Saturday, June 9, 2012

Super Tweeter

I've been coaxing more and more from a pair of old Bozak B-302a speakers. They started out sounding interesting but I remember the sound of a friend's Concert Grands back in 1972 and could not get them out of my mind. He built many models, all smaller than this refrigerator-sized speaker, and the B-302a was one of his best selling models. It had a 12" woofer, a 6.5" midrange, and two 2" tweeters. First-order crossover networks (-6dB/octave) were used throughout the entire line.

Rudy Bozak tried his best to push the state of the art forward but his designs suffered from infancy in technology. Many things were done very, very right and others - well not so much. Thiele-Small did not come along until the 1980s so he was just basically winging a lot of things. The biggest issue was high frequency response, that above 8KHz. He also did not flush mount drivers so his understanding of first reflections was missing. But he did understand low-mass cones and to some degree dispersion.

I thought about trying to keep the original design and after looking at it for a few days decided only to keep the original box for the woofer and move everything else out to a satellite. If I decided later to changer the woofer box, the satellite would already be finished. So the mid and tweets were removed from the original box as was the crossover network (moved to the back panel allowing easy access for tweaking) giving the most volume to the woofers as possible. I also bought another pair of woofers and four more tweeters so I had a total of eight tweeters, two midranges, and four woofers to work with.

The 2" tweeters, although an attempt at high frequency extension by using aluminum center caps, just had too much mass to be considered state-of-the-art today. They did work well up to about 7-8KHz and started to die pretty sharply after that (see Before SPL graph). I used an asymmetrical truncated pyramid box and stuffed the tweeters in a vertical array yielding the best horizontal dispersion pattern. To improve the vertical dispersion, I also arced the array and added a large chunk of Bondo body filler to the baffle board thus providing a low-reflection surface and smooth transition.

Although quite good sounding in their limited range, the lack of HF response was disappointing. I temporarily added a super tweeter (ST) I had lying around by resting it on top of the cabinet.  From its limited contribution, I knew that this was the correct direction to go. Searching for a ST that would be efficient ans still sound like music proved to be a challenge. The 1-watt sensitivity on this system was about 96dB/W/m so finding something with a sensitivity above that was a long process. Finally, a friend recommended considering the Audax TW025A28, a gold dome that if nothing else would be aesthetically pleasing.
Testing this ST by sitting it on top of the satellite again provided significant HF content at a sensitivity that approached the Bozak 2" tweeters.  Instead of adding a new location, I removed one of the existing 2" tweeters (second from the top). This provided a smooth transition while maintaining good direct radiation. BUT I had to build a new faceplate since the round one was too big to fit into the slot.  I fashioned one from some thin oak paneling and the rest is history.

So what does it sound like? My wife is the best test to such changes since she is an artist, not an engineer. Her reaction was an immediate thumbs up tossing phrases like "amazing" and "it's like listening to a whole new speaker" since she was truly unaware of what was measured to be missing.  She listens to them louder than before, another sign that she is listening to the details rather than merely tolerating the sound. Speaking of measurements, here are the before and after RTA results.



Now given these are RTA measurements taken with my cell phone and an app called RTA Pro, the overall accuracy is in question, especially below 200Hz and above 10KHz, but the relative changes can be considered reasonably representative of what happened and also reflect our subjective impressions. The top octave is there as noted by the marked increase in HF content above 6KHz and interesting there is an additional smoothness in the 500Hz-2KHz region unexplained at this time. The 4KHz peak is still there but not as pronounced (wide) as before, possibly because of the addition of the top octave.

So what does this mean to the high end? Adding super tweeters adds enjoyment (duh!) but the physical location of these drivers determines how well the system will sound. Adding the ST and resting it on top to "test" its effects did not permit the driver to perform to its full potential. Putting it in the proper place made a world of difference in the tonal balance.

Yours for higher fidelity,
Philip Rastocny

I do not use ads in this blog to help support my efforts. If you like what you are reading, please remember to reciprocate, My newest title is called Where, oh Where did the Star of Bethlehem Go? It’s an astronomer’s look at what this celestial object may have been, who the "Wise Men" were, and where they came from. Written in an investigative journalism style, it targets one star that has never been considered before and builds a solid case for its candidacy.

My other titles include:
Copyright © 2015 by Philip Rastocny. All rights reserved.

No comments:

Post a Comment

To comment on this blog, you must first be a member. All comments are moderated.

Note: Only a member of this blog may post a comment.