In my view this box gives the best compromise between f3 and linearity of response. I consider this the optimal box.
Author Archive
Vifa_TC14WG-49-08_optimum box
Friday, February 13th, 2009Dons problem.
Wednesday, February 11th, 2009Don (Moron) posted on the Audioholics forum about adding a sub woofer to his speakers.
His speaker system was a reflex loaded MTM system, using dual Focal 7W4411 woofers and Focal axiom TLR tweeter in MTM configuration. Don had built the speakers which had been professionally designed and the crossovers supplied by the designer.
Don put this post on the Audioholics Subwoofer forum on 12/19/2008. Don’s main complaint about his speakers was severe lack of bass. However in further correspondence via Email, it became there were other issues. Don complained, that the sound was generally poor and that he did not listen to music much anymore. Speech clarity was so poor, and he had had to resort to using an old cheap RCA speaker as center channel to make speech intelligible.
Obviously with drivers of that quality something had to be seriously amiss. Attention was paid to the bass tuning. On information supplied by Don it seemed that the bass alignment was optimal. The F3 was an excellent 41 Hz, so the speakers should not be bass shy. Don did however volunteer that all internal surfaces were covered with 1″ Black Hole sound absorbing material. I advised him to reduce coverage to 50% of the internal surface, but to make sure the rear wall remained covered.
From Don’s plans it seemed the enclosure was well designed and braced. Don reassured me about the constructional integrity of the speakers.
So suspicion naturally turned to the crossovers. These were Don�s crossovers. They are fourth order Linkwitz/Riley crossing over at 2.4 kHz There is a frequency response chasm from 200 Hz to 4 kHz, centered on 1.5 kHz. It is easy to see why Don was unhappy with his speakers. The major part of the dip is across the speech discrimination band from 400 Hz to 2 kHz. So it is easy to understand how speech intelligibility was compromised. The tweeter level is two high, so the speaker is poorly balanced, leading to the perception of lack of bass.
So new crossovers were designed. A second order impedance compensated crossover was designed crossing over at 3 kHz. The response in the crossover region is +/- 2db. when combined with the driver roll offs, this constructs a composite fourth order crossover. The tweeter is 90 degrees ahead of the woofers at crossover, so tweeter polarity is not reversed. The crossover is impedance compensated, with impedance of 4 to 5 ohms except for a rise to 16 ohm at crossover.
The crossovers were constructed, and Don had a friend help with installation.
Don has been very pleased with the result. We have confirmed Fb is very close to 35 Hz. With Dons Radio Shack spl meter, the speakers would apear to be 4 Hz down at 40 Hz. There is useful audible output to 30 Hz.
Since this modification Don has used his speaker more, and the old RCA center channel speaker is gone. Speech clarity is reported as excellent, with a good stable center image. Don now finds the bass adequate and no longer feels a need for a sub.
This is Don’s impression of his “new speakers,” in his own words.
“I can barely believe the sound of my “new” stereo system. The crossovers you designed and made, and your suggestion of removing about half of the insulation, has turned a rinky-dink system into a concert of beautiful music. I can now easily understand voices – which I couldn’t before – and the music is beautiful, AND, you saved me hundreds of dollars, because the bass is so much deeper and clearer that I don’t feel the need of a subwoofer. I use the stereo soley for music and TV.
I paid a considerable amount of money to have a “professional” chief engineer of an audio company design the cabinets and crossovers for my componts, and the sound was pretty lousy: Couldn’t understand voices and pretty much a jumble of music. Seemed like the tweeters and woofers were in conflict.
As I admitted, I’m a moron regarding the topic of stereos, so I have no idea what the terms you used mean, or why your crossovers have made such a difference, but the sound that my speakers now reproduce is stunning. I had no idea a home stereo system could produce such beautiful music.”
Don’s post is interesting and shows that things are not always what they seem. If Don were so inclined his system would now likely have improved performance with a sub, were as previously a sub would have provided little benefit.
It is gratifying when a long range call for help can have such as satisfying outcome.
Adcom GFA-555 II power amp circuit and service manual.
Tuesday, February 3rd, 2009Here is the Circuit.
The Audax AP170Z0 Aerogel mid/bass driver.
Tuesday, January 20th, 2009The Audax Aerogel drivers are excellent mid bass woofers. However they have a big peak first break up mode, at 3.8 kHz of 9db in magnitude. This is associated with a rise in response at 1kHz. This very large peak requires notching out. However with careful crossover design, excellent performance is obtainable from this driver. At first sight this driver does not look promising, but with notch filter and tweaking the Qs of the filters superior performance is possible. A speaker with good bass performance and excellent smooth detailed mid range is posible, in a very cost effective speaker.
