slot port design |
Post Reply | Page 12> |
Author | ||
snowflake
Old Croc Joined: 29 December 2004 Location: Bristol Status: Offline Points: 3118 |
Post Options
Thanks(0)
Posted: 10 August 2022 at 4:05pm |
|
Hi
I'm trying to design a port for a double 15" ported horn using B&C 15DS115. 190L volume 500cm2 port cross section and 25cm port length gives a tuning of about 40Hz and port speed of about 40m/s. So it needs to be flared to reduce the port noise.
cabinet is 57cm wide so one option is to have a slot port that is 57cm*8cm and flared with a large radius (~4cm) along the long edges at each end. other option is to make a wider (57cm*16cm) slot out of two parallel pieces of 18mm and then used curved ply in between these to create a flare in the other direction that would be 26.5cm*16cm at the narrowest point. anyone able to say which of these will have lower port noise? |
||
fatfreddiescat
Young Croc Joined: 15 October 2010 Location: N.E.Wales Status: Offline Points: 1081 |
Post Options
Thanks(0)
|
|
Probably not a lot in it between the two options I guess.
Circular or square port cross sections should have lower resistance which suggests the second option. The first option is the type favoured by L Acoustics and others, possible advantages are an increase in end correction due to two walls of the port extending the full depth of the cabinet, possibly allowing for the same tuning with a larger open area or a shorter port length, it also has a larger area of the port profiled which should help, plus better cooling for the drivers when the cab is laid on it's side. Is it possible to fit the ports as 'cassette' modules? Would allow for future retuning etc. Re port noise, I don't think that will be an issue with this cab as the overall output will be so high it will mask the port noise, the main issue I think will be the port choking and so transitioning to a sealed rear chamber design, this can be measured easily although bloomin loud with that cab
Edited by fatfreddiescat - 11 August 2022 at 11:09am |
||
Elliot Thompson
Old Croc Joined: 02 April 2004 Location: United States Status: Offline Points: 5172 |
Post Options
Thanks(0)
|
|
Slot ports generally offer less turbulence noise than circular ports. The issue is not the type of port but, the loudspeaker itself. When the loudspeaker is incapable of producing sound from the voice coil in which, the excursion is the main source of reproducing the sound at the given frequencies, you will encounter a lot of turbulence noise.
The key is finding the right loudspeaker (based on the TS Parameters) that will achieve the lower frequencies desired without relying heavily on excursion. Bear in mind Xmax has nothing to do with the measured TS Parameters. Focus on loudspeakers designed for Reflex & Horn cabinets and not solely Horns. Best Regards,
Edited by Elliot Thompson - 16 August 2022 at 3:38pm |
||
Elliot Thompson
|
||
snowflake
Old Croc Joined: 29 December 2004 Location: Bristol Status: Offline Points: 3118 |
Post Options
Thanks(0)
|
|
there aren't many very high xmax drivers to choose between. the 15SW115 has a higher qts but wants a much larger horn throat and rear chamber. In the same enclosure as I've designed for the DS the SW is limited to lower output 40Hz-50Hz by it's xmax |
||
Elliot Thompson
Old Croc Joined: 02 April 2004 Location: United States Status: Offline Points: 5172 |
Post Options
Thanks(0)
|
|
The problem is the B&C 15DS115 will never take advantage of the rear chamber to attain 50 Hz much less 40 Hz. It is designed to offer a lot of 60 Hz on up. That is what the TS Parameters is saying. You need a balance amongst the two. Disregard the horn for the moment and just scale the rear chamber in which, the driver will be sitting in with the port. In other words, scale a reflex cabinet with the desired port to determine if driver can achieve 50 Hz - 40 Hz @ -3dB. If your first -3dB point is around 80 Hz - 60 Hz, you have already loss half of your amplifier power trying to achieve 60 Hz. Matters will be worse if you are trying to reach to 40 Hz. The TS Parameters and the Free Air Graph speaks for itself. The speaker is -10 dB @ 60 Hz from a starting point of 95 Hz. With a driver offering a Qts of 0.17 and a VAS of 3.32 Cubic Feet.... Best Regards, Edited by Elliot Thompson - 17 August 2022 at 1:37am |
||
Elliot Thompson
|
||
toastyghost
The 10,000 Points Club Joined: 09 January 2007 Location: Manchester Status: Offline Points: 10920 |
Post Options
Thanks(0)
|
|
This is going to be the trade-off, unfortunately. The DS series of drivers was originally developed to provide a higher BL at a lower unit cost than the SW, and appears to have been made possible primarily due to tooling improvements that resulted from a large manufacturer’s order requirements. The SW might need more breathing room for the horn element, but they're very effective in even relatively small ported enclosures. Check the B&C 215-DCX reference design as an example; it has an unprocessed on-axis -3dB point of 50Hz from two 15SW115 in separate enclosures of approximately 120 litres each, with a slot port of 204 x 180 x 180 mm on either side of each driver. I'm not sure what your target overall cabinet size is, but you might be able to give a little more space to the horn with a similar arrangement. A horn throat with lower compression ratio will generally sound better at full chat, anyway. That might be useful - the horn's low pass filtering effect can mask the nastiest throat distortions, but they may instead be audible through the diaphragm and out via the port. |
||
snowflake
Old Croc Joined: 29 December 2004 Location: Bristol Status: Offline Points: 3118 |
Post Options
Thanks(0)
|
|
just noticed the 15DS115 spec sheet recommends a rear volume of 80L and tuning of 40Hz - which by coincidence is exactly what I arrived at to give suitable rear chamber for the horn and a vented chamber that can have a 40Hz tuned port that isn't ridiculously long.
