Crossover design

[VDS]

Hi, in order to lower crossover point from 600 to 4-500 would I change inductor from 2.5 mh to 3ish mh?  Would I need to adjust capacitor also (currently 20uf)?

[VDS]

.

[Maz4bz]

3 hours ago, VDS said:

how important XO’s are to overall sound, 

 

The crossover is EVERYTHING to the final sound.

 

As Dr Geddes says, drivers are commodity products, and that he can make any sound relatively great with a well designed crossover.

 

Don't kid yourself that you can change the crossover effectively without measurements. This is like trying to drive a car with the windscreen blanked out. You can operate the contolls but you have no real way of knowing where you are going!

 

Get some measurement gear. Post your measurements. There are many here that can help you once you start posting data and being scientific.

 

Hang in there it's well worth the effort.  

 

 

[VDS]

2 minutes ago, Maz4bz said:

 

The crossover is EVERYTHING to the final sound.

 

As Dr Geddes says, drivers are commodity products, and that he can make any sound relatively great with a well designed crossover.

 

Don't kid yourself that you can change the crossover effectively without measurements. This is like trying to drive a car with the windscreen blanked out. You can operate the contolls but you have no real way of knowing where you are going!

 

Get some measurement gear. Post your measurements. There are many here that can help you once you start posting data and being scientific.

 

Hang in there it's well worth the effort.  

 

 

Great analogy! I agree, I wish I had measurements, it would be easier to get advice.  Umik at Parts Express? 
Ted

[Deang]

3 hours ago, VDS said:

Don't kid yourself that you can change the crossover effectively without measurements.

 

The Type A, B, and E have four elements. They all use the same low pass coil, autotransformer, and tweeter capacitor. The only thing that changes is the value of the primary high pass capacitor, which changes based on crossover point and reflected impedance through the autotransformer. So, you can keep it simple. It’s much easier to get away with when the system has inherently low distortion and linear operation. What you give up is an even power response and smoother sound due to lower distortion at high volume levels. The simple filters work well with most non-CD horns.

 

We know a lot about the stock drivers, horns, and boxes. They’ve been measured ad nauseum. There is now good data on both the DE-120 and DCM-50 (and several other drivers). What we don’t have are measurements of the DCM-50 on the K-510, but we do know it’s a CD horn, and most of these type horns exhibit reduced output beginning at around 2kHz, and are generally rolling off at 6dB/octave. They are not the ideal candidates for these type filters (even the ALK types) due to the lack of properly applied EQ. This is where measuring and design software comes in.

[Deang]

6 hours ago, VDS said:

Hi, in order to lower crossover point from 600 to 4-500 would I change inductor from 2.5 mh to 3ish mh?  Would I need to adjust capacitor also (currently 20uf)?

The idea was to get you in the ballpark. You shouldn’t evaluate any recommended change until you get them properly placed. 
 

With the 1.5uF capacitor on Tap 1, you are at 800Hz, which is about a full octave above the Fc of the horn. Output is roughly 97dB. With a 3uF capacitor, you are too close to the Fc of the horn, and the driver may not be loading properly, which will increase distortion.

[Coytee]

18 hours ago, VDS said:

 I replaced k600, with a k510

 

I don't know much about crossover recipe's....but have two questions....

 

Since he's got the K510, wouldn't he need to tilt the response curve to fit it and, can this be done correctly with a passive?

 

VDS, if you have the K510 (and I don't know anything about your driver)  If your driver is similar to the K69/691, what about going active and skipping the tweeter?

 

Back to the shadows I go.

 

 

[Deang]

1 hour ago, Coytee said:

Since he's got the K-510, wouldn't he need to tilt the response curve to fit it and, can this be done correctly with a passive?

 

Last part first: Since Roy designed a passive for the K-510 Jubilee, you should know the answer to that.

 

First part last: Yes. What he's doing isn't ideal, which I alluded to. However, he's gone three-way, and he can cheat it a bit. He should for sure get a microphone and REW and measure.

[VDS]

Umik ordered today, well see what’s going on. It will be interesting to see quantifiable info about the sound in order to make improvements. Stay tuned 

[VDS]

3 hours ago, Deang said:

 

Last part first: Since Roy designed a passive for the K-510 Jubilee, you should know the answer to that.

 

First part last: Yes. What he's doing isn't ideal, which I alluded to. However, he's gone three-way, and he can cheat it a bit. He should for sure get a microphone and REW and measure.

Dean, any horn suggestion, don’t say k600! Lol... is horn performance something I will be able to see when my mic arrives? Probably a dumb question I know.  I tried to ask for horns that have improved Cornwalls, I thought after 50 years there would be some consensus, didn’t seem to be.  

[PrestonTom]

In a very friendly way I am going to make some very honest statements.