Last fall while in England, my father asked me to put together speakers for their sun room. There are always a collection of drivers around my parent’s house. I used the Audax AP170Z0 and the Scanspeak D2905/95 tweeters. I used these because they were handy. A superior speaker resulted from this design.
Here is the woofer alignment, for an optimal 1 Cu. ft. box.
Here is the alignment for a smaller 0.43 cu.ft. box
This is the Crossover Circuit.
I have modified the circuit for the Vifa D27TG 05/06. This is available from parts Express.
The driver has an F3 of 46Hz in a 1 cu.ft box. This is good performance, although the driver is displacement limited from 50 to 110 Hz. In this range spl is limited to 100db. above that spl. of 107db is possible.
There are two impedance humps at 30 and 75 Hz, with the minimal impedance at F3, so tuning is optimal.
A vent flared at both ends, 3″ in diameter, six inches long, slows air velocity to 18 m/sec.
The crossover is electrical second order, however it combines with the driver slopes to make a composite fourth order filter. Both tweeters are down at least 24 db at resonance (Fs).
The woofer impedance is equalized by Req and Ce and the break up mode peak dealt with by Rm, Lm and Cm.
16 SWG air core inductors should be used for L1 and L2. A good quality iron cored inductor should be used for Lm. Good quality polypropylene caps should be used, such as Solen caps. Resistors should be wire wound 15 watt.
The phase response of the crossover is unusually good. The mid band response is excellent, including across the crossover region. Some degree of diffraction compensation has been achieved, about 3db. This is usually what is achievable without adding a second driver. Room gain should in any event do the rest, unless the speaker is a very long way away from room boundaries.
The impedance is 5.5 ohm though most of the range with a peak to 24 ohms in the crossover range. These speakers should be regarded as 6 ohm speakers.
Allowing for component insertion loss, notching the woofer peak, and providing for some diffraction compensation, the sensitivity of the speakers is 86db 2.83 volts at one meter. This is very typical for a two way using drivers of this size.
This makes a very nice speaker.
I have also worked out a crossover that has minimal changes for the SEAS 27TDC. This tweeter is about $10 more, but well worth it.
If you want to do a two and a half way tower, then here is the optimal box alignment with an F3 of 46 Hz.
I have designed a crossover using the Scanspeak tweeter. This should be a very good speaker and well worth the cost of the Scanspeak tweeter. The diffraction compensation starts at the optimal point of 600 Hz, and provides +6db of compensation for the diffraction loss by 175 Hz. This speaker should have good robust tenor register.
Here is the 2.5 way crossover using the SEAS 27 TDC tweeter. The phase response is excellent.
I’m adding and MTM design for this driver with a choice of two tweeters. This requires a third order crossover at 1.8kHz.
Here is the box alignment. Allowing for the volume of driver crossover and braces,the final volume Vt should be around 1.3 cu.ft. The port should be 4″diameter X 10″long flared at both ends. You can also use the two cu. ft. box from the two and a half way tower, above, for extended bass.
The crossover circuits for the Vifa D27TG 05/06 and the SEAS D27TDC are the same.  The SEAS D27TDC is excellent value for money, and in my view worth the slight extra cost.
This is the response with the Vifa D27TG 05/06
This is the response for the SEAS D27TDC
Any of these designs should make an excellent set of speakers. The crossovers for the MTM set up should have optimal lobing for this configuration. Vent velocity of these MTMs is an excellent 18m/sec.
Allowing for insertion loss, and notching the woofer peak, overall sensitivity of these MTM speakers will be 90 db 2.83 volts 1 meter and 87 db 1 watt 1 meter.
The impedance is very uniform at 4 to ohms throughout, except for a slight peak t0 15 ohms at crossover. This design is impedance compensated and should be a pretty easy load for any amp or receiver able to drive a four ohm load.
50% of the internal surfaces of these enclosures should be covered with a Rockwool like product or Black Hole. The internal surface of the back of the enclosures should be fully covered.
Any of these speakers if properly constructed should give results equivalent to speakers in the 1 to 2 K range per pair.
Here is a parts list and sources, for the two and a half way with the Scanspeak D29. The component values that change when using the SEAS 27 TDC tweeter are in the brackets, and those alternate values that change when using that tweeter can be found by clicking within the brackets. Note the specified values and source values in some cases differ slightly. However there is no significant change in performance.
C1Â Â Two required
C2Â Â Omitted
C3  Two required. (version with SEAS 27 TDC tweeter)
C4Â Â Two required
C5Â Â Two required
Ce  Four required
Cm Four required
L1 Â Two required
L2Â Â Omitted
L3Â Â Two required (version with SEAS 27TDC tweeter)
L4Â Â Two required
L5Â Â Two required
Lm Four required
Rp1 Two required (version with SEAS 27 TDC tweeter)
Rp2 Two required (version with SEAS 27TDC tweeter)
Req Four required
Rm Four required
Circuit Board Two required
Terminal cups Two required
Ports 4″ flared Two required
The total cost of the above list is $323.94
Crossover parts for the MTM. Note that the parts for the Vifa and SEAS tweeters are same.
