What beats the R2904/7000 ring radiator?

[Pan]

Among tweeters he has tested so far, the lowest distortion units regardless of price are the Scan-Speak 6600

Did you miss the C23? It has lower distortion on my screen if I read the graphs right.

Anyone knows what's up with the comblike patterns on some of Zaph's measurements?

Also does someone know what microphone Zaph uses for his tests?

At the level and distance he use the contribution from the mic can start to enter the picture.


/Peter

 

[Pan]

Also, as Jay_WJ said, distortion products at 5 kHz are hard to hear, you'll barely hear the 2nd order, you'll barely hear 3rd order, and every other order are inaudible because over 20 kHz... not to mention that again, most tweeters are good over 5 kHz, low distortion, at any price. [/B]

Harmonic distortion products yes, but sum and difference tones will potentially end up all over the place if you listen to other things than sine waves. ;-)

Also there use to be signficant difference even above 5k.


/Peter

 

[simon5]

Also on the comments about the cross section, every person who puts the music reproduction over the measurements knows that high order crossovers are not usable, it is important to keep the crossover as simple as possible. I would never use anything over 2nd order, this is why I don`t like metal and ceramic cones.

Well if you crossover 1st order... that's why you can't crossover low... basic physics. Did you ever try a well built and well designed LR4 crossover that sums well with a good tweeter crossed low ?

Also, I'm not imopressed by the use of Argumentum ad populum in your argumentation ! 😉

 

[Mario Pankov]

Zaph looks at the measurements only, not to the sound produced by the final product. C23 is a good tweeter, better than it`s cheaper equivalent the KE25SC ( which is also a very good product ). My opinion is it is not better than the 2904/7000. A friend of mine built a pair of monitor speakers using the C23 and a ceramic cone midbass and a few weeks ago he said that will throw them through his window...I listened to them, I know they are not my tweeter. 🙂
------------------------------------------------------------------------------------------
Well if you crossover 1st order... that's why you can't crossover low... basic physics. Did you ever try a well built and well designed LR4 crossover that sums well with a good tweeter crossed low ? - I do not crossover low 😉 No, I have not tried 4th order because I have listened to enough speakers that employ a 4th order. I simply choose drivers that have an excellent cone, excellent motors and do not need anything over 2nd order ( Scanspeak Revelator units ).

I will say it again - 2-way designs are a compromise, the addition of a midrange driver is always a better way to go.

 

[simon5]

Anyone knows what's up with the comblike patterns on some of Zaph's measurements?

Also does someone know what microphone Zaph uses for his tests?

At the level and distance he use the contribution from the mic can start to enter the picture.

/Peter

My mic is now the popular Behringer ECM8000, and my pre-amp is a Behringer EURORACK UB1202. Note the UB802 will work fine, but the UB502 will not since it doesn't have phantom power for the mic. These items were purchased at 8th Street Music. (8thstreet.com) I rounded out the package with a boom mic stand from Rat Shack and a good but inexpensive low Z XLR mic cable. The ECM8000 is a very flat mic. You can probably get by without even using a calibration file, as it's only about 1.5 dB down at 20kHz. On the other hand, the common calibration files for this mic are very consistent and will probably work fine for you. My mic is calibrated and the cal file includes the mic cable and the speaker cable.

Here you can read about his test setup.

http://www.zaphaudio.com/setup.html

You mean combing of the fundamental in some graphs ? That was caused by a faulty setup, but it only affected the fundamental measurement, the fundamental didn't have combing in it, only the measurement have combing in it.

It doesn't affect at all the distortion measurements of the 2nd 3rd 4th and 5th harmonic. Zaph corrected that problem down the road, so you don't see it on newer tweeters tested

 

[simon5]

Zaph looks at the measurements only, not to the sound produced by the final product.

False, Zaph listen to all tweeters. He just prefer most of the time to find objective measurements to backup what he thinks instead of subjective ones.

Often, you can read his subjective thinking in his blog, shortly after he tests a new tweeter.

 

[Pan]

Zaph looks at the measurements only, not to the sound produced by the final product. C23 is a good tweeter, better than it`s cheaper equivalent the KE25SC ( which is also a very good product ). My opinion is it is not better than the 2904/7000. A friend of mine built a pair of monitor speakers using the C23 and a ceramic cone midbass and a few weeks ago he said that will throw them through his window...I listened to them, I know they are not my tweeter. 🙂

It's a fine tweeter and that anecdote gives an indication of pilot error. Also some people (very common) blame tweeters for upper mid problems either due to distortion from the mid or poor implementation/frequency response.


/Peter

 

[Pan]

It's a shame that he uses that mic for distortion tests as it has high distortion.

