Klippel: SB Acoustics 17NRXC35-4

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

Just another aspect not mentioned for pushing the xover point as low as possible: in a 2 way woofer's distortion also rises with frequency, because of Le(x) and Le(i) caused distortion and made worse by being modulated with excursion and current from low notes. So the levels of distortion from tweeter maybe better than the levels of distortion from woofer at the xover point selected.

Excellent point, Feyz. You're right, in many cases the midwoofer has higher harmonic distortion at 2khz than the tweeter has. I think this point is often missed as well. The bottom line is that the distortion argument against crossing tweeter in this range just doesn't hold up in most small two-way speakers.

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

I touched on this in my reply above with the comment abut decreased polar of mid woofers at this range. I also understand that plenty of people cross tweeters at 2khz. I still don't think it makes my statement regarding increased distortion any less valid. Yes, some are better than others, but a cursory look at available data for a fair amount of tweeters verifies the notion. The fact that most mids have higher distortion than a tweeter crossed low doesn't make the higher distortion of a low crossed tweeter any more arbitrary. It is something that may or may not be an issue in a particular design. That's all I'm saying.

That said, I'm not going to argue about it. IMO, you do what you need. I don't have extensive home audio builds in my resume but I have bought and tested a lot of various drivers for car audio application and my own personal goals usually mean the tweeter is crossed above where some very nice home audio designs are. I've actually left a gap in the 2-4khz range to some varying degree. We do what works best for our goal. Honestly, the polar response and my issue with low crossing tweeters in 2-ways is why I've typically used a 3-way setup in my car where on axis aiming is a rarity.

As I am sure you are aware, listener and speaker placement, as well as the environment itself are very different in car audio; so much so, it is difficult to translate well between the design goals of the two.

Here's where I'm coming from - I believe distortion data, although an important piece of information, is highly misinterpretted on the internet in many of the threads I read. Speaker design is loaded with compromises and tradeoffs, we begin with the tradeoffs of Hoffman's Iron Law and go from there.... What we have drifted into here in this discussion is one of those: The tradeoff between tweeter distortion (of a well designed tweeter) and a low crossover point. What is seldom discussed is the relative audibility of what lies on both sides of the tradeoff. In this case the audibility of the effects of lobing error, poor polar response, and the hash of a midwoofer cone break-up far exceed the audibility of the distortion of an RS28 at 2kHz. If the scale is the audible effects of the design choices then the balance tips squarely in one direction.

I know there is another thread on why people cross tweeters low. That doesn't disqualify some discussion here. The question was raised of the usability of the SB17 and the difficultly in crossing it over due to its frequency response. That's a fair question in this thread. I simply tried to answer from my experience with the driver (the 8 ohm version), and that led to this discussion, which is just as relavent. Threads do this sort of thing, so do conversations ;).

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

My last DIY home system had horn loaded 15" midbasses and horn loaded compression drivers up top...so you can understand why I have a hard time believing 6s with a dome tweeter crossed near where my horns were crossed could reach the same levels with the same dynamics that I am used to having.

Obviosuly, that's a very different design goal. Neither a dome tweeter, nor the SB17, nor any small two-way will have an output level that would compare.

[ErinH]

Re: Klippel: SB Acoustics 17NRXC35-4

If we can, let's try to stay on topic. I believe there is a thread currently up regarding tweeters crossed low. Partly my fault for engaging the subject.

[christianb]

Re: Klippel: SB Acoustics 17NRXC35-4

My last DIY home system had horn loaded 15" midbasses and horn loaded compression drivers up top...so you can understand why I have a hard time believing 6s with a dome tweeter crossed near where my horns were crossed could reach the same levels with the same dynamics that I am used to having.

No one ever said they would reach the same levels as horn-loaded 15's with compression drivers. Are you saying these are a requirement for non-background music?

[ErinH]

Re: Klippel: SB Acoustics 17NRXC35-4

Actually you need the wide directivity of the tweeter near the crossover point If you are not wanting to have a dip in this region of the power response due to the narrowing directivity of the woofer. Less lobing error at the crossover along the vertical axis, and more uniform polar response along the horizontal axis are both generally considered desirable characteristics. A higher crossover point would drive both of these in the wrong direction.

I touched on this in my reply above with the comment abut decreased polar of mid woofers at this range. I also understand that plenty of people cross tweeters at 2khz. I still don't think it makes my statement regarding increased distortion any less valid. Yes, some are better than others, but a cursory look at available data for a fair amount of tweeters verifies the notion. The fact that most mids have higher distortion than a tweeter crossed low doesn't make the higher distortion of a low crossed tweeter any more arbitrary. It is something that may or may not be an issue in a particular design. That's all I'm saying.

That said, I'm not going to argue about it. IMO, you do what you need. I don't have extensive home audio builds in my resume but I have bought and tested a lot of various drivers for car audio application and my own personal goals usually mean the tweeter is crossed above where some very nice home audio designs are. I've actually left a gap in the 2-4khz range to some varying degree. We do what works best for our goal. Honestly, the polar response and my issue with low crossing tweeters in 2-ways is why I've typically used a 3-way setup in my car where on axis aiming is a rarity.