The cost of this MTM crossover and other parts is $130, so for a pair, $260, for mains and center $390.
This is the parts list for the crossover for the 1 cu. ft and 0.43 cu.ft book shelves. This is for the Vifa and SEAS tweeters. The different values for the Seas tweeter are in brackets.
L1 (SEAS 27TDC)
The cost is around $100 each. So two mains or surrounds $200. Two surrounds and rear backs $400.
So the total parts cost for a system using 2.5 way mains, MTM center and four bookshelves is $1228 using the Vifa tweeter and $1298 using the SEAS.
Using MTM mains and centers the cost is $1164 using the Vifa tweeter and $1234 using the SEAS tweeter.
So the most expensive system averages $185 per speaker for all parts, and the cheapest $166 per speaker.
So building with MDF, you should be able to build two towers, a center and four surrounds for around the $1500 price point, which is outstanding value.
Dayton RS 150S-8 and Beston RT002A crossover and box alignment
Sunday, January 11th, 2009This combination works best with a fourth order Linkwitz Riley crossover.
Here is the vented woofer box alignment.
MTM CONFIGURATION
Here is a vented box alignment for two Dayton RS 150S-8s.
Note that it requires a 3″diameter vent flared at both ends 7.25″ long. This gives a max air vent velocity of 20 m/sec.
The max. spl is 109db, however output is limited by driver excursion from 50 to 100 Hz, reaching an spl of 103db at 65 Hz.
The cabinet should be well braced. The volume displaced by the drivers, braces and crossover should be added to Vb to come up with Vt. A rule of thumb estimate is Vb + 15 to 20%.
Place the drivers as close together as possible.
Half the internal surface of the enclosure should be covered by acoustic absorbing material. The back wall should definitely be covered. Rockwool type products are best.
The vent can be placed on the front or back. Keep the vent 4″ away form any wall boundaries and drivers.
Here is the crossover for two Dayton RS 150S-8s in MTM configuration with the Beston RT002A.
If you want to notch the 15 kHz peak of the tweeter, use this parallel notch filter in series with the tweeter.
C = 2.002 µF
L =Â 0.05 mH
R =Â 7.95 ohms
Note: R should be adjusted by sub-
tracting from it the DCR of the coil.
Put the above components in parallel and out the parallel network in series with the tweeter. At the high frequency range of the peak centered at 15 kHz, I doubt the sonic effects of this peak will be very noticeable. I would build the speaker without this notch filter first and see how it sounds.
The crossover is third order for this MTM configuration, which will give the correct lobing tilt for this MTM configuration.
The Thor line with the SEAS Prestige CA18RNX and 27DTC Drivers
Friday, December 12th, 2008The cost of the SEAS EXEL drivers is too great for some to bear in these economic times. I have explored the possibility of using the above drivers as a less costly alternative.
The Thiel/Small parameter are an acceptable match. Fs is 4 Hz higher, VAS is virtually identical, and Qts is slightly lower. I consider the mismatch however to be acceptable. The line as built is slightly long. However, I think the line will function well.
The SEAS CA18RNX bass/mids are $75 from Madisound, so that would be $300 for the pair.
The SEAS 27DTC Tweeters are 39.91 from Madisound, around $80 for the pair, bringing the driver coast to an acceptable $380.
Here is the pdf for the woofers.
Here is the pdf for the tweeters.
I have designed a crossover for these drivers.
This crossover is elegant and simple. Although the pdf states that these are second order Linkwitz/Riley filters, they are actually highly customized.
The low pass filter is actually first order from 1KHz to 5 KHz. At5 Hz, the composite electrical and acoustic roll off becomes fourth order. The woofers are down 18db an octave above the crossover frequency of 2.8 KHz. Req and Ce provide impedance compensation to the woofers.
The high pass filter is also first order 8 KHz to 1.8 KHz and transition to fourth order after that. The program does not let me show R1, which changes the Q of the filter. Note that C2 shows a resistance of 2.005 Ohms. Now no cap will have that resistance. So a 10 watt 2 ohm resistor needs to be placed in series with C2.
This is an easy to build seven element crossover. You can see that the filter is first order in the crossover region, as the phase angle is 90 degrees, or a quarter cycle in the crossover region. Also this crossover puts the tweeter 48 db down at its resonance of 550 Hz. So there should be no tweeter harshness. A standard first order filter would have put the tweeter down only 14db, at resonance. Since this is an odd order filter at crossover, it will fulfill the lobing requirements of the d’Apollito configuration.
