Matching Woofers and Tweeters

0 Members and 1 Guest are viewing this topic. Read 14238 times.

sts9fan

Matching Woofers and Tweeters
« on: 13 Mar 2012, 05:33 pm »
I am looking at a 2way project and have been selecting drivers.  What do you think are the important attributes when pairing a tweeter and woofer?
Frequency response?
Impedance?
Sensitivity?

Tyson

  • Full Member
  • Posts: 11481
  • Without music, life would be a mistake.
Re: Matching Woofers and Tweeters
« Reply #1 on: 13 Mar 2012, 05:45 pm »
How low the tweeter can cross over, and how much beaming the woofer has at that point.

S Clark

  • Full Member
  • Posts: 7464
  • a riot is the language of the unheard- Dr. King
Re: Matching Woofers and Tweeters
« Reply #2 on: 13 Mar 2012, 06:02 pm »
What are your goals?  Is simplicity important?  How will you do measurements?  When assigning speaker projects for my physics classes, students had to fill out a questionnaire.  Then I could assign woofers and tweeters that best matched what they wanted to accomplish.  We need more info to make meaningful recommendations.

Scott

sts9fan

Re: Matching Woofers and Tweeters
« Reply #3 on: 13 Mar 2012, 06:08 pm »
I am using a SEAS Prestige L18RNX/P (H1224) 7" Aluminum Cone Woofer in a 0.5ft^3 box.  I am not really lookin for specific recs.  More just wanted to see what people think are the important synergies.

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #4 on: 13 Mar 2012, 06:17 pm »
You might want to check out the Parts Express Forum. Loads of interesting designs discussed over there. In particular there are a lot of interesting waveguide designs which allow you to cross over the tweeter mighty low (800-1200 Hz), provide great off-axis response and are nice low distortion designs (since the woofer doesn't have to go much higher than 1200 Hz).

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #5 on: 13 Mar 2012, 06:20 pm »
If you are using that woofer, you might want to look into the Aether Audio (formerly SP Tech) waveguide designs. They used the SEAS L18RNX/P in their SP Tech Minis, which I have and love. Think the tweeter used may have changed over time in the SP tech designs but maybe you can look into what that combo has been.

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #6 on: 13 Mar 2012, 06:27 pm »
Looking back at my notes, one of the tweeters used in the SP Tech Minis at one point was the SS9300.

sts9fan

Re: Matching Woofers and Tweeters
« Reply #7 on: 13 Mar 2012, 06:36 pm »
I am not using a waveguide.  Again I am looking for theory.  Its easy to look up what is paired with what.  I am looking personal whys.

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #8 on: 13 Mar 2012, 06:45 pm »
Fair enough. For the theory side of things:

•Don't ignore speaker distortion (often called one of audio's "dirty little secrets").

One way of reducing this distortion is to have a pairing or a design (yep like a waveguide for example) that will allow for a low crossover point. This leads to less distortion from your woofer.

•Depending on your listening set up, the off-axis response would be something else I would attend to in a pairing, though that may mainly pertain to the tweeter side of the equation.

OP, hope this is helpful. Even if not to you, perhaps to others reading this thread looking for important principles of speaker design.

EthanH

Re: Matching Woofers and Tweeters
« Reply #9 on: 13 Mar 2012, 07:12 pm »
I am not very experienced in speaker design at all, but to my knowledge metal cone woofers like the L18 demand particular attention be paid to the woofer's distortion profile.  If I recall correctly the L18 has a big peak around 7khz, however, it also has smaller odd-order harmonic distortion peaks lower in frequency (I believe there's a substantial odd order peak around 1.5khz).  So, in my opinion a robust tweeter that is capable of handling a low crossover point (~1.5khz, maybe lower) would be necessary to keep the L18 in it's ideal operating range.  I've seen a few designs for the L18 with crossover points at 1.5khz, 1.7khz, and 2khz, respectively (I own a pair of L18s in a .5 cubic foot box). 

sts9fan

Re: Matching Woofers and Tweeters
« Reply #10 on: 13 Mar 2012, 07:53 pm »
I am planning on addressing those issues.

cheap-Jack

  • Full Member
  • Posts: 760
Re: Matching Woofers and Tweeters
« Reply #11 on: 19 Mar 2012, 02:24 pm »
Hi.

How low the tweeter can cross over, and how much beaming the woofer has at that point.

Human ear hearing sensitivity is at its maxminum at around 3.4KHz. So best to avoid our ears detecting the dales & hills of the audio response of a loudspeaker in box is to avoid having the X-over point located there.

IMO, lower the better as our ears tend to need more power to get same hearing level at lower frequency, plus lower the frequency, less soundwave
beaming our ears can sense.

Like my 2-way KEF bookshelvers, the X-over corner frequency is at 1KHz. The tweeter is large DOME diaphram to provide wide soundwave dispersion.

c-J

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #12 on: 19 Mar 2012, 02:46 pm »
Hi.

Human ear hearing sensitivity is at its maxminum at around 3.4KHz. So best to avoid our ears detecting the dales & hills of the audio response of a loudspeaker in box is to avoid having the X-over point located there.

