Copyright © 1998
Cadman Enterprises Ltd
Welcome to the Graham Slee Audio Products Owners Forum ![]() Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here) This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd |
High Fidelity Speaker Design |
Post Reply
|
Page 123 7> |
| Author | |
Ash
Senior Member
Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4360 |
Post Options
Thanks(0)
Quote Reply
Topic: High Fidelity Speaker DesignPosted: 08 Dec 2015 at 11:07pm |
|
So now I need to make some decisions about speakers. I have already bought two pairs of Mark Audio Alpair 12PW and two pairs of Mark Audio Alpair 7P, so eight speaker drivers in total. The point of this is for some experimentation; I could try a 4.0 setup with the four 12PW/7P pairings or try a stereo vertical line array with two pairings on left and right. I could use the extra drivers to experiment with different enclosure/baffle materials/sizes/shapes/designs giving me the chance to compare two designs side-by-side with the same driver types. Will be mounting the 12PW and 7P as close together as physically possible, as recommended by Mark Fenlon. The closer they are, the more they will imitate a point source in the nearfield/midfield.
The traces below show why I chose to use the 12PW and 7P combination instead of using a lone pair of 12P. The combo would give better results on open baffle, I think; deeper bass extension and weight, a more balanced, even frequency response with a more consistent treble response. ![]() Not sure about crossovers though. I have no idea how to design passive crossovers using electronic components in simple circuits. Not sure where to start here... Want to keep it simple though, with 12PW and 7P volumes well matched.
|
|
![]() |
|
Ash
Senior Member
Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4360 |
Post Options
Thanks(0)
Quote Reply
Posted: 09 Dec 2015 at 8:28pm |
If the interaction between adjacent 12PW and 7P drivers were to be entirely constructive wave interference, the resultant superposition waveform would look pretty desirable. Depends on their phase relationship, I guess. Would not want a very complex crossover network for a driver pair that is this synergistic. Two of these pairings per channel would move twice as much air but vertical dispersion would be reduced in a vertical line array.
|
|
![]() |
|
Drewan77
Senior Member
Joined: 25 May 2013 Location: Chester,England Status: Offline Points: 1544 |
Post Options
Thanks(0)
Quote Reply
Posted: 10 Dec 2015 at 11:51am |
|
It sounds like some sort of digital crossover may be the best solution for you Ash. I know from experience that differing crossover points and slopes can make a big difference to both phase and timing. It is fascinating and informative to experiment and learn - the end result in my case has been wholly positive.
Maybe something like mini DSP is worth researching because you can tweak and compare numerous setups without the niggling doubt that the passive components you have chosen could be 'improved' upon.
|
|
|
Older than I once was, younger than I'll be
............................. Andrew |
|
![]() |
|
Ash
Senior Member
Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4360 |
Post Options
Thanks(0)
Quote Reply
Posted: 10 Dec 2015 at 1:22pm |
|
I'll probably just use my K1000 as a listening reference to get my speaker playback as close to it as possible. I don't want to get too carried away with correction or it will just turn into another expense/obsession.
My goal now is to find a good balance of stereo image coherence, sound radiating area, minimized room interaction, frequency neutrality and maximum frequency extension, especially in the low bass. I'm certainly enjoy speakers more than headphones now that I have fantastic speaker drivers and no box enclosure characterizing their performance. I'll never go back to a box design for serious reproduction.
Edited by Ash - 10 Dec 2015 at 3:15pm |
|
![]() |
|
Fatmangolf
Moderator Group
Joined: 23 Dec 2009 Location: Middlesbrough Status: Offline Points: 9695 |
Post Options
Thanks(0)
Quote Reply
Posted: 11 Dec 2015 at 8:50pm |
|
With two speakers putting the signal out on each side you may get 'lobing' where the sound varies as you move up and down, and possibly side to side.
|
|
|
Jon
Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC. |
|
![]() |
|
Ash
Senior Member
Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4360 |
Post Options
Thanks(0)
Quote Reply
Posted: 21 Dec 2015 at 10:00pm |
|
Andrew, I have a favour. Would you be able to reveal the brands/model numbers of the drivers used in your speaker system. May I also see some images of the drivers/baffles as you have them set up in the room so I can make some careful observations. I want to get an idea of the size and shapes of your hardware. PM or email is also fine if you don't wish to share it right here. You're much closer to speaker mastery so I'd like to stand on the shoulders of giants again if that's okay.