There is also around 3db of diffraction compensation with this crossover.
Since the phase and time response of this filter is well above average, transient response should be superior.
Good parts should be used for the crossovers. Air cored inductors and polypropylene caps, such as Solen caps should be used.
I think where funds are a problem, these drivers and crossover should be given serious consideration.
I will work on a TL for these specific woofers. I think this design carries the promise of unbeatable value for money. The line will end up being a little smaller than the Thor line. Power handling will be less, but a respectable 160 watts.
Report from the 2008 SOTU event: – Orlando
Thursday, November 13th, 2008I attended the SOTU event in Orlando last month. It was an opportunity to listen to some of the equipment talked about on the Audioholics forums and see equipment off the drawing board. Apart from my Rotel, pre pro, my Direct TV HD DVR and one Sony Bravia TV, I don’t think any of the rest of my extensive equipment list is talked about on the forums.
It was a chance to hear what is “on offer” so to speak, and make some acquaintances. Also I attended as many of the didactic sessions as I could, but I gave up the lions share of the last day, to try and carefully evaluate the sound in the demo rooms. By that time others had already had their chance to to hear the movies, booms crashes and explosions.
I wanted to hear the quality of the sound on offer. In all of the rooms, I used mainly for evaluation a disc of keyboards suites by Jean-Philippe Rameau. (Hyperion CDA 67597) This disc is in a large series of superbly engineered discs by Hyperion devoted to the outstanding Canadian pianist Angela Hewitt.  For this disc she chose an Italian Fazioli piano. These pianos are beautifully voiced and balanced, with a very clean delicate articulation. Recordings like this absolutely lay a speaker naked.
In a couple of rooms I played the opening of Sir Edward Elgar’s oratorio The Apostles. The London Philharmonic Orchestra, choir, Choir of Downe House School and six soloists are conducted by Sir Adrian Boult. This recording (EMI CDS 7 49742 2) is widely considered one of the finest recordings ever made. It comes from the pre digital era and was superbly recorded in the Kingsway Hall before modern acoustic engineers ruined it. Now this is a recording I’m loath to use for an audition, as only very exceptional reproducers can do any justice what so ever to it. So I think it generally unfair practice, unless I hear a reproducer that might stand a chance, to offer this recording up. Two were of that caliber, but unfortunately I only got a chance to play it on one. It was played on another at the request of the demonstrator. This recoding has a wondrous sense of space and at the same time inner detail. It is beautifully balanced. There is tremendous deep bass detail, which many otherwise fine reproducers just leave out. Unless you have heard these disc on a reproducer with a very extended and accurate last octave, it is impossible to know what other reproducers leave out.
The Pioneer 3-EX: –  This is a new speaker from Pioneer. This is in a line of speakers build round a mid/tweeter coaxial driver. This is a scaled back version of the $18,000/pair 1-Ex. These speakers use coaxial mid/tweeter units. The woofers have woven carbon fiber cones. These speakers are therefore three way designs, with crossover frequencies sensibly chosen at 400 Hz and 2KHz. They claim to be time aligned. Nominal impedance is stated to be 6 Ohms, however with two woofers in parallel below 400 Hz, they have to be regarded a essentially four ohm speakers. I can not ascertain that these speakers are actually yet for sale, or confirm a price. But $6000 per pair was quoted.
Unfortunately I only got to play the Rameau disc on these speakers, but it was an extended listen. The sub was kept operative.
I have to say I was extremely impressed by these speakers. They were very well voiced and balanced. The image was rock solid and the piano placed in an entirely believable acoustic environment. The detail and articulation of the Fazioli piano was captured nigh on perfectly The only minor blemish, I was aware of a very slight bloating of the lower strings, by I think the port. However I’m used to the non resonant output from TL ports. I did notice in their movie presentation that the speech clarity was by far and away the best of the show. Speech was still highly intelligible, even in very noisy parts of movies. I think this is largely due to the coaxial driver. It goes along with Chris Seymour’s presentation of his measurements on center channel speakers.
Chris Walker of Pioneer gave a figure of $6000/pair. If that is so, I rate them very good value for money. I did not get a chance to play the Elgar, but I would like to have. I would just love to get my hands on at least a couple of those coaxial drivers. I noted the mid range cone was very shallow, and made just about a perfect waveguide for the tweeter.
I’m not usually impressed with far Eastern Speakers, but these really caught my attention. I though they produced the best sound of the show. The quality of construction and finish were superb. I understand the designer was formerly associated with KEF.