 

1) crossovers are not easy to design. The pros spend a good deal of time setting up a crossover (in spite of having education, training, software, experience, to do this). 2) what makes a speaker sound good is not just the crossover,  it is also the dispersion and frequency response of the system (see Toole for a nicely presented discussion of this). 3) Constant dispersion horns sound great but they also require a high-end boost. Equalizing with passive components usually means attenuating the low end in order to "boost" the high end. This is required for CD horns (such as the K-510). It can be done, but do you want to throw away the efficiency?. 4) Using horns frequently requires time alignment and this is not easily done with passive components. 

 

 Where does this leave us. 1) you will need measurement equipment, 2) you will need knowledge about these passive components (the software will only give you an approximation) 3) very importantly, you will need to design toward a goal. This last point has some subtlety. Toole among others has emphasized the importance of a smooth frequency response both on -axis and off-axis. Will your network provide that and will you be comfortable trying to measure that? It is not easy stuff. Defining the goal is essential. 

 

Trying to do this with passive components is expensive. Trying to do this without measurement and relying on "wait ... it sounds good to me", is a fool's errand and you will only go in circles. You might want to consider an active solution (DSP and multiple amps). It will save money in the long run, but you are still faced with having to define the "goal." Understanding the goal will be helped by a careful reading of Toole's book (this is just one perspective). 

 

My friendly suggestion is if you are willing to do the homework (and I don't know what your technical background is) it may be to use a tried and true solution by someone who has a history of competence in this work. IOW, it ain't easy stuff. I believe the Klipsch offering are well-designed. I believe some of the Crites suggestions are well-designed (with his specific choices on drivers and crossovers). I can not speak for the others - they may or may not be. 

 

These are "System Issues" and it is not a matter of swapping drivers in and out. It requires a system design - approach. I would suggest going with a proven system  design. 

 

I know this may not be what you want to hear, but many folks underestimate what is involved. 

 

Good luck,

-Tom

[VDS]

54 minutes ago, PrestonTom said:

In a very friendly way I am going to make some very honest statements.

 

1) crossovers are not easy to design. The pros spend a good deal of time setting up a crossover (in spite of having education, training, software, experience, to do this). 2) what makes a speaker sound good is not just the crossover,  it is also the dispersion and frequency response of the system (see Toole for a nicely presented discussion of this). 3) Constant dispersion horns sound great but they also require a high-end boost. Equalizing with passive components usually means attenuating the low end in order to "boost" the high end. This is required for CD horns (such as the K-510). It can be done, but do you want to throw away the efficiency?. 4) Using horns frequently requires time alignment and this is not easily done with passive components. 

 

 Where does this leave us. 1) you will need measurement equipment, 2) you will need knowledge about these passive components (the software will only give you an approximation) 3) very importantly, you will need to design toward a goal. This last point has some subtlety. Toole among others has emphasized the importance of a smooth frequency response both on -axis and off-axis. Will your network provide that and will you be comfortable trying to measure that? It is not easy stuff. Defining the goal is essential. 

 

Trying to do this with passive components is expensive. Trying to do this without measurement and relying on "wait ... it sounds good to me", is a fool's errand and you will only go in circles. You might want to consider an active solution (DSP and multiple amps). It will save money in the long run, but you are still faced with having to define the "goal." Understanding the goal will be helped by a careful reading of Toole's book (this is just one perspective). 

 

My friendly suggestion is if you are willing to do the homework (and I don't know what your technical background is) it may be to use a tried and true solution by someone who has a history of competence in this work. IOW, it ain't easy stuff. I believe the Klipsch offering are well-designed. I believe some of the Crites suggestions are well-designed (with his specific choices on drivers and crossovers). I can not speak for the others - they may or may not be. 

 

These are "System Issues" and it is not a matter of swapping drivers in and out. It requires a system design - approach. I would suggest going with a proven system  design. 

 

I know this may not be what you want to hear, but many folks underestimate what is involved. 

 

Good luck,

-Tom

Tom, thanks, that’s exactly how I feel, I started with the attitude of throwing drivers in and calling it an upgrade.  But I know I have 3 good drivers, I’m open to changing mid horn,  now I’m gearing up to slowly learning about crossovers to maximize the potential of my 3 drivers .

And even though I know they can get much better with careful electronics, they already sound good and enjoyable, but I want them to be startling good.  I’ll just keep moving forward smarter, with measurements instead of ,as you say, going in circles,  without a method.

im always open to hear constructive criticism and experienced perspective!

[PrestonTom]

Okay, let me clarify, I did not say "without a method". Since this also  that is necessary. My emphasis was "without a goal". That is even more critical. 

Good luck,

-Tom

[VDS]

1 minute ago, PrestonTom said:

Okay, let me clarify, I did not say "without a method" that is necessary. My emphasis was "without a goal". That is even more critical. 

