The Crossover - Brain of your Loudspeaker System

[croseiv]

Is there a specific characteristic in the sound of your speakers that you don't like?
.

Honestly, no. I was just wondering if the current crossover could be improved upon to produce even better sound.

 

[avaserfi]

Oh BTW, Gluing anything or sticking crap to the driver's diaphragm for Delta mass testing is not very advisable nor accurate. A test box is recommended.

I was able to get extremely accurate, consistent, results using weight on the drivers diaphragm. I used blu-tack to adhere accurately measured weights in an equally distributed fashion upon the drivers diaphragm.

 

[R-Carpenter]

I was able to get extremely accurate, consistent, results using weight on the drivers diaphragm. I used blu-tack to adhere accurately measured weights in an equally distributed fashion upon the drivers diaphragm.

How do you work woofers with Phase Plug?
Dayton Reference for example. I always had problems with drivers like that and ended up building a small test box with interchangeable baffle.

 

[avaserfi]

How do you work woofers with Phase Plug?
Dayton Reference for example. I always had problems with drivers like that and ended up building a small test box with interchangeable baffle.

I attached the weights such that they weren't touching the phase plug. It worked without issue when I measured the RS 150.

 

[matthewdclarkaz]

3 way speakers

I've never understood why companies make 3 way speakers with only 1 crossover point (I mean besides cost.) It would seem to me that you would lose much of the benefit of a mid-range if it was running the same frequencies as your woofer.

Am I way off base on this?

 

[TLS Guy]

I've never understood why companies make 3 way speakers with only 1 crossover point (I mean besides cost.) It would seem to me that you would lose much of the benefit of a mid-range if it was running the same frequencies as your woofer.

Am I way off base on this?

I think you probably have some misunderstandings. There are situations were the low pass filter of a midrange is acoustic. However the woofer will still need a low pass filter. This type of set up is rare.

I suspect you are confused about 2.5 ways. Here the lower driver is a fill driver providing step loss diffraction compensation for narrow fronted cabinets. The crossover between the upper bass mid and the tweeter is then high pass/low pass.

These speakers are very common, and is the usual situation with tweeter on top, two bass mids below. Your confusion probably arises because more often than not only the tweeter/bass mid crossover is specified and not the 3 db point of the low pass filter for the fill driver.

What is happening is that the driver below the tweeter is reproducing bass and midrange, but the lower driver only bass.

Here is a crossover of that type of arrangement.

 

[Seth=L]

The fact your end quote is "you still have a lot to learn" is a pretty good indication you already know more than most of us. Arrogance and pride about what we know is usually a good indication we do not know all that much. Your point that it is easy to screw up an otherwise good system by using a mediocre crossover is taken. That said, driver design is a more difficult and time consuming task than crossover design. I have seen hundreds upon hundreds of bad drivers and crossovers, and I think driver expertise is more uncommon than crossover expertise. What you may be amazed to learn is how often you can conceal and minimize driver and system flaws with a sophisticated crossover. I learned how to do this by building crossovers for very cheap speakers. I do this once per week for one of my clients. I did it today. I designed 7 crossovers in 6 hours. I refer to it as "making Chicken Soup out of Chicken feathers". I keep looking for white meat, and the client keeps bringing me tails....

I honestly thought this was more so the truth. However I, lacking the knowledge, don't really have a lot of data or experience to back this up. I do follow philosophies that usually follow the lines of the most logical or simple scenarios.

Technology revolving around amplifiers hasn't had any "breakthroughs" or how most amplifiers sound identically under most listening situations. Electronic circuits is almost an exact science in those levels of design and implementation. Tackling room acoustics, driver design that will function well in the cabinet and cabinet design itself are all too variable or costly to "put your finger on". Making a linear, truely linear, loudspeaker is no small feat. If it was, everyone would be making them just like everyone can make a receiver.

Technology.....I thought I would just say that word again. Bose uses technology a lot apparently.

 

[Seth=L]

Okay, say the speakers, and cabinet are obviously of good quality, but the crossover seems to be the weakest link. Would any audible benefit come from building a better more robust crossover (I think the answer is yes)?

