the flared port. Particle velocity will be acceptable at this point. by restricing the port a bit in the center, we lower tuning (allowing us to shorten the port, to maintain the tuning freq).
dealing with pure sims, you will see no "performance increase" from using the flared port method vs straight port, but since we know that port compression exists, we must go by our rough " guideline" of 15 m/s port velocity in order to confirm that the box will perform "as simmed" at high power.
BFMs Ports chuff. I have 4 omnitop12s. my *biggest* complaint with them is the ports. I've rejiggered his plans for a revised version (trapazoidal, slot ports on either side of the horn, with plenty of port area).
calculate. go to power response
go to output, change to "port", or if you have modelled like I have (using horn segments in par), change it to horn, to provide you with only the output of the port. this will turn velocity option black so you can actually use it.
so.
heres my reasoning between the flared port being "shorter" for a particular tuning.
vs a straight port the same size as the smallest part of the flared port, the straight port will be shorter, HOWEVER, particle velocity will be above desired 15 m/s.
If I repeat this, with a straight port the same size as the mouth of the flared port, the straight walled port will be LONGER than the flared port. Particle velocity will be acceptable at this point. by restricing the port a bit in the center, we lower tuning (allowing us to shorten the port, to maintain the tuning freq).
dealing with pure sims, you will see no "performance increase" from using the flared port method vs straight port, but since we know that port compression exists, we must go by our rough " guideline" of 15 m/s port velocity in order to confirm that the box will perform "as simmed" at high power.
BFMs Ports chuff. I have 4 omnitop12s. my *biggest* complaint with them is the ports. I've rejiggered his plans for a revised version (trapazoidal, slot ports on either side of the horn, with plenty of port area).
You can use Flare It program to find acceptable "core limit" and "chuffing limit" port and flare areas respectively.
Also, chips, remember that you can't flare out too rapidly or the flare won't really be effective at controlling velocity regardless of what the sim says. Don't flare out faster than a roundover bit does. You can flare out slower than that if desired, but don't flare out too rapidly.
Flaring the ends and using smaller ports can save a lot of space, it's a really good idea.
Ah yes, I see the distinction now! If I recall, you were often recommending a Dayton PA385 in a BR alignment in the past as I was still struggling to accept physical limitations of sound haha
This seems to be the dream driver for my applications (decent output, lowish 35-40Hz Fb, and mobile) the THAM15 looks killer but simply doesn't go low enough. Many of my tunes I like to play have pretty prominent kicks in the high 30's which I'm not willing to give up with the 40Hz F10!
From what I recall, your OTH40C didn't go quite as low with this driver as it did using the FaitalPRO, also I believe you weren't keen on the assembly procedure!
After Playing on Hornresp for a while it seems there simply isn't a better driver with ANY significant advantage over this unit that will cater to what I'm looking for. EXCELLENT recommendation! (I should have just listened haha)
In fact I'm about to start SolidWorksing a model for this cab for once I have some extra disposable income!
A last question,
the very knowledgable Matthew Morgan J has commented that in practice he has found that Bass Reflex units don't quite live up to sims as he'd like and that he's had more luck/consistency with quarter wave style cabs of the same size.
The alternative to this reflex I've arrived at (80L internal, 24x20x18" external) is a Karlson aperture offset driver based cab. Are you aware of any advantages to that cab over a bass reflex???
Thank you! checking port velocity now
EDIT: YIKES, port velocity with 10mm Xmax was well into 30m/s. Using the same size port length divided into thirds with a doubling of CSA @ each end (only a 21 degree flare) It is down to 20.5 m/s with no 4th order Butterworth.
Sine, were your results favorable when simming 19m/s?
This seems to be the dream driver for my applications (decent output, lowish 35-40Hz Fb, and mobile) the THAM15 looks killer but simply doesn't go low enough. Many of my tunes I like to play have pretty prominent kicks in the high 30's which I'm not willing to give up with the 40Hz F10!
From what I recall, your OTH40C didn't go quite as low with this driver as it did using the FaitalPRO, also I believe you weren't keen on the assembly procedure!
After Playing on Hornresp for a while it seems there simply isn't a better driver with ANY significant advantage over this unit that will cater to what I'm looking for. EXCELLENT recommendation! (I should have just listened haha)
In fact I'm about to start SolidWorksing a model for this cab for once I have some extra disposable income!
A last question,
the very knowledgable Matthew Morgan J has commented that in practice he has found that Bass Reflex units don't quite live up to sims as he'd like and that he's had more luck/consistency with quarter wave style cabs of the same size.
The alternative to this reflex I've arrived at (80L internal, 24x20x18" external) is a Karlson aperture offset driver based cab. Are you aware of any advantages to that cab over a bass reflex???
