Instrument tests on 5 tweeters

[Deang]

So you think Selenium would decide to run distortion tests on a driver, and then pick the horn that would deliver the worst results. Good marketing!

Throw out the Selenium plot and just look at the Beyma plot -- I don't see anything there that would compel me to quit liking that driver.

[BEC]

I think Selenium may have been trying to show off the low frequency response of the D220Ti on that 600hz horn.

Bob

[Al Klappenberger]

Dean,

The argument about measurements versus listening tests has been going on for ever! I believe you need to measure what you can measure and use you ears for what you can't measure. If it measures good and still sounds bad, you need to measure something else!

Also.. Look at the "h2" plot on the Beyma. It shows about 20 dB down at 10Khz. That is exactly what the 10KHz spike shows on my intermod plot. That's the 2nd harmonic of 5 KHz. I didn't post it, but I did a single tone test at 5 Khz at 100 dB. The harmonics where still there. Now look at the harmonics on the B&C and the Selenium tweeters. The spike at 10 KHz is about 55 dB down on them. That's significant! For that one harmonic alone, it's a difference of 10% distortion versus 0.3%.

Al K.

[Guest David H]

Nevermind, Looking back on the test I found the answer to my question.

Dave

[Deang]

Anyone ever tell you guys that you're irritating? [8-|]

[Al Klappenberger]

Dean,

I have been called worse!

Reality stinks, don't it?

Al K.

[greg928gts]

Conclusions:

The cear winners are the B&C DE10 and the Selenium D220TI. Both of these are very inexpensive and nearly the same price.

I am VERY disappointed in the distortion from the Beyma CP25. I will no longer be recommending them and will probably replace them in my own speakers in the future. I suspect the problem may be the small round aperture in the center of its horn. This might be to improve dispersion. I think that's a secondary consideration compared to modulation distortion!

Al K.

So let me get this straight. The tweeter you used to recommend to people for years, that is now sold by one of your competitors, now has all kinds of distortion that you didn't realize was there before, and the tweeter you are now selling as a dealer is testing the best of all five of these?

R r r r r r r i i i i g h t.

Greg

[ZAKO]

Rrrrrrrrrrrrriiiiiight !!! And so where has the masking effect been all these years???

[jacksonbart]

Don't worry, its the room acoustics that are to blame.

[Al Klappenberger]

"The tweeter you used to recommend to
people for years, that is now sold by one of your competitors, now has
all kinds of distortion that you didn't realize was there before, and
the tweeter you are now selling as a dealer is testing the best of all
five of these?"

Yep.. That's exactly it!

I suggest you duplicate my instrument tests. You might be able to prove me wrong and make me look like a fool. Now is your chance!

This is what happens when you depend on just your ears. Distortion often sounds "Lush" or "silky". That sounds good to the untrained ear, but it's WRONG!This is why I hardly ever give my subjective opinion on how something sounds.

Al K.

[greg928gts]

I'm not so arrogant to think that I can do a better job at testing someone elses drivers better than they can. The companies that built those tweeters all have testing equipment that would make yours look like a battery checker, but somehow you're testing is going to reveal something that they didn't find.

R r r r r r i i i i i g h t.

Audio is a subjective artform. The testing gets you just so far, and then from there you have to use your ears. There's nothing revealed in your testing that's going to change what I hear.

You might be able to prove me wrong and make me look like a fool.

I can't do that any better than you. []

Greg

[jacksonbart]

I'm arrogant.

[Al Klappenberger]

To all:

I am bothered by the thought of all the people who bought the CP25 tweeter just because I recommended it. I really owe them an apology. There's a certain amount of responsibility that goes along with a reputation. I should have done these instrument tests before I promoted the CP25.

Consider this my apology. All I can say is that I deceived myself as well! I fully expected the CP25 to be the run-away winner in the tests. I was very wrong.

As I have said before, be careful how you choose your experts!

Al K.

[Johnny's Jill]

Consider this my apology. All I can say is that I deceived myself as well! I fully expected the CP25 to be the run-away winner in the tests.

No apology necessary, distortion or not the CP-25 is a great sounding tweeter and far superior to my tin ears than the K-77. I have not heard or tested the Selenium therefore I can't comment on it.

I don't think you steered anyone wrong by reccomending a product that subjectively sounds good to you.

Craig out 73

[Nat Denkin]

Here is another paste-up of the 5 tweeters for IM distortion. Harmonics of the 5 KHz tone can clearly be seen too.

Al K.

Al,

I have been using your trachorn (Atlas driven) and unviersal crossover along with the Beyma. People are impressed with the sound and the lack of beaming (large listening sweet spot).

