PR Audio

Driver Matching Service

 
PR Audio offers driver matching on many of the speakers (drivers) we sell.  This eliminates the strong possibility(1) that you may receive drivers with significantly differing curves or efficiencies.  Some vendors offer matching in high end models, but, we offer it for much more affordable units.  We can offer matching of pairs to within +/- 2 dB over most(2) of the range of almost any driver we stock in quantities of 10 or more.  In many cases we can offer matching to within +/- 1 dB for a little higher cost.  On a few models we even occaisionally match up pairs to +/- 1/2 dB, though this is a hard specification to “make” on a consistent basis.

Below are curves on a set of our PR1C9SD-HE1 tweeters (modified), matched to within +/- 1 dB over the entire audible range (they actually are better than +/- 1/2 dB from below 2500 Hz to 18 kHz.)

Matched Pair, PR1C9SD-HE1, 15º off axis, +/- 1/2 dB from 2.5-18 kHz

(Click image to enlarge)

http://praudio.com/site/pr1c9sd-he1-high-efficiency-1-soft-dome-tweeter/

Note that the above units were surface mounted:  Flush mounting would smooth out the ripples above 5 kHz.  Matched pairs are normally curved at 15º off axis, with 1/3 octave smoothing, for the most “realistic” results.  Please Contact Us if you have a different requirement.

Sets of 3, 4, 5, or 7 (as for home theater) can also be matched.  However, it is much more difficult to find sets of 4,5 and especially 7*, units, matched within +/- 1DB, so we generally offer +/- 2 dB matching on the larger quantities.  Fewer of our drivers are available with 5 or 7 driver matching, as well.  A possibility is to have one pair more closely matched, for “main” / front speakers, and then have the other 3 or 5 within +/- 2 dB of the first pair, and each other.  Again, Contact Us for any special requirements you have.

*Eg., 5 or 7 tweeters such as for a 5.1 or 7.1 surround sound system.

Notes and more notes:

(1) Looking at it another way, we’d estimate that if you buy a pair of medium price tweeters, say, in the $20 – $50 each price range, the odds of getting a match within +/- 2 dB anywhere in their operating range is probably  50% or so (it’s usually but not always better on the high end of that price range.)  But, that means that you have maybe a 40% chance of getting a pair that does NOT fall within that tolerance.  If you buy 4, intending to send 2 back, you probably have more like an 80% chance of getting that match on two of them, but then you have to do the testing (if you can) and pay for return shipping (assuming the seller even allows returns of non-defective units, and / or does not charge a restocking fee.)

(2) When we match to within, for example, +/- 2 dB, that means that after 1/3 octave smoothing, all parts of the two drivers curves will be within +/- 2 dB of each other, over the entire range the matching is specified over.(3)  As a practical matter, we will keep the overall sensitivity match of two such matched drivers to within +/- 1 dB (or very very close to that.)

(3) Curves tend to be a little more erratic at the frequency extremes.  For example, in the VERY well matched tweeter set above, the “matching range” specified is 2.5 kHz to 18 kHz, with the curves’ largest divergence above 18 kHz.  This is very typical, and, generally a divergence at over 18 kHz is not nearly so audible as a smaller divergence at, say, 5 kHz.  However, we DO also look at the curves / comparisons outside of the stated matching ranges, and, all else being equal (or nearly so), we minimize those differences as well.

(4) We match drivers using an expanded scale such as what you see if you click on the curves image above.  Additionally, our LMS package allows us to compare curve levels at any frequency selected, to within 0.01 dB, which is far more resolution than needed, but is very handy.  And, in our final determination, we analyze curves manually:  By that, we mean that Paul actually looks at every unit’s curve, and the comparisons, himself, to determine curve trends and other details hard for most analysis software to pick up on.  This assures the best possible audible match(s).  (See what THAT would cost you from a high end manufacturer!)

——————————————–

Additionally, part of an e-mail we recently sent a customer may be useful to you (see below).  Prices for pairs matching, where available, is usually available on the particular item’s page, but, again, if you don’t see it provided on a particular unit, or if you have some other requirement than is listed, please Contact Us . 

——————————————–

Hi, Mathias,
 
To more fully answer your question, we have several of the
MB Quart 95-7048 left, as well as quite a few of the 95-6814.
Depending on how closely you need them matched, I am fairly
sure we could provide 4 pcs. of the 95-7048, all matched to
each other, and I am quite sure we could provide 4 pcs. of
the 95-6814, all matched to each other.  I will need to run
curves on some more units to be sure, but +/- 1 dB matching
of 4 units looks “doable”, and +/- 1/2 dB matching of 4 units
may be possible (especially in the case of the 95-6814′s).
 
It’s quite a bit harder to come up with 4 units, all matched to
each other, to a tight tolerance, than simple “matched pairs”.
To be honest, I have not yet worked out pricing for something
like this, but for now, I would say that if you want 4 pieces,
all matched to within +/- 1 db, a fair price would be the same
as: two matched pairs, +/- 1/2 dB, with the pairs not necessarily
matched to within +/- 1/2 dB.
 
Ie.:
 
Unit 1 = Unit 2 = Unit 3 = Unit 4, all +/- 1 dB. >>> $220.00 for all 4.
 
OR, if
 
Unit 1 = Unit 2, +/- 1/2 dB, and Unit 3 = Unit 4, +/- 1/2 dB,
but Units 1=2 would not necessarily be within 1/2 dB (or even 1 dB)
of units 3=4*.   >>> $220 for all 4.  (Per the website pricing.)
 
*In the 2nd case, we would try to kep the overall match close to
  +/- 1 dB, and certainly within about +/- 1-1/2 dB, but it might be
  difficult to guarantee better than +/- 1-1/2 dB overall, as the
  +/- 1/2 dB pairs matching reduces the choices a bit, overall,
  unless we further enlarge the sample group, which entails more
  work**.
 
**It’s relatively convenient for us to compare up to 20 units at a time
    / in a group.  After that, it gets to be considerably more “tedious”.
 
   (Comparisons or matches to a “golden” (reference) unit are a little
   different “ballgame”, easier in some ways, more problematic in
   others, but that’s not really what we are looking at, here…)
 
It might be possible, with some extra work and a bit of luck, to come
up with Unit 1 = Unit 2 = Unit 3 = Unit 4, all +/- 1/2 dB.  I’d really
have to see just how much time that would take, but, I’ll go ahead
and say that IF we can come up with matching that tight, the price
will not go over $275 for the 4 pieces, and if we do a little better
on the time than I am thinking, maybe it’d be $260 for all 4.