The lossy Le model in hornresp adjusts the BL for the 15DS115 to 28 which implies a qts of 0.3 - doesn't that mean this driver is more suitable for vented than spec sheet qts suggests? the compression ratio on my horn is very high but the horn flare is 65Hz 0.7T with symmetrical throat loading and a rapid flare in the first 10cm. DS115 should be robust enough to deal with that. pushing things in terms of throat distortion at full ouput though, ~1W/cm2 @200Hz means ~4% air distortion. I would prefer lower but a lot of speakers are worse than that due to motor and suspension non-linearities. |
||
Elliot Thompson
Old Croc Joined: 02 April 2004 Location: United States Status: Offline Points: 5172 |
Post Options
Thanks(0)
|
|
It does not. This is due to the Le being based on 1 kHz on up. It does not play a factor from 100 Hz down. I have no idea how much this driver will cost you. In the States, the street price is nearly $700 for one driver. There are other options (Eighteen Sound, Faital Pro, RCF, BMS) that would take advantage of the rear chamber to achieve a better low frequency extension that would be a fraction of the cost. So there is no reason to limit yourself only to B&C. Reflex cabinets rely on VAS, fs and, Qts to achieve optimum results as it uses the rear wave of the driver to amplify sound (Loading in the internal chamber of the cabinet). Horn drivers follow the opposite principal using the front wave of the driver to amplify sound (Loading in the Horn). This is why a driver suited for both types of cabinets will give you the best of both worlds. Realistically, a medium between the two. Best Regards, |
||
Elliot Thompson
|
||
toastyghost
The 10,000 Points Club Joined: 09 January 2007 Location: Manchester Status: Offline Points: 10920 |
Post Options
Thanks(0)
|
|
A relatively unknown fella called W. Marshall Leach would strongly disagree with you: https://leachlegacy.ece.gatech.edu/papers/vcinduc.pdf There are further examples, plus a MathCAD script for the modelling method used in Hornresp ,from Plantefeve here - use a translator: http://jm.plantefeve.pagesperso-orange.fr/acou.html Le is typically reported on driver marketing spec sheets at 1 kHz, but it varies with frequency & is dependent on the excursion. Some driver manufacturers such as Scanspeak share the results of two models publicly, others require you to ask for the full .kdbx file from Klippel Analyzsr. L(x) and Li(x) are nonlinear parameters reported by a LSI3 test, which is based on these standards that Klippel contributed to: IEC 60268-5 Sound System Equipment, Part 5: Loudspeakers IEC 62458 Sound System Equipment – Electroacoustic Transducers - Measurement of Large Signal Parameters They have a lovely illustrated poster about nonlinearities in loudspeakers too, with a section for Le(x) and Le(i): https://www.klippel.de/fileadmin/_migrated/content_uploads/Klippel_Nonlinearity_Poster.pdf It's too big to screenshot and share here, but if you click one link on this post - make it this one ☝️ I think people sometimes put a bit too much weight on the Hornresp Lossy Le results - it is a nonlinear parameter in a linear model, after all - but it absolutely has an effect on Bl and more in the real world. Edited by toastyghost - 18 August 2022 at 8:56am |
||
Elliot Thompson
Old Croc Joined: 02 April 2004 Location: United States Status: Offline Points: 5172 |
Post Options
Thanks(0)
|
|
Bear in mind that I measure loudspeakers in which I have the capability to select the starting and ending points. You will not achieve an Le response from 100 Hz down. This is why people disregard Le for Bass applications.
The biggest problem is using a horn method topology on a reflex concept. It just creates confusion for those who have very little understanding on achieving optimum results designing a reflex box. Best Regards, Best Regards,
Edited by Elliot Thompson - 18 August 2022 at 12:15pm |
||
Elliot Thompson
|
||
toastyghost
The 10,000 Points Club Joined: 09 January 2007 Location: Manchester Status: Offline Points: 10920 |
Post Options
Thanks(0)
|
|
Oh, is this another post where you’re just going to ignore a whole pile of evidence showing that you’re wrong?
Cool; obviously you’re a much more established engineer of renown & contribution than Leach & Klippel |
||
snowflake
Old Croc Joined: 29 December 2004 Location: Bristol Status: Offline Points: 3118 |
Post Options
Thanks(0)
|
|
it's not really a reflex box though - I'm only getting the port output so the response of the rear chamber doesn't really matter like it does with a reflex design. The only thing that seems to affect port output level significantly is the motor strength - which implies low qts drivers are better! And the Sd affects the q of the port output - larger drivers giving a broader port output peak. the rolloff of the horn+chamber is mainly governed by the reactance annulling of the horn.
EDIT: higher motor strength also gives more group delay I've just realised that if the effective BL is reduced due to lossy Le then my reactance annulling calculations might be well off Maybe there is a driver with a strong motor and high xmax but lower Le... "I think people sometimes put a bit too much weight on the Hornresp Lossy Le results - it is a nonlinear parameter in a linear model, after all - but it absolutely has an effect on Bl and more in the real world." will bear this in mind Edited by snowflake - 18 August 2022 at 5:59pm |
||
Post Reply | Page 12> |
Tweet |
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |