CSS LAR1 Ribbon, Audax PR170N0, Usher 15HM

[roger_lew]

Hey All,

The weather is finally starting to get nice up here in the northwest which means it is time to start building speakers. I'm getting ready revise my 3 way actives. Currently they are loosely based on the Linkwitz Pluto; in that they use the NSW2 and are fairly omnidirectional. They have Usher 15HMs on the bottom up to 350Hz, and Dynavox 5.25" polycones up to 1000Hz. The bass is great and they have a very wide sweet spot thanks to the polar response, but I've come to the conclusion that their top end is lacking something and the midrange doesn't grab me nearly as much as a set of 2 ways I have with BG Neo8PDRs powered by a Sonic Impact 5066.

My primary goals are higher efficiency, better dynamics, high end frequency extension (compared to what I have now), and compactness (again, comparatively). I'm not going for 120dB, but I want the efficiency so I can use higher quality amps that are easier on my pocket book. Right now I have a NAD 304 up top and a Pioneer SX-6 in the middle. The NAD 304 is listenable, the SX-6 isn't so hot, even between 350Hz and 1000Hz. I'm probably going to build some 41Hz tripaths and keep my Pioneer SX-1050 (120 w, ~60lb) on the bottom.

I've already got the drivers sourced. The Usher 15HMs will continue their bass service. They are going in a 2.7 ft^3 sealed enclosures (qtc=~.7) and will be eq'd to an F3 of 35Hz (12db Shelf, which is why I'm keeping the SX-1050 on the bottom). In the future I plan on having real subs, but in the short term the Ushers have plenty of displacement for my needs/room. They will still do 100db at 30HZ without room boost on paper with 128 watts. That isn't so shabby.? Some of you are laughing smugly now, that's ok.

For the midrange I bought a pair of Audax PR170N0s from madisound. They are a coated version of the PR170M0s. They are a bit of gamble, but the response plot just looked to pretty to pass up. I plan on putting them in a sealed enclosure and crossing them at 350Hz 4th order. Since I'm not going for 120db they should be fine even with their .5mm xmax. If anyone is actually bothering to read this they are probably wondering why I'm not going open baffle. Well, I'd love to but SAF and room placement dictate against it.

For the top end, Bob from CSS gave me a bit of a deal on a pair of their new LAR1 ribbon tweeters. They have a honeycomb foils which are suppose to reduce distortion. I choose them for their efficiency and hopefully dynamics (I haven't got to play with them enough to judge their dynamics). My initial plan is to cross them rather high as the PR170N0s don't seem to have any harsh breakup modes. Hopefully next week I'll have time to mount the ribbons and the mids on open baffles temporarily so I can play with them more before I start the final enclosures.

For SAF I'm planning on building them as narrow floorstanding towers with the woofers side mounted. I can use my DCX2496 for time alignment. I've measured them at 90 degrees and tested them in my listening room and they seem to work fine that way (they actually seem better to tell you the true, possibly because they are farther from the rear wall). Dimensions will be 12" wide, 43" high, and 19" deep. All walls will be 1.5", the cabs will weigh about 130lb without drivers. I was going to use OSB/MDF sandwich construction, but there was a lot of wasted material so I'm going all MDF. It'll take 3 sheet of 3/4" to build both cabs.

In the long term I wouldn't mind retiring my DCX2496 but in the short term I need it at least for time alignment and bass eq. Any thouhts or Suggestions are welcome, there are still some things that are up in the air.

Here's a rendering:

[roger_lew]

x-ray:

[roger_lew]

plan view:

[roger_lew]

cut list

[roger_lew]

cut plan

[roger_lew]

So I found some time to build router jigs and temporary open baffles for the LAR1 and the PR170N0. They are 17.75" tall and 18" wide. The main panel is 11" wide and there are 5" wings at 45 degrees on either side. See picture below. Currently I'm using crossover points at 500 Hz and 5kHz (all LR24) and set levels at listening position using 1/6th octave smoothing. The vertical dispersion of the tweeter doesn't seem to be an issue. It is actually much better than I anticipated. The mids had a terrible hiss with my SX-6 so I swapped it out for the NAD 304 which proved to be much quieter, and am using my 25 watt TDA7265 chipamp for the Ribbons. Below are some of the initial nearfield measurements I made on the LAR1 and PR170N0.

My measurement setup consists of Behringer ECM 8000 microphone, Behringer DSP110 Shark microphone preamplifier, Echo Indigo I/O PCMCIA soundcard, TDA7265 chipamp, and SynRTA. I exported the data to csv files from SynRTA and plotted them in Excel with no smoothing. The Indigo I/O isn't super clean. Loopback shows .5 dB hysteresis all the way across the frequency band. I also don't trust the LF response it is reporting. It is showing a strong rolloff at 100 Hz with my Ushers which is inconsistent with measurements I made with my EMU-0202.

[roger_lew]

closer:

[roger_lew]

back:

[roger_lew]

LAR1 nearfield measurements. I was pleasantly surprised with how closely the frequency responses matches with their datasheet. The datasheet can be found here:http://www.creativesound.ca/details.php?model=LAR1

[roger_lew]

PR170N0 nearfield measurements. These are still breaking in so I don't know how much they will change once they settle. I'm guessing the little dip at 350 is related to the baffle.