Good luck,

-Tom

That’s a point I’m learning, I’m trying to be more critical and develop goals, like most of us I just started with wanting them to “sound great” now I’m trying to understand   What that means to me, and how my speakers are falling short of that goal

[VDS]

12 hours ago, PrestonTom said:

Okay, let me clarify, I did not say "without a method". Since the is also  that is necessary. My emphasis was "without a goal". That is even more critical. 

Good luck,

-Tom

Preston Tom, or anybody,  

regarding horn selection, I’m very baffled, basically, how does one choose a horn? Is it to emphasize certain driver characteristics,  add certain coloring?  I seem to understand, say, if you want to use a mid driver down to 4-500hz you need to pick a horn that will handle that frequency without getting all “wonky”.

i mean, I hear a lot of praise fro elliptical horns, Dave Harris Fastract(?), Greg’s Volti horn, are there situations where this elliptical configuration is better suited and situations where it’s not?

Is there any general agreement on all around 2” horns for home 2 channel, given that the drivers spec is 400-10,000hz?  
 is this way to general of a question? Horn performance seems immensely complex, how much of that translates to audible differences?

thanks, the knowledge of this group is indispensable! Ted

[Coytee]

I'm curious if you're using the K510 have you considered taking the speaker to a 2-way? 

 

 

[PrestonTom]

If your goal is to have a system that is relatively CD (controlled dispersion) then Roy's modified tractrix horns will do this (note that not all "tractrix" horns are CD - the "modified tractrix" ones from Klipsch are CD). Two obvious choices are the K-402 or the K-510. Other horns may or may not have well-measured dispersion patterns available (if they were even measured at all). 

 

A reading of Roy's occasional posts will highlight the advantages of this goal. A reading of Toole will provide a needed background and lead to a more developed and detailed explanation about the importance of CD (along with some other issues).

 

Fundamental to the design is a careful decision on whether the system should be 2-way or 3-way (with all the pros and cons of each). The crossover point(s) and steepness are a consequence of the the dispersion, amplitude & phase response, and distortion. 

 

My general view is that most forget about the issues of distortion and dispersion and set their crossover points too low. Also keep in mind that setting up a crossover is difficult!. Time alignment and CD horns can make the task easier (but not trivial). Trying to do this with analog components will be filled with difficulties and compromises. In the end, developing an analog crossover can burn up a good deal of cash. I hope I have given a strong hint to think about a DSP solution. 

 

Good luck,

-Tom

[VDS]

Have not really considered 2 way, I’m sure he crossover would be easier.  At this point I’m open to all ideas.  I want to lean more about active crossovers, it theoretically seems like a better, more flexible driver control system.  My main concern is to now measure what I have, possibly change horns if it’s warranted, and increase the cohesiveness with a passive crossover.  Then I’d like to build a set from scratch and possibly explore active.  I feel like I want to “finish” this Cornwall project by getting it to a high level, but possibly not as high as possible, and the start fresh and possibly try a 2 way active as a first from scratch project. 

If I can get them to sound at least noticeable better by measuring and modifying the Crites b2, based on those measurements, that may be all I can hope to achieve this round.  I can always come back and retweek later when I have more experience and knowledge. 

[Coytee]

20 minutes ago, VDS said:

 I want to lean more about active crossovers, it theoretically seems like a better, more flexible driver control system.

 

Count me as a fan of actives.....so, with that said.

 

There was a nice chap, Rigma who had a pair of Jubilee's (2-way) and used a passive crossover.  He spent about $3,000 in parts alone to make a really primo crossover....basically, spared no expense.

 

Little while later, he upgraded his stock K69 driver for a TAD-4002 driver....which meant he had to re-do his crossover.  As I recall, he spent another $3,000 for new parts and made a no expense spared crossover for the new setup.

 

As it turns out, I had the same speakers....with the same K69 driver.....  I had an active, cost me about $1,250 new as I recall.  Dialed things in and it sounded nice.  I ended up also swapping out the stock drivers for a pair of TAD-4002 drivers....yet all I had to do was go to my active and dial in the new parameters. 

 

So, you have $6,000 in parts to get to his "end game" verses my $1,250 to get to my "end game" with the same speaker.

 

"To each his own" and all that fine stuff....  He sits and focuses more on his sound than I do....consequently, I'll take that $4,750 difference as cash in my pocket any day.  I also get to adjust the delay to better align the whole speaker....  he doesn't get to do that.

 

[Coytee]

Here are his works of art.....took two of us to carry a single unit in (I wasn't going to be responsible for carrying it, tripping and damaging it!!)

 

I think it weights something like 40 pounds (??)

 

Also, I think it's on a board that approaches 23" x 36"