The measurements presented at the begining of this thread suggest that the Beyma disperses the power better than the other driver/horn combinations. While the other tests appear to have been done at constant power, the conditions of the distortion test set SPL=100dB on axis at 2 feet. If the Beyma is dispersing the power better, the test would be driving it harder as it radiates a over a much greater angle. I have to wonder what the results would be if the measurement were made with 1W input measured at 1 meter and averaged over the power delivered in all directions.

Are there industry standards for these tests? I recall in another thread that you used tones with differing powers for testing midrange horns.

Thanks for taking the time to do independent measurements. Still, I have to wonder why providers would not publish their own test results if they had distortion measurements as low as 0.3% .

nat

[Al Klappenberger]

Mr. Roberts,

You have no room to criticize anyone else's testing. You have no electronic instruments and would not know how to use them if you did. The only instruments you have are your ears. Those, like all others need to be calibrated. The human ears are calibrated by listening to live music. I have very few opportunities to do that, so I restrict myself to testing with instruments I CAN calibrate using known standards. Your ears have been calibrated by listening to hammers, power saws and sanders. You should stick to doing what you know how to do well, building and referbishing cabinets rather then backing into other peoples areas of expertice.

To all,

The levels I used (100 dB SPL at 2 feet) may be too severe, but I used the exact same levels with all 5 tweeters in exactly the same way. If I were going to do any fudging, it would have been to try to make the T-35 / K77M look bad. It, in fact, looked very good. I have always been impressed with the horizontal dispersion of the CP25. At lower levels, distortion reduces. Tweeters generally are not required to handle a lot of power at normal listening levels. This is probably why the CP25 has been accepted. It may also have been because I suggested it and made brackets available to use it with my Trachorn frames. I bear some responsibility. I have never been happy with the sound of symbols reproduced by my speakers. I assumed the problem was simply the state of the recording art. I assumed the same thing about the smearing I used to hear on strings in recordings of symphony orchestra performances until I moved to the extreme-slope squawker / tweeter crossover and it went away. Maybe, when I replace the CP25s with lower distortion tweeters, symbols will sound better. I will find out as I plan to replce them soon.

AL K.

[Nat Denkin]

....

The levels I used (100 dB SPL at 2 feet) may be too severe, but I used the exact same levels with all 5 tweeters in exactly the same way.....

AL K.

Al,

What was the input power to each tweeter for this test?

Thanks,

nat

[Al Klappenberger]

About the levels used for the intermod testing:

PWK did some tweeter testing using the same tone frequencies, 5 and 12.5 Khz. He chose to apply a level of 1 Volt. Most of the tweeters I tested displayed 105 dB SPL at 1 meter with 1 Watt applied. 100 dB at 1 Meter (about 3 feet) would require .316 Watt. That computes to 1.59 Volts (RMS). At two feet, it would require slightly less power to generate 100 dB SPL. I think it’s safe to assume that the total power applied to each tweeter during the tests was about .5 Watt including both tones. I did measure the input voltage while I made up my mind about what levels to use in the tests. Once I decided on 100 dB SPL at 2 feet, I quit measuring the input and simply adjust to get that acoustic level. I’m not sure if this level represents a fair evaluation of reality or not. At least, all the tweeters were compared using a level playing field!

Al K.

[ngk]

Al,

Please help explain a few things as I only have a vague understanding of the measurements that you made:

The frequency plots for the 5 drivers would indicate (to me) that the Selenium driver has the smoothest and most extended response, followed by the Beyma CP25. Am I correct, and is this extended resonse one of the reasons the Beyma was recommended over the K77?

From what I can understand, the IM distortion plots show that the K77 is not too bad, and the Selenium is the best compromise unit for frequency response and distortion. The K77 distortion seems better than the Beyma (I think) - why did noone really notice this? I have the feeling, which is how you always qualify all your findings, that our hearing does deceive, but more importantly, IM distortion at most listening levels is concealed by music quite effectively except in very poor quality drive units.

I have lived with 1st order crossovers - type A and the dhaxovers, homemade - for several years without being entirely happy. I recently moved over to your universal design and the khorns sound so much better at all volumes. Besides reducing drive unit overlap, I have the feeling that the steeper crossover on the tweeter is reducing distortion in the Beyma. Am I correct in thinking that a steeper crossover will help reduce IM distortion as less energy at lower frequencies is reaching the tweeter?

Will the sensitivities of the B&C and Selenium units necessitate the use of your tweeter attenuator which I already own but do not at present find necessary with the Beyma?

Al, I can see that you are going to cost me another pile of money in upgrades if you are correct.

Nick

[greg928gts]

In the never-ending quest for improving the sound of my system, I'm really looking forward to hearing the Selenium and the B&C tweeters mentioned here. I just ordered a pair of each and will be doing a subjective comparison with my Khorns. Since I actually own Khorns and listen to them regularly, I think my customers will appreciate the information I provide to them regarding my subjective evaluations.

Greg