IMO, lower the better as our ears tend to need more power to get same hearing level at lower frequency, plus lower the frequency, less soundwave
beaming our ears can sense.

c-J

+1

My SP Tech Minis use the L18 and they are crossed over at 800Hz(!).

nodiak

  • Full Member
  • Posts: 1083
Re: Matching Woofers and Tweeters
« Reply #13 on: 19 Mar 2012, 04:03 pm »
Not trying to be pushy or to irritate by reiterating about waveguides. I'm another who has had a lot of 2 ways over the decades and the best improvement I have come across is using a waveguide loaded tweeter to escape the narrow sweet spot and spread even response off axis.
So to your question it's "constant directivity" or "power response" when the drivers are combined. Includes xo point where the mid/woofer isn't heavily beaming and tweeter isn't distorting. As in Geddes approach, Mark K's E18DXT, Jon Marsh's Modula MT EX, Zaph's WGTMM.
Imo it's worth whatever pain to find and use an appropriate waveguide for the tweeter.
I won't personally kid myself that I can design a great xo by ear so unless you have experience and equipment, the challenge is to find one using the drivers you have...or selling and switching to something you feel seems well designed.
Beyond all that... smooth upper roll off of mid/woofer, low fs of tweeter = simpler xo.
I know you're holding back from saying much, but are you designing the xo yourself or looking for other's work. Why that driver, etc. What are you trying to achieve - using sub, etc.
I know the discussions get complicated but I'm interested in what may come of the efforts.

cheap-Jack

  • Full Member
  • Posts: 760
Re: Matching Woofers and Tweeters
« Reply #14 on: 19 Mar 2012, 05:12 pm »
Hi.

.. smooth upper roll off of mid/woofer, low fs of tweeter = simpler xo.

Yes, you have pointed out the key points of loudspeaker system design. The best sounding one, IMO, is to use simple X-over, i.e. lowest filter order possible to mininize phase shift around the corner frequency.

It is always better to find out how a loudspeker system once built, will behave by reading its I/P impedance/frequency response curve. Such curve is not that hard to map with the help of an audio frequency generator, digital voltage meter & a dummy load.
 
Ideally, it should get an impedance curve with lesser dales & hills the better. Needless to say, 2-way will be smoother than 3 way as there will be more impedance resonance peaks with more ways of driver units.

c-J

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #15 on: 19 Mar 2012, 05:28 pm »
Good point about impedance charts, especially if a tube amp is used. And speaker designs are about tradeoffs. So it may be worthwhile to have a more complex crossover in order to achieve a smooth response and lower crossover point (like the 4th order one on the SP Tech/Aether speakers, which is in-phase BTW), or it may be that a higher crossover point is worth it in order to have a simpler first order or series crossover (as seen in some Fritz and Zu speakers, IIRC).

Æ

  • Full Member
  • Posts: 859
Re: Matching Woofers and Tweeters
« Reply #16 on: 19 Mar 2012, 05:35 pm »
Good point about impedance charts, especially if a tube amp is used. And speaker designs are about tradeoffs. So it may be worthwhile to have a more complex crossover in order to achieve a smooth response and lower crossover point (like the 4th order one on the SP Tech/Aether speakers, which is in-phase BTW), or it may be that a higher crossover point is worth it in order to have a simpler first order or series crossover (as seen in some Fritz and Zu speakers, IIRC).

A complex crossover may produce a better frequency response, but it isn't better for the impedance. Multiple component (complex) crossovers makes the impedance highly reactive, which is technically worse for a tube amplifier.
« Last Edit: 19 Mar 2012, 06:42 pm by Æ »

Æ

  • Full Member
  • Posts: 859
Re: Matching Woofers and Tweeters
« Reply #17 on: 19 Mar 2012, 05:42 pm »
I am not using a waveguide.  Again I am looking for theory.  Its easy to look up what is paired with what.  I am looking personal whys.

Generally speaking with a two way loudspeaker, you want a tweeter that can go as low as possible. Typically it would have an Fs below 1Khz. It's not important to match sensitivities because you can always attenuate (pad) the tweeter, but you never want to pad the midwoofer.
If you will be using Baffle Step Correction (for the midwoofer), then you can take that into consideration when choosing the tweeter too. If you were adding 3dB to 6dB of baffle step correction, then you could choose a less sensitive, less efficient tweeter by the same 3dB to 6dB.

Matching the timber of the woofer and tweeter usually has to be done by ear. Or rely on someone else's recommendation.

roscoeiii

Re: Matching Woofers and Tweeters
« Reply #18 on: 19 Mar 2012, 05:50 pm »
A complex crossover my produce a better frequency response, but it isn't better for the impedance. Multiple component (complex) crossovers makes the impedance highly reactive, which is technically worse for a tube amplifier.

No argument there. Wasn't able to dig up an impedance plot, but that is likely one reason that tube amps don't tend to be recommended for the SP Techs (though they pair beautifully with the Butler tube hybrid amp, this pairing is used by a number of SP Tech owners and was used at some of the audio shows that their speakers were shown at).

Ralph at Atma-Sphere also often posts a link to an article he wrote about power vs. voltage paradigms in speaker and amplifier design, which may be useful for readers of this thread. It starts of on amp design characteristics but then gets to the issue of loudspeaker design as it relates to these two paradigms:

http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

Æ

  • Full Member
  • Posts: 859
Re: Matching Woofers and Tweeters
« Reply #19 on: 19 Mar 2012, 06:02 pm »
How low the tweeter can cross over, and how much beaming the woofer has at that point.


The beaming of the woofer is gradual with increasing frequency and highly dependent on the diameter of the cone.
My personal formula, rule of thumb is to use the number 8658 divided by the actual cone diameter in inches. Cone diameter only, not including the entire surround. The number you end up with would be the approximate frequency where beaming would just begin to get noticeable.