I would say that is certainly the truth. As I said before each part really has equal "importance". Paul took another subject and expanded on the difficulty of designing x-overs relative to the design off speaker drivers. It seems very likely that the cross-over would be easier, cheaper, and far less time consuming to design than a driver or good cabinet with low as possible resonant affect. Just because the cross-over is easy to design doesn't detract from it's importance.

Debating on which part of the speaker is the "most important" seems to be a fools erand, I don't honestly believe there is a clear choice or that their should be. Saying the cross-over is the most important part of the speaker is akin to saying the engine is the most important part of the car. It's a crucial and significant part to the operation of the finished product, and no part should be discounted when it comes to the function and/or performance of the product.

 

[TLS Guy]

I would say that is certainly the truth. As I said before each part really has equal "importance". Paul took another subject and expanded on the difficulty of designing x-overs relative to the design off speaker drivers. It seems very likely that the cross-over would be easier, cheaper, and far less time consuming to design than a driver or good cabinet with low as possible resonant affect. Just because the cross-over is easy to design doesn't detract from it's importance.

Debating on which part of the speaker is the "most important" seems to be a fools erand, I don't honestly believe there is a clear choice or that their should be. Saying the cross-over is the most important part of the speaker is akin to saying the engine is the most important part of the car. It's a crucial and significant part to the operation of the finished product, and no part should be discounted when it comes to the function and/or performance of the product.

We seem to have gone full circle. Drivers and crossovers are tough to design.

However it is the crossover that generally puts the designers stamp on it and determines the type of sound field the designer wants as long as driver selection has been competent.

 

[matthewdclarkaz]

I think you probably have some misunderstandings. There are situations were the low pass filter of a midrange is acoustic. However the woofer will still need a low pass filter. This type of set up is rare.

I suspect you are confused about 2.5 ways. Here the lower driver is a fill driver providing step loss diffraction compensation for narrow fronted cabinets. The crossover between the upper bass mid and the tweeter is then high pass/low pass.

These speakers are very common, and is the usual situation with tweeter on top, two bass mids below. Your confusion probably arises because more often than not only the tweeter/bass mid crossover is specified and not the 3 db point of the low pass filter for the fill driver.

What is happening is that the driver below the tweeter is reproducing bass and midrange, but the lower driver only bass.

Here is a crossover of that type of arrangement.

I think I got the gist of that. Basically even though they don't list it, the two drivers are actually handling different ranges? Is that right?

 

[R-Carpenter]

I think I got the gist of that. Basically even though they don't list it, the two drivers are actually handling different ranges? Is that right?

Correct. Drivers have their own natural Frequency response that is also effected by the box they are put in and ELECTRIC crossover put behind it.
It is difficult but not impossible to use drivers natural roll off (acoustic crossover) to combine it with other drivers. Of cause sound purists will adore it but as well as with anything else in speaker design, there's no free lunch.

In a 2.5 system one of the woofers has an additional coil behind it. This woofer gradually becomes more active with lower frequencies. That is not acoustic crossover just so you don't mix tomatoes and oranges.

 

[TLS Guy]

I think I got the gist of that. Basically even though they don't list it, the two drivers are actually handling different ranges? Is that right?

In a way. The upper driver is handling as much bass as the lower driver, however the upper driver handles mid range as well. The lower driver starts shelving in somewhere between 400 and 600 Hz depending on the width of the cabinet.

 

[matthewdclarkaz]

gotcha. makes much more sense now. thanks.

 

[Paul_Apollonio]

One word of warning, crossovers can often be downgraded by substituting inductors with heavier gauge wire, and therefor lower resistance. The specific resistance of the inductor selected is often vital to the performance of the crossover.

Mr TLS Guy, I do not think that designers by and large calculate the resistance of the series inductors used in their designs as you suggest. I think that budget considerations prevail in the choice of components. I have designed hundreds of crossovers and drivers since 1975. When I was working for an OEM manufacturer, never once was I asked by a customer to optimize the Qes of the driver to be "such and such" with the warning that I must compensate for the series choke resistance. I think it is a good thing to always minimize choke resistance, except if it comes at the cost of added non-linearity. Iron or ferrite core chokes will trade a linear distortion (series R) with a non linear distortion (hysterisis and saturation).