The measurements should be taken with a healthy doze of salt because of that.

The frequency response is often ok but with a 1-1.5dB hump at 10-15k though.


/Peter

 

[Jay_WJ]

Originally posted by smellygas Are you suggesting that all tweeters sound the same, as long as they are crossed over above 4kHz with 4th order rolloff (and have reasonably flat FR)???

Definitely, not. There're no two tweeters with different motors and dome structure that have perfectly the same linear/nonlinear distortion profile. That's why they all sound different. What I meant is, if you are to cross tweeters with a LR4 or higher order filter at 4 kHz or higher, there's no justifiable reason to use a $300 tweeter. It will not necessarily sound better in any technically meaninglful sense.

BTW, if you want to cross above 3 kHz with LR4, look no further than the Seas 22TAFG. The lowest possible distortion and better dispersion than 1" domes.

Was this quote published in a peer-reviewed journal or written in a textbook to reflect a conensus opinion? If so, I'd be interested.

If you value opinions of experienced speaker designers . . . For example, post yourself the same quote at PE Tech Talk board. And see what experienced DIYers who frequent the board, like John Kreskovsky, Jeff Bagby, John Krutke, Mark Krawied, David Ralph, Pete Schmacher, etc will say to you.

-jAy

 

[Jay_WJ]

It's a shame that he uses that mic for distortion tests as it has high distortion.

The measurements should be taken with a healthy doze of salt because of that.

The frequency response is often ok but with a 1-1.5dB hump at 10-15k though.


/Peter

He should know better than you. 🙂 His mic is calibrated regularly, and the distortion/noise floor in all the signal chane of his measurement setup has been checked up.

 

[Jay_WJ]

Ring radiators do have an advantage over conventional tweeters and this is in the radiating area. The total radiating area of a dual concentric tweeter may be up to 2.5 times the one of a regular dome.

Uh oh, exactly the opposite. The Sd of the 7000 and the XT25 1 inch ring radiators is smaller than 1" domes. For example, see the spec sheets of the 7000 and the 7100:

http://www.tymphany.com/r2904_700000
http://www.tymphany.com/d2904_710002

5.6 sq cm versus 7 sq cm.

-jAy

 

[Pan]

He should know better than you. 🙂 His mic is calibrated regularly, and the distortion/noise floor in all the signal chane of his measurement setup has been checked up.

I have measured the distortion of the mic and I know it has poor performance. Calibration does not deal with nonlinear distortion.

The stock ECM8000 is not suitable for distortion measurements of high performance speakers.

If he have modified the mic and verified performance linear enough it is good of course and I would love to see the method and results.


/Peter

 

[Jay_WJ]

Did you miss the C23? It has lower distortion on my screen if I read the graphs right.

Anyone knows what's up with the comblike patterns on some of Zaph's measurements?

Also does someone know what microphone Zaph uses for his tests?

At the level and distance he use the contribution from the mic can start to enter the picture.


/Peter

I didn't miss the C23. Its 3rd products can't match the SS. Slightly higher 2nd order productt are not offensive at all.

Zaph uses nearfield measurment for the tweeter distortion test. So, virtually no contribution of the mic. The 90 dB / 1 m said in the web page is a converted SPL to 1 m distance.

-jAy

 

[Jay_WJ]

I have measured the distortion of the mic and I know it has poor performance. Calibration does not deal with nonlinear distortion.

The stock ECM8000 is not suitable for distortion measurements of high performance speakers.

If he have modified the mic and verified performance linear enough it is good of course and I would love to see the method and results.


/Peter

He knows that, too, Peter. He is one of the most meticulous persons I've known in this hobby.

 

[Pan]

I didn't miss the C23. Its 3rd products can't match the SS. Slightly higher 2nd order productt are not offensive at all.

Nonlinearities give IMD so the assumption that 2nd are better than 3rd is slightly flawed.

Zaph uses nearfield measurment for the tweeter distortion test. So, virtually no contribution of the mic. The 90 dB / 1 m said in the web page is a converted SPL to 1 m distance.

It's the other way around. For a given output of the tweeter, mic distortion will be higher the closer to the source the capsule is.


90dB Spl @ 1m translates to aprox. 102dB SPL @ 0.25m. Mic distortion will be a factor.


/Peter

 

[Pan]

He knows that, too, Peter. He is one of the most meticulous persons I've known in this hobby.

Most persons overlook this fact and I have been participating on the boards for a long time and have "talked" to Zaph way back at the Madisound board. I have never heard anything about modification to the mic or verification of the distortion.

It would be intersting to know how this was done if it was done.


/Peter

 

[Jay_WJ]

Nonlinearities give IMD so the assumption that 2nd are better than 3rd is slightly flawed.