Of interest is the fact that the speakers were powered by their SC-07 receiver. This is their mid range receiver using class D amps employing B & O ICE amp technology. I asked about their suitability for powering four ohm loads, especially given the fact that although the speakers were nominally 6 ohm, given the fact there are two drivers in parallel below 400Hz, they had to be four ohm below 400 Hz, except for the dual bass reflex impedance peaks. They stated that for most speakers of four ohms they will be fine but refuse to put it in print. I asked what they would do for warranty in the event a customer had a problem.  They said they had extensive data on the loads offered by a lot of loudspeakers. I have to say I’m suspicious of this assertion. However they stated that if a speaker had an impedance curve they deemed unsuitable it could void the warranty. I could get no commitment that any list of unsuitable speakers would be forthcoming. They did however state that their flagship receiver the SC-09TX was unconditionally stable into all loads.
The Snell D7 : – This is a two and a half way tower with a switchable rear firing driver. It is designed round modest SEAS Prestige 6.5cm paper coned woofers and a SEAS prestige tweeter. I have modeled these units previously, and would seem to be able to provide a level of performance beyond what their price point would suggest. I was told that these speakers retail for $4000 each, however I see they can be obtained for just under $2,200 on the Internet. The speakers were auditioned with the sub.
I first auditioned these speakers with the Rameau disc. They gave a very creditable performance. Articulation and sense of space ere above average. The lower mid and upper bass were a little more colored tan the Pioneers. However the tonal balance was pretty accurate, with superior articulation.
I tried the Elgar disc on these speakers. Perspective, balance and voices were reproduced well. There as fairly good bloom on the brass. The violins, had a reproduced quality though. This recording has very natural violin sound, on these speakers the violins were a little wiry, and lacked softness.  The deep bass was the biggest difference between these speakers and my rig. The deep effortless natural foundation was lacking. The sound of the softly played tymps lacked depth of tone. The deep organ pedals made only a fraction of the effect they should have. However having said all that, the performance as very listenable, and had a good sense of space and perspective. For speakers in this price range they acquitted themselves well in what is a huge mountain to climb for any reproducer.
I could not help but notice how affected the staff were by Elgar’s music. It seemed to affect them deeply. Elgar was a master at conveying feeling and emotion. So often on hearing his music, you say to yourself, “yes I’ve felt like that.” Elgar is known only to most Americans for the march from his Pomp and Circumstance March No. 1 played at graduations. That is a pity.
The HSU HB-1H: – These speakers at $179 each are absolute gems, and outstanding value for money. They were being used, five of them with an HSU sub, to demonstrate the Sherwood R-972 receiver. This receiver contains the French Trinnov optimizer. This obviously is in competition with the Audyssey system. This system uses a cluster of microphones for set up. I have to say this system was able to achieve some excellent perspectives. Also you can set it to EQ below 300 Hz only, which I think is a big advantage. I noted on both systems the effects of EQ above the bass area were not entirely satisfactory. The system was ably demonstrated by Jeffrey Hipps, senior vice president.
This system with its five identical speakers, was able to locate Angela Hewitt at her Fazioli piano just about anywhere in the room. She was moved over to the right corner with a commendably realistic soundstage.
The sound of the piano was articulate and detailed and the bass better integrated with the sub than any other system.  The lower octaves of the piano were reproduced very effectively. These speakers use a horn loaded tweeter to help achieve a superior sensitivity 92 db/1meter/2.83 volts. Horn aberration was remarkably slight, with a trace of brittleness to the piano sound. The speakers had a good tenor range, indicating good step response compensation, even though there is only one woofer. The impedance does drop to four ohms though, to achieve this, again indicatiog there is no free lunch.
At Jeffrey’s insistence we played the Elgar disc. The system was able to achieve excellent perspective. The horn artifact marred the choral sound giving the chorus an “awe” character to their sound. This wars especially notable in the choral peaks. Bass was well integrated, but despite the excellent HSU subs, could not achieve the depth and authority of my system.
For speakers, in this size and price range I thought they were outstanding performers, and remarkable value for money. Unfortunately horns are renowned for exciting higher order modes, and diffraction problems in the waveguide which become more pronounced at higher volume.
The SVS MTS-01 and Audyssey: – In this room Kris Kyriakakis from USC and Audyssey was using this SVS set up to demonstrate Audyssey. I was able to spend some extended time with Kris. I was convinced that the system could improve the lower octaves, however I as not impressed with what happened above that, and Kris had to do some manual over rides to reproduce the piano to my satisfaction.
The sound of the SVS system was very pleasant. I thought the bottom end still a little embellished, despite Audyssey. Perspective and detail were not quite of the standard of the previous system mentioned. However I have to stress listening to the performance as very pleasant with nothing overtly standing out to overtly irritate. At under $1000 dollars each I consider they represent good value for money.