Thank you! checking port velocity now
EDIT: YIKES, port velocity with 10mm Xmax was well into 30m/s. Using the same size port length divided into thirds with a doubling of CSA @ each end (only a 21 degree flare) It is down to 20.5 m/s with no 4th order Butterworth.
Sine, were your results favorable when simming 19m/s?
Last edited:
After experimenting with different ideas, it occurred to me how awesome it would be to have an optimization system like excel. Not that I have anywhere near enough skill to program that but minimizing port volume and port velocity for a given box size and tuning would be pretty cool.
MMJ had posted a figure showing the difference between port pressure and velocity at the beginning and end of the port.
Is there any disadvantage to using a straight 1 segment flare horn style port? (Linearly increasing CSA with a max at the outside end?
Sent from my iPhone using Tapatalk
MMJ had posted a figure showing the difference between port pressure and velocity at the beginning and end of the port.
Is there any disadvantage to using a straight 1 segment flare horn style port? (Linearly increasing CSA with a max at the outside end?
Sent from my iPhone using Tapatalk
What you are describing here is a back loaded horn (or BVR - big vent relex). It's different than a flared port, but calling it's characteristics "advantages" or "disadvantages" depends on your goals. A horn shaped port will have to be a lot longer and larger for the same tuning as a flared port, and for that reason will exhibit more sensitivity gain. Is that an advantage or a disadvantage? It depends on if you want loud or small. It's clearly different though.
You can check velocity at both ends of the port or horn flare, so it's easy to keep it under control.
An optimization tool isn't really required, Hornresp's sliders make iterations fast and easy, and it takes literally only a few seconds to check on velocity. An optimization tool would have to know all your goals to be effective - that can be tricky unless your goal set is very narrow and easily defined.
Last edited:
Chips:
THe faital pro 15 1030hp and the BC 15tbw100 are both superior to the dayton (but not for the dollar).
remember, it is only possible to check the particle velocity WITHOUT any filters in place.
Thus, while at "max spl" (voltage in + filter means you stay withing xmax), your "partical velocity" readings using hornresp will be higher than they would under real use (with a filter). I am fond of going a litttttle bit big with the rear chamber (causeing a slight peak at tuning) and then using a 12db/oct or 18 db/oct BW hipass. not only does this minimize the added group delay inherent with using any filter at all (higher orders, more delay), but it reduces power above the xover freq a little bit more than a higher order filter. this coupled with your trusty limiter will reduce the "simulated port noise".
I'm *very* happy with this bass reflex box. It absolutely bangs, and has been entertaining an average dancefloor of 100 people since its installation, with a bridged ancient crest amp on it. (think about 1kw into 8 ohms).
I recently paired it with my 2 OTH40c (not in the fashion shown in the pic, ,that was just a beauty shot) at the venue its installed in. took a little work with the delay, filter settings, and placement (ended up with about 40 ft between the 2 different designs, both of them wall loaded) for a party we hosted (Innamind Recording US tour featuring LAS and Kursk). bass was clean (not muddy as I feared it would be mixing TH with BR) and coverage was surprisingly even throughout the bassment venue (about 45ft x 75 ft, 12 ft ceilings)
THe faital pro 15 1030hp and the BC 15tbw100 are both superior to the dayton (but not for the dollar).
remember, it is only possible to check the particle velocity WITHOUT any filters in place.
Thus, while at "max spl" (voltage in + filter means you stay withing xmax), your "partical velocity" readings using hornresp will be higher than they would under real use (with a filter). I am fond of going a litttttle bit big with the rear chamber (causeing a slight peak at tuning) and then using a 12db/oct or 18 db/oct BW hipass. not only does this minimize the added group delay inherent with using any filter at all (higher orders, more delay), but it reduces power above the xover freq a little bit more than a higher order filter. this coupled with your trusty limiter will reduce the "simulated port noise".
I'm *very* happy with this bass reflex box. It absolutely bangs, and has been entertaining an average dancefloor of 100 people since its installation, with a bridged ancient crest amp on it. (think about 1kw into 8 ohms).
I recently paired it with my 2 OTH40c (not in the fashion shown in the pic, ,that was just a beauty shot) at the venue its installed in. took a little work with the delay, filter settings, and placement (ended up with about 40 ft between the 2 different designs, both of them wall loaded) for a party we hosted (Innamind Recording US tour featuring LAS and Kursk). bass was clean (not muddy as I feared it would be mixing TH with BR) and coverage was surprisingly even throughout the bassment venue (about 45ft x 75 ft, 12 ft ceilings)
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.