Edited by Ash - 21 Dec 2015 at 10:03pm |
|
![]() |
|
Drewan77
Senior Member
Joined: 25 May 2013 Location: Chester,England Status: Offline Points: 1544 |
Post Options
Thanks(0)
Quote Reply
Posted: 22 Dec 2015 at 7:29am |
|
Hi Ash
I’m sitting in an airport lounge waiting for a flight back to the UK & that’s no problem but pictures will follow when I’m back… I have listed everything below for you but I’m not sure how much use this will be because the speakers are controlled by a DEQX processor. For completeness, I have sketched out every aspect of my speaker setup even though only part (or none!) of this may be relevant to you. Before deciding on these, I looked into the pros & cons of open baffle speakers compared to more conventional designs: Pros: They are easier to construct & there is no “cabinet coloration”. Transient speed & ‘attack’ can be spectacular, as can imaging & ‘presence’ in the room. Cons: An extremely large baffle is required for low frequencies & ‘room coloration’ of the sound can then become problematic. On that basis, I designed around my existing subwoofers which were already well integrated (digitally corrected) into the music room & I was happy with the depth and speed they provide. In my speakers, there are no crossovers and without digital phase and time alignment, these Open Baffles are basically unlistenable. During the prototyping and development stage, DEQX allowed me to measurably & methodically tune and tweak, including on-the-fly whilst listening to music. Without that, the process may have been theoretical and random with more erratic results. I am NOT a speaker designer and although it is kind of you to say, neither am I a ‘master or a giant’ ! - it is my extensive experience with DEQX that had led me down this particular path.The OB speakers handle frequencies from 100z upwards, with the subwoofers below that. All crossovers/phase & time alignment are digital & speakers/subs were firstly measured and corrected ‘anechoically’ outdoors and then repeated in-room at the listening position. The speakers and closest sub are then manually time aligned to the furthest & ’slowest’ sub at the listening chair using step response measurements. Subs: - Miller & Kreisel MX200 (2 x 12” drivers). Deepest, tight/controlled bass, with good definition. 0-100hz, crossover @ 72dB slope - B&W PV1D (2 x 8” drivers, high pass filtered above 20hz). Fastest bass 20-100hz, crossover @ 72dB slope OBs (per speaker): - 4 x HiVi M5a 5” Aluminium/Magnesium mid bass drivers (no longer available, replaced by Hi-Vi M5N) 100-3100hz, crossover @ 72dB slope - 1 x Fountek Neo CD3.0 Ribbon Tweeter. 1 x Jantzen Silver z-cap 3100hz>, crossover @ 300dB slope Baffle diffraction around the tweeter is managed with a two step wool surround, various shapes/sizes were tested and I have taken care to minimise lobing issues, given the extreme brick-wall crossover being used here (I also experimented with reverse tweeter phasing because this is sometimes used with higher order crossovers. However the DEQX algorithms manage phasing within speaker sets so this proved unnecessary). Driver cabling: - GSP Spatia to separate mid & treble binding posts. To maintain correct impedance, the four midrange drivers are wired in series: the + binding post from the + channel of the amp is wired to the + of the first speaker, & the - of that speaker is connected to the + of the next etc. Baffles for the OBs: - Originally prototyped in various thicknesses of MDF, adapting until I was able to arrive at a sound and measured frequency response that I was happy with. The final product uses a dense/acoustically dead material - 18mm CNC cut zebrano bamboo (I commissioned ’Timboo’ in Halnaker West Sussex for this). The front baffle has one vertical edge and one slightly tapering inwards, with a large radius to the top corners. The baffle has a side edge which gradually widens as it approaches the base of the speaker. This constrains (open) rear diffraction from bleeding and interfering with direct sound from the front baffle. The M&K sub has also been externally clad in the same material for aesthetic reasons which may add to the already immense rigidity. The finished appearance of speakers and sub, to my eyes is superb but the effect is rather wasted in my ‘man cave’ music room. The M5a drivers are only just able to manage the 100hz crossover point without a cabinet enclosure for assistance - the fact that there are eight of these compensates to a large degree but I needed them to manage the mid treble frequencies so the larger HiVi driver options were ruled out. In the final configuration of the system, I also use a small amount of digital eq on frequencies below 250hz but no more than +/- 1.75dB which removes a couple of nasty room nodes at the listening chair. The use of two subwoofers with different room placements can assist with lessening these anyway. Final measurements at the chair confirmed a flat response +/- 2dB from 16hz to +20khz but I subsequently tweaked that a little to provide exactly the sound I was after. In conclusion, by using OBs with Alu/Mag drivers and very high order crossovers, all the amps and driver sets are operating in a less stressed or coloured fashion. The level of realism and ‘attack’ this achieves is always satisfying and quite frequently startling. A more knowledgeable acquaintance who has listened to my system has described the end result as akin to an Airbus aircraft or modern fighter aircraft - basically unstable and unable to fly without fly-by-wire assistance but with it, able to achieve great or even spectacular results. My speakers, unassisted would never even get off the ground (let's hope I do in a short time from now, my gate is called !) Edited by Drewan77 - 22 Dec 2015 at 7:32am |
|
|
Older than I once was, younger than I'll be
............................. Andrew |
|
![]() |
|
Post Reply
|
Page 123 7> |
|
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 |