Emotiva: – Dan Laufman and general manager Lonnie Vaughn put on a good show in the Emotiva room. I got to spend significant time with Lonnie. I appreciate his generosity with his time. There was a a mouth watering amount of electronics on display. The hood was open so one could see the standard of construction. Emotiva’s boards, still are stuffed with through wire components, rather than robotically placed surface mount types. Boards were not overly crammed, and components were identified on the boards. I would think any competent service tech would have an easier time than usual for boards of modern manufacture isolating and repairing a fault.
There has been talk on these forums of reliability issues with Emotiva electronics of late, so I asked Lonnie about it directly. He told me their failure rate as in the neighborhood of 1%. That is very good for solid state electronics. I have no reason to disbelieve his statement. He was not someone who I would have thought would have glossed over a problem. The trouble is the 1% of customers with a problem are inclined to vent on the forum. Like the vast majority of manufacturers they do not make their semiconductors and other components. Small levels of contaminants in the manufacture of semiconductors, have always been a problem. This I doubt will ever be entirely solved. Problems caused by these impurities will show up usually in the first few months of use.
Now to the speakers. The movie demo was impressive, however I find it impossible to evaluate and know the worth of speakers on a diet of movies. So Lonnie put the Rameau disc on the ERT-8.3 towers at the back, driven form their CD player.
First auditioning was extremely disconcerting, with no coherent image what ever. So much so, that I had to ask Lonnie to check the phasing of the speakers. Sure enough they were out of phase! I don’t want to make an issue of this, we have ALL done it. After correcting the phase, things considerably improved. However I have to report that I found this speaker had shortcomings. The image of the piano was still not well defined. The speaker was not bright and had good tonal balance. However, it had a slightly muddled quality to it. These speakers did not create for me a believable acoustic space. There was significant coloration in the lower middle and upper bass. At the asking price under $800 a piece, they can not be called a disaster or even poor value for money. However Lonnie’s laudable aim is to make a speaker excellent value for money, and not a speaker just good for the money.
I discussed my feelings at length about these speakers. I suspect the problems are caused, by the fact that there are two mid range drivers, the fact that the low pass/bandpass crossover is 300 Hz, into the problem range for passive crossovers and some degree of cabinet resonance. I felt that the lower crossover was particularly problematic. The reason for using two mids, was to get sufficient spl to meet THX specs for spl.
Unfortunately three ways are hard to execute, especially on a tight budget. In that price range, I would think they would be better off with a well executed two and and half way.
Another attendee later told me their bookshelf speaker was a real winner, and so I later returned and asked Lonnie to please wire them up, which he kindly did. Unfortunately it immediately became apparent that the right speaker had gap rub. Dan immediately got the blame for overpowering them. I suspect the woofer had a dropped a voice coil turn. However on trying to listen through the buzz I think they may well have a winner here.
On another note, the Emotiva CD player produced faint clicks on the Rameau disc, not evident on other players and my two players. I’m just mentioning this, and don’t want to make a mountain out of it. It is a fact of life that some discs will have problems on some players and not others. That is why I always keep at least two players hooked up. Simon Perry the late founder of Hyperion used to say, that the ability of a player to play a wide range of discs, was in inverse proportion to the cost of the player!
I have to say I had an interesting time in the Emotiva room.
EMP : – EMP is a division of rbh. They had some very attractive speakers on show. Hooked up were their prototype EF 70Ts EF70Cs and EF70, mains, center, surround/presence respectively.
The EF 70T towers have a tweeter and six nice looking 5.25 inch aluminum coned drivers. Because of my long experience with the Jordan Watts driver, I’m very interested and partial to small aluminum coned drivers. So I wanted to like these speakers. I regret to have report though that in my view they are a disaster and not at all ready for prime time.
Angela Hewitt’s Fazoili piano was unrecognizable. There was no sparkle, there was a gross excess of lower mid range and upper bass, giving this delicate instrument a bloated presentation. The whole presentation of the piano had a flat dull presentation.  I noted that the movie presentations were at much too high a level, often a tip off there is something to hide.  Movie voices had an unnatural quality and clarity left a lot to be desired. In short I could not live with these speakers at all. If I ordered them, they would ship out on the next UPS truck.
I think speakers are making great strides, and I was impressed as much by the similarity of the speakers as differences. However EMP was”odd man out” by a mile.
The designer Shane Rich, who also was generous of his time, was obviously disappointed by my assessment. I did suggest that he effectively has a line source with his six drivers with the problems that entails. I shared with him that I encountered similar problem using the JW drivers as a line source down into the bass. I suggested that, the upper frequency limit of each driver be reduced as you move down the line. This might well clean things up.