[winslow]

Re: Klippel: SB Acoustics 17NRXC35-4

My last DIY home system had horn loaded 15" midbasses and horn loaded compression drivers up top...so you can understand why I have a hard time believing 6s with a dome tweeter crossed near where my horns were crossed could reach the same levels with the same dynamics that I am used to having.

[dlr]

Re: Klippel: SB Acoustics 17NRXC35-4

I don't think my Continuum, Piccolo, Mandolin, or Dreydel sound spitty, and I doubt any of the dozens of builders of these speakers do either. Maybe some of them can chime in with their long-term impressions. They are all example of what I am talking about.

Actually you need the wide directivity of the tweeter near the crossover point If you are not wanting to have a dip in this region of the power response due to the narrowing directivity of the woofer. Less lobing error at the crossover along the vertical axis, and more uniform polar response along the horizontal axis are both generally considered desirable characteristics. A higher crossover point would drive both of these in the wrong direction.

I fully agree with you Jeff. I'm crossing the SB17 with the original SB25 at 2K. It was well received at DIY NE last year. I'll have it up again this year and will probably buy the SB26 to update it for the current version.

With the UE I'm able to cross the DXT in my 3-way LR8 @1200. I'll have that up again at DIY NE this year, so we'll see what others think of it. I've been more than happy with it. This also provides a very smooth polar response.

dlr

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

Except that used like that (assuming a 180deg waveguide, aka flush mounting) they will all sound unacceptably spitty to someone accustomed to the sounds of unamp'ed live music and high-fidelity reproduction.

Properly used (i.e. with some means of controlling their low-end directivity) you're probably right.

I don't think my Continuum, Piccolo, Mandolin, or Dreydel sound spitty, and I doubt any of the dozens of builders of these speakers do either. Maybe some of them can chime in with their long-term impressions. They are all example of what I am talking about.

Actually you need the wide directivity of the tweeter near the crossover point If you are not wanting to have a dip in this region of the power response due to the narrowing directivity of the woofer. Less lobing error at the crossover along the vertical axis, and more uniform polar response along the horizontal axis are both generally considered desirable characteristics. A higher crossover point would drive both of these in the wrong direction.

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

I want live play back type levels. Such a setup wouldn't do it.

If you are ever in the vicinity, you are welcome to drop by. Ask Wolf, I can do live playback levels just fine. :D

My point is still the same. In a two-way with a midwoofer the tweeter is not the controlling element for output. It's the midwoofer.

[Pallas]

Re: Klippel: SB Acoustics 17NRXC35-4

You guys are making it out to be much more difficult than it really is - the SB29, Usher 9930 & 9950, RS28F, RS28A, Seas 27TDFC, will all cross easily below 2khz with low distortion at a reasonable price; not to mention the numerous higher end Scanspeak and Seas tweeters.

Except that used like that (assuming a 180deg waveguide, aka flush mounting) they will all sound unacceptably spitty to someone accustomed to the sounds of unamp'ed live music and high-fidelity reproduction.

Properly used (i.e. with some means of controlling their low-end directivity) you're probably right.

[internova]

Re: Klippel: SB Acoustics 17NRXC35-4

What would be the problem if I want to cross this woofer at 2.5K? I think the break up in this case could still be well controlled. The off-axis responses of the woofer isn't that bad either.

[winslow]

Re: Klippel: SB Acoustics 17NRXC35-4

I want live play back type levels. Such a setup wouldn't do it.

[Jeff B.]

Re: Klippel: SB Acoustics 17NRXC35-4

I understand where Winslow is coming from.
Many tweeters have a rising distortion trend typically below 4khz that is pretty rough at 2khz. This can be seen on Zaph's testing.

I'm not saying you can't or shouldnt cross below 4khz. I am saying that, in general, it's dicey here down. Of course, generally, I don't make general statements. ;)

It's tough... The polar of 6-7" mids starts going bad in this area so you need a tweeter that can make it that low. Finding one is tough for listening at higher volume due to distortion.

You guys are making it out to be much more difficult than it really is - the SB29, Usher 9930 & 9950, RS28F, RS28A, Seas 27TDFC, will all cross easily below 2khz with low distortion at a reasonable price; not to mention the numerous higher end Scanspeak and Seas tweeters.

I stand by my statement - in a two-way loudspeaker the midwoofer would easily by overdriven long before any of the tweeters I just named if they are crossed over properly. (All of which I have actually used and crossed over below 2khz).

I have to disagree, finding one is not tough when the woofer is the limiting component.

[ErinH]

Re: Klippel: SB Acoustics 17NRXC35-4

I understand where Winslow is coming from.
Many tweeters have a rising distortion trend typically below 4khz that is pretty rough at 2khz. This can be seen on Zaph's testing.

I'm not saying you can't or shouldnt cross below 4khz. I am saying that, in general, it's dicey here down. Of course, generally, I don't make general statements. ;)

It's tough... The polar of 6-7" mids starts going bad in this area so you need a tweeter that can make it that low. Finding one is tough for listening at higher volume due to distortion.