May 4, 2003 -- PeeCreek Project has been published!


    Please find the complete project here.

    I have prepared a brief tutorial on how to mount a PVC vent pipe.  You can find that here.

May 2, 2003 -- A few things...


    Based on the volume of email I've received over the past two days, quite a few of you would like to see my alternative lowpass for the AR diy kit designed by Ed Frias.  The common complaint has been a lack of midrange clarity and "thin" sound.  This lowpass should solve that problem by better controlling the rolloff of the woofer and adding a bit more baffle step compensation.  You'll need to increase your tweeter padding a bit (I put a 6 ohm resistor in place of the 4 ohm resistor in the highpass and it seemed to work nicely).  Here's the lowpass (18 awg inductors are called for, although you can go with 14 awg for the 1.4mH inductor with, perhaps, a hair of improvement in the bass performance).

    Also, many of you have asked to see what I found for the 850122's TS parameters.  They do differ significantly from the manufacturer's published data.  The following is representative of the 8 woofers I tested.


(click to enlarge)

    I'll be finishing the PeeCreek write-up over the weekend, so stay tuned.

April 27, 2003 -- Is there a PeeCreek in your future?


    This is turning out the be a potential all-star in budget DIY designs!  This picture shows the PeeCreek prototype finished in PE's vinyl laminates -- cherry on the top and sides and the new textured black on the front, back, and bottom.  The picture doesn't do justice to just how nice a finish this inexpensive, easy material can produce

    The final crossover is still in testing, so stay tuned.

April 22, 2003 -- What's in a box?


    In this box is a Peerless 850122 6.5" woofer and the new North Creek D25 1" tweeter.  So what's so interesting about it?  Well, it's got the makings of an excellent budget speaker with a wee bit of a twist (this is pretty dry stuff, folks, so we stretch the definition of "twist" sometimes to keep the readers from lapsing into mild comas).  The twist being that the design is optimized for use in the new Parts Express prefabricated enclosures.  These are some beautiful enclosures and will certainly be a godsend to those who are avid about building their own speakers, but, for whatever reasons, do not build their own enclosures.  As you can see, rather than buy the enclosures myself, I chose to build exact replicas.  Detailed enclosure construction plans will be available when the project is complete, for those of you who prefer to build your own enclosures.

    One problem with many DIY design is that the low-frequency alignment (a fancy way of referring to the bass response) is not well-conceived.  When designing a speaker enclosure, the low-frequency alignment must be carefully chosen based on the parameters of the woofer, the desired power handling, the linearity of the driver, and quite a few other things.  The alignment, itself, is the combination of box size and tuning (whether it be sealed or vented).  The chosen alignment determines how the woofer will rolloff on the low end.  In other words, the box is a lowpass filter for the woofer.  This isn't anything new to those of you who have read even introductory texts on speaker design.

    Ok, now that we've established that the box is a lowpass filter.  Why don't most people give it the same level of attention as they do for the various filters in their electrical crossover networks?  Give up?  It's because it's a mechanical lowpass filter.  Maybe that will helps to put it into perspective.

    Those of you who aren't reading this site for the first time probably realize the importance and difficulties associate with proper crossover design when it comes to integrating the drivers frequency responses in a manner that sounds good.  The enclosure is an extension of that same process, but it only affects the woofers... but the effect is quite large.  A larger box and lower tuning frequency gives greater bass extension, but it lowers system efficiency.  Power handling also suffers.  All these things combine to determine how loud a speaker can play cleanly, i.e., with the drivers operating within their linear pistonic range.  Maintaining driver linearity at sound pressure levels approaching that of live performances is a very important objective in designing a true, high-fidelity speaker.

    So why the lecture on box design?  Simple.  Many DIY designs don't pay that much attention to the enclosure and, quite often, simply use a reasonable tuning frequency in a conveniently-sized prefabricated enclosure.  Be wary of these designs.  Don't dismiss them out of hand, either -- many of them are worth your time and effort to build.  A few aren't.  As always, do some research before you build.  And by all means, if you're going to publish a project for use in a prefab enclosure, make sure to choose your woofer such that it's performance will be optimized for the enclosure you recommend.  Normally, you choose the woofer first, then you build an optimized enclosure.  When you're buying "off the shelf", you don't have that luxury.

    So, since I've had a lot of requests for suggestions for using the Peerless 850122, I've decided to hook it up with the North Creek D25 and see how it goes.  So far, I'm quite impressed with the combination, although neither driver has (up until now) impressed me much.  They're both good at everything, but neither excels at anything.  I am extremely impressed, however, at just how good the North Creek tweeter performs for the price -- matched pairs for $36 is an exceptional value.  The increased imaging accuracy that can be obtained with well-matched drivers is not insignificant.

    More details soon....

April 7, 2003 -- Too good to be true...


    In my last update, I proclaimed the Usher 8945w as the best "bang-for-the-buck" driver out there.  Then, the price went up (it's now listed at $96/each at Euphase Audio).  This makes it nearly 150% of the price of the Vifa PL18.  That's rather interesting, because I'd estimate that it provides approximately 150% of the performance of the PL18.  On a cost performance basis, the 8945w is now about on a par with other drivers -- the PL18, the SS 8545k, Audax HM100's, etc.  Although I'd still recommend this driver over the SS8545k in any three-way application.  In a two-way, the Scan Speak woofer has a bit better clarity and (perhaps) slightly less coloration, so if money is no object I'd go with the Scan Speak (bah -- in a two-way, I'd go with the Scan Speak 7" Revelator when cost is not the issue!).  When money IS a factor, the Usher will provide exceptionally good performance for the price.  Chances are, the design will dictate how good the Usher will really sound in any application, so research the design you choose to build carefully before you decide to invest your time and money.

    The good thing is that it doesn't change my opinion of this driver's capabilities -- the Usher 8945 is an excellent addition to the DIY arsenal, where deep, well-resolved bass is required from a 7" driver in a moderately sized enclosure.  In a two-way, it produces excellent midrange with a somewhat dark, sterile quality.  It's only in direct comparisons with the Scan Speak 7" Revelator does it show any signs of coloration.

    Now, onto news.  Project X is complete.  The prototypes were finished in a very industrial truck-bed liner finish, to protect the enclosures during transport.  They're spending most of their time on the road, traveling to auditioning sessions.  You'll notice that I reduced the amount of felt on the baffle.  When I finally got down to voicing these speakers, I preferred the sound of the midrange without the felt, but using felt around the tweeter definitely clears up some baffle reflections in the high-end.  I'll be posting the final crossover, measurements, and enclosure details soon.