I enjoyed Chris Seymour’s exhibit, and enjoyed talking to him. His 1000 watt class D mono blocks using class D ICE amps built by B & O have a superb build quality and a very classy finish.  They ooze quality. As stated above, Chris gave a well researched presentation on the shortcoming of horizontal MTM center channel speakers. Those of you who know my posts, will know that I have posted about this on quite a few occasions.
As far as other lectures, of those I heard, I would commend Jeff Pravia of Epson on projectors. I particularly enjoyed Jonathan Novick’s presentation from Audio Precision on Amplifier measurements. His data was obtained with resistive loads. I of course made a plea for tests with loads having up to 30 degree phase angles. The THX, and Audyssey presentations were also notable.
This SOTU event, I thought was somewhat subdued as a result of the prevailing eccenomic conditions.  It would have been nice to have had more presenters. Gene thought that the economy will have to show good signs of recovery before hosting another event of this nature. I find that regreteble but sensible and understandable.
It was good to meet Gene and his lovely family. At the Banquet I sat with Rickster 71, Wayde Robson, and their wives.
I dressed daft enough to get the best male costume award.
Majorlooser is sporting the horns, I’m the Matador! It was a great evening.
Three way speaker using Dayton DC380-8_Eminemce Alpha 6 and Hi-Vi Research RT 2H-A
Monday, September 15th, 2008This speaker is modeled at the request of Loren42.
The woofer is the Dayton DC380-8. This driver is available from Parts Xpress. This has an F3 of 30 Hz in a 5.5 cu.ft cabinet tuned with a single 4? diameter port 9.2 inches long flared at both ends. Spl is 111 db. The speaker is slightly limited by xmax between 30 and 55Hz, however overall performance is excellent. The vent air speed velocity is 18 m/sec so there should be no issues with vent chuffing.
The mid range cabinet should have all walls lined with mineral fiber such as Rockwool. The cavity should be filled with Polyfill without compression.
The bass enclosure should have 50% of each panel lined with 4? thick Rockwool. Special attention should be payed to placing Rockwool on the panel behind the driver.
Here is the box alignment.
The Crossover is a three way all-pass Linkwitz Riley second order. The composite electrical and driver slopes are third order 18db per octave in the low pass/band pass filter and fourth order 24 db per octave in the band pass/high pass filter.
Zobel networks have been provided to equalize the impedances of the LF and band pass drivers.
The woofer has a sharp cone break up peak at 1500Hz which has had to be dealt with in the woofer compensation network.
The mid band response with this network is excellent and unusually smooth for inexpensive drivers with this network. With the polarity of the band pass driver inverted, the phase response is excellent. Group delay is very acceptable. The impedance curve has been equalized and this driver should be a very easy amplifier load, and not cause any problems.
Allowing for insertion loss of the network, the sensitivity of these speakers should be 92 db 1 watt I meter. This accounts for the high spl.
As is is typical of three ways using a large woofer, the crossover network is complex with a part count of 20 components per crossover. Because the low pass/band pass crossover has to be at 400Hz,some inductor and capacitor values are high. This cost of the components for these crossovers will be significant.
The ports and all crossover parts are available from Madisound.
Even though the drive units are all in the lower price range, with a properly braced and damped enclosure and this crossover network, this should be a superior speaker.
Here is the alignment for a Peavey 15 black widow in a 6 cuft box. It requires 4, 4inch diameter vents flared at both ends, 11.4″ long. Optimal box id 2.2 cuft.
The Bemidji mini
Tuesday, September 2nd, 2008
Here is a simple simple cost effective bookshelf speaker design that will yield superior results compared to commercial offering from manufacturers at similar or greater price tags. The performance of this speaker should be superior to the sum of the parts.
Â
I chose the above drivers carefully to transition well together with a low part count in the crossover. I have selected a small Vifa polypropylene woofer. The on and off axis response is excellent, and there is nothing troublesome with the out of band response. There is a small I KHz peak and as is the norm with small polypropylene bass/midranges a step response. Here is the manufacturer’s specification.
Â
The tweeter is by Morel. This is one of their cheaper offerings. The roll off and dispersion characteristics are excellent. Fs is 900 Hz. This is the full specification.
Â
The crossover needs to make a smooth transition at crossover without ripple, correct the step response, as far as possible take care of response irregularities. It should have a respectable, impedance, phase, and time response. It should also be the most elegant solution with the smallest part count compatible with achieving goals. These crossovers have five components on each board.
Â
Now this crossover has a combined fourth order acoustic and electronic slopes.