Note that Zaph's HD sweeps don't compensate for a tweeter's natural rolloff. Given this, the difference between C23 and 6600's 3rd order distortion levels at 2 kHz and below is night and day.

It's the other way around. For a given output of the tweeter, mic distortion will be higher the closer to the source the capsule is. 90dB Spl @ 1m translates to aprox. 102dB SPL @ 0.25m. Mic distortion will be a factor.

Zaph is aware of the distortion profile of his mic. The ECM8000's distortion at high SPL is commonly known. It's not only you. That's one of the reasons why Zaph chose different, optimal settings for different driver groups so that the mic distortion becomes a non-issue. Also note that he uses exactly the same settings within a driver group for consistent comparision to be valid within that group. The result would be still meaningfully interpretable even if the same mic distortion profile were added in a consistent manner. And I've seen several distortion measurements of the same drivers done by people using the ECM8000. They have all looked consistent with each other.

 

[smellygas]

What I meant is, if you are to cross tweeters with a LR4 or higher order filter at 4 kHz or higher, there's no justifiable reason to use a $300 tweeter. It will not necessarily sound better in any technically meaninglful sense.

Have you actually tried this? I have. In one of my setups, I swapped a SS D2904/6000 dome with a modified 7000. 3kHz crossover with 48db/oct DSP slope. Profound improvement. Mostly because the D2904 seems to roll off in the extreme highs. This is also well-described. I have also, just for fun, connected the SS 9800 revelator in leiu of the tweeter in my B&W CDM-1's. (4kHz, 18db/oct I think) Also, profound improvement. You should try it and see if you can't tell differences in quality. Honestly, if tweeters didn't affect sound quality when crossed over >4kHz, don't you think every loudspeaker manufacturer out there would be designing 3-ways with a $20 tweeter and a midrange that goes up to 4khz??

If you value opinions of experienced speaker designers . . . For example, post yourself the same quote at PE Tech Talk board. And see what experienced DIYers who frequent the board, like John Kreskovsky, Jeff Bagby, John Krutke, Mark Krawied, David Ralph, Pete Schmacher, etc will say to you.

-jAy [/B]

Listen, if distortion products of the 1-2.5kHz range were the defining characteristic of tweeter sound quality, then this phenomenon could easily be tested and verified. Easily and cheaply. Which means there would be lots of papers on it and it should be all over the books. Truth by assertion by experienced hobbiests who post on internet forums doesn't quite carry the same weight for me. I do appreciate your perspective on this, though.

SG

 

[Pan]

Note that Zaph's HD sweeps don't compensate for a tweeter's natural rolloff. Given this, the difference between C23 and 6600's 3rd order distortion levels at 2 kHz and below is night and day.

But 6600 still is more nonlinear which leads to more distortion playing wide band material. Third order only means a more symetrical suspension and motor. At 2kHz the 6600 has about 10dB more distortion.

Zaph is aware of the distortion profile of his mic.

How do you know that?

That's one of the reasons why Zaph chose different, optimal settings for different driver groups so that the mic distortion becomes a non-issue.

But you are mistaken when you believe that this is the case and that a closer placement of the mic makes the situation better.. when it's the opposite.

Also note that he uses exactly the same settings within a driver group for consistent comparision to be valid within that group.

But it does not matter since distortion can cancel or add depending on level and phase.

The result would be still meaningfully interpretable even if the same mic distortion profile were added in a consistent manner.

Nope!

And I've seen several distortion measurements of the same drivers done by people using the ECM8000. They have all looked consistent with each other.

Well I guess they all modified the mic for lower distortion then since the results are identical.. ;-)

Ok, it's no meaning debating this since we don't know if Zaph did improve his mic (but I assume he did not) so please digest what I have explained to you and maybe we can take it up later when we got more meat?

Cheers!


/Peter

 

[simon5]

Have you actually tried this? I have. In one of my setups, I swapped a SS D2904/6000 dome with a modified 7000. 3kHz crossover with 48db/oct DSP slope. Profound improvement. Mostly because the D2904 seems to roll off in the extreme highs. This is also well-described. I have also, just for fun, connected the SS 9800 revelator in leiu of the tweeter in my B&W CDM-1's. (4kHz, 18db/oct I think) Also, profound improvement.

SG

You can't swap a tweeter with another just like that. The crossover need to be optimized, it needs to be tweaked until it sums perfectly and the levels need to be matched perfectly.

Your profound improvement could be just because the modified 7000 and the 9800 are slightly louder than CDM-1 and 6600. About 0.5 dB louder can play with your mind. Some even say 0.1 dB...

You didn't play with the L-Pad, nor the x-over, your tests are pointless, sorry. Please calibrate everything and report back.