In Pro, where driver designers must deal with customers who routinely abuse the drivers, Q is calculated for elevated operating temperatures, and these same drivers seem overdamped when modeled with their VC at room temperature. That is often a seat of the pants approach, were we guess how badly the end user is going to abuse the product.

Do you have a single system example of where the designer calculated the driver Q with the notion that the series L used in the crossover would be raising the system Qts? It would prove your assertion. Thanks, Paul A.

 

[TLS Guy]

Mr TLS Guy, I do not think that designers by and large calculate the resistance of the series inductors used in their designs as you suggest. I think that budget considerations prevail in the choice of components. I have designed hundreds of crossovers and drivers since 1975. When I was working for an OEM manufacturer, never once was I asked by a customer to optimize the Qes of the driver to be "such and such" with the warning that I must compensate for the series choke resistance. I think it is a good thing to always minimize choke resistance, except if it comes at the cost of added non-linearity. Iron or ferrite core chokes will trade a linear distortion (series R) with a non linear distortion (hysterisis and saturation).

In Pro, where driver designers must deal with customers who routinely abuse the drivers, Q is calculated for elevated operating temperatures, and these same drivers seem overdamped when modeled with their VC at room temperature. That is often a seat of the pants approach, were we guess how badly the end user is going to abuse the product.

Do you have a single system example of where the designer calculated the driver Q with the notion that the series L used in the crossover would be raising the system Qts? It would prove your assertion. Thanks, Paul A.

I think most crossover programs do this for you. I leaned this from Joe D'Appolito, from his articles. I have improved my skills over the years by following a lot of his advice. The resistance of the inductor is very important in the parallel inductor in the high pass filter, were it can change he Q of he filter significantly. I know Joe will sometimes save money two ways in the instance of situations were a resistor is required in series with the inductor. Joe will use a cheaper inductor with a smaller gauge of wire, and also save the cost of the series resistor. I can't believe Joe is the only guy who works that trick.

 

[Paul_Apollonio]

I think most crossover programs do this for you. I leaned this from Joe D'Appolito, from his articles. I have improved my skills over the years by following a lot of his advice. The resistance of the inductor is very important in the parallel inductor in the high pass filter, were it can change he Q of he filter significantly. I know Joe will sometimes save money two ways in the instance of situations were a resistor is required in series with the inductor. Joe will use a cheaper inductor with a smaller gauge of wire, and also save the cost of the series resistor. I can't believe Joe is the only guy who works that trick.

Dear TLS, I understand now. You are talking about parallel chokes. If you require a series resistance then you replace the larger wire with smaller, and save money on the choke and resistor is built in. True. That said, in a parallel choke, that resistance manifests itself and a limitation on the stopband rejection. More than 99% of the time, I do not want it. There are many instances where I need a small resistance in series with a parallel capacitor so preclude the impedance magnitude dropping too low. I do not recall a single instance when I ever NEEDED to have series resistance with a parallel (shorting) choke. I can see how with a fourth order network, that may be useful.

 

[ozmedia]

T-30

Holy SHAMOLEY!! I've never seen the T-30 LSE crossover before, we have a power station down the street that looks less robust! Shane and crew did a pretty nice job with that one!

 

[DJKen]

Paul,

I have been building audio systems for thirty years, albiet a whole lot more informally than you, and this is one of the best explainations of the parts that work to make a crossover.

One thing however, I can tell by the writing style that somebody else mispelled the title of the section "Maths"; your demonstrated syntax and style wouldn't produce such an uncomfortable assemblage of letters on purpose!

Cheers!

Ken

 

[gene]

One thing however, I can tell by the writing style that somebody else mispelled the title of the section "Maths"; your demonstrated syntax and style wouldn't produce such an uncomfortable assemblage of letters on purpose!

Its kinda an inside joke between Paul and myself. I used to be a math major when I went to school to actually become an Electrical Engineer and I had a math teacher that always used to say "check your maths". Its kinda a slang among geeks I suppose.

 

[nosferatu]

here is link to Vienna acoustics woofer and crossover , the crossover don't look too Hi-End to me

link