The crossover is very symmetrical without ripple. As you work with a design you find the design begins to tell you where the crossover needs to be. In this case it is 3 KHz. This is a very good place for a two way crossover point by the way. There is no zobel impedance equalization.  This is for a good reason. I took advantage of the rising impedance to correct the step response. The low pass filter is in fact first order over most of it operative range. It changes to second order. Now please note that the program does not let me write in and show R1. If you look at the component values you will see that C2 shows an internal resistance of 4 ohms. No cap has this value, so you need to put a 4 ohm resistor in series with C2. This is VERY important. This resistor is operative in the model.
Â
The HF filter has the Q of the filter aligned to combine with the acoustic roll off of the tweeter and make for symmetry with the woofer response and make a very smooth transition. This crossover also optimizes the off axis response of the speaker. The tweeter is 33db down at F3, which is excellent.  This filter also switches order.
Â
The step response is corrected. A small remnant of the small woofer peak remains, however the Q of the peak is changed. It is lower in magnitude and moved down to 750 Hz.  It is 400 Hz broad from 600 to 1000Hz, and less now about 2db. I have made the judgment that in that range and at that magnitude it will do no arm. Also I suspect that if the speaker is around 10 inches from room boundaries that the response below the slight peak will be lifted by that small amount and help compensate for cabinet diffraction loss.
Â
A word about sensitivity, note that the average sensitivity is now 86db 1 watt one meter. That is because with any passive network does not allow for boosting a signal, only attenuating it. This reduction in sensitivity is a consequence of correction of the step response.
Â
Now impedance.  Now the bass curve will be the one shown in the woofer alignment. The published spec is sealed, and has one impedance hump. This woofer requires reflex loading and there will be two impedance humps. One is at 20 Hz the other at 73Hz. There is a dip to 6 ohms between the peaks. Now although these are 8 ohm drivers, correcting the step response, which in my view is essential, always drops the impedance. The impedance is 6 ohms from 200 Hz to around 650 Hz and then rises. This is where a good deal of the power is, so regard these speakers as 6 ohm. There is an impedance rise to 14 ohms in the crossover region.
Â
These speakers should be an easy amp drive. Any amp the over heats driving those, should never have seen the light of day.
Â
Now to phase, this crossover puts woofer and tweeter out of phase at crossover. There is an unusually good phase and time response. The shift at crossover is 2.2 inches or 0.17msec. Now a woofer�s acoustic center is the geometric center of the cone, which for me is impossible to visualize and certainly calculate. However the tweeter will be ahead of the woofers acoustic center, so will modify this. I would start with the wiring as shown, but when the speaker is constructed see which polarity has the greatest output at crossover. I have a hunch that with the drivers lined up on a flat panel recessed flush, the polarity switch will not change the output. If that is the case, I would wire the tweeter in phase. This is something that is hard to predict from modeling.
Â
For the crossover use good quality air cored inductors and mount them as far away from each other as possible, with one horizontal and the other vertical. This will minimize cross inductance. Please select polypropylene caps.
This should be an easy low budget speaker to construct. I would welcome comments from anyone who builds a pair.
Polk RT3 speaker mod.
Monday, September 1st, 2008This is a mod for the Polk RT 3 speaker. It uses this woofer. This is a pdf. of its specification sheet.
It uses this tweeter. Â Here is the spec. sheet.
This is the woofer alignment.
This is the crossover pdf and simulations.
The crossover is at 3 KHz. However the component values have been significantly modified. There is 2 the 3 db of diffraction compensation and the response irregularities of the woofer between 200 and 2000 Hz have been significantly blunted.
L1 and C1 form the high pass filter. This starts to attenuate first order at around 6.5 KHz and switches to second order around 1200 Hz. The tweeter is down 33db at resonance. This should mitigate against any tweeter roughness. Rp1 and Rp2 pad the tweeter output.
L2 and C2 form the low pass filter. This starts as a first order filter around 400 Hz and starts to perform diffraction compensation. It changes to second order just before crossover at 3 KHz.  Req and Ce perfrom impedance compensation. Without it there is significant rise in response above crossover.
Note that providing diffraction compensation has sacrificed about 3 to 4 db of driver sensitivity. Although the sensitivity is around 87db 1 watt 1 meter which is good for a small speaker like this.
There are impedance peaks of 50 and 40 ohms as a result of reflex loading at 25 and 100Hz respectively. Because of the equalization of diffraction and smoothing of the response the impedance is 6 ohms over most of the operating range. There is a rise to 12 ohms at crossover.
The phase response with the polarity of the tweeter reversed is excellent. with the drivers having a maximum pahse angle of 45 degrees in the crossover region. Driver offset of the front baffle will modify this a little. As result group delay is only significant on the driver tuning range.
The existing vents should be made 7.62 inches each.
The crossover can be mounted in and out of the cabinet.
This mod to the RT 3 should be able to be accomplished with minimal cabinet modification.
If a bigger box were to be built, then F3 could be lowered to 59 Hz.