Home Theater Forum and Systems banner

1 - 20 of 22 Posts

·
Registered
Joined
·
704 Posts
Discussion Starter #1
Here's the first of what I hope to be a growing series of articles concerning designing and building passive crossovers. I hope you find this useful.

David Yohn

****
Crossover Components and Assembly

I’ve been designing loudspeaker systems since 1977 and have done so both as a DIY’er and professionally for commercial companies. Crossover design is one of the most difficult parts of loudspeaker building, and can be quite confusing for some people. I don’t claim to know it all, nor do I claim that my opinions are the only valid ones: there are as many different ways to approach loudspeaker design as there are designers. This is simply a few ideas I’ve learned over the years in an attempt to help you get started.

This first article is not a tutorial to teach you how to design crossovers. I will not discuss the merits of crossover topologies or choosing the correct points and slopes or whether series or parallel or Linkwitz or Bessel is the best sounding. Those articles may come later. This guide simply outlines a few ideas that I think are important to keep in mind when choosing and laying out your crossover components, and offers a few tips and ideas about construction and component selection.

First, you need familiarity with electrical schematics and you must understand wiring concepts such as parallel and series, the concept of polarity and an understanding of Ohms law. You should understand the basic characteristics of necessary components such as resisters, capacitors, inductors, and electrical insulation. You need to understand how to solder and have all the proper soldering tools, how to strip wire and have those tools, etc. You need basic electrical technical skills - you don't need to be an electronics engineer. These skills are easily learned and I recommend if you are new to them that you practice a bit on some scrap wire and components to build up your chops.

Design your crossover using a software modeling tool to predict performance. There are many such tools available, ranging from expensive packages such as LEAP or Praxis to free tools like the Passive Crossover Designer from FRD consortium, Speaker Workshop from AUDA, and many on-line calculators. If you understand how to use them they all can be used effectively. Even our own REW software has been used to measure the effect of crossover components to optimize a design, as described here. Don't forget to take into account the series resistance of inductors when modeling your crossover. Be aware that if you use something like an on-line calculator you'll get "text book" component values based solely on the math, not based on the characteristics of the particular driver you intend to use. These generic values will never be "perfect" for your speakers unless you get really lucky.

Most software packages will print out the crossover schematic after it's designed, or you can draw it onto a piece of paper. I find it is easier to hand-draw it out as it helps make it "mine." But that's just me and I'm old. :) Realize you are simply drawing the electrical schematic to become more familiar with the connections in the circuit, not drawing the physical layout of the components. Minimizing the number of connections is generally a good idea.

After the crossover is designed make a list of all the required components and the target values. Remember that you can add capacitors together by wiring them in parallel. So if your design calls for, say an 83uF capacitor, you have the choice of finding an 83uF cap or of wiring four 20uF caps and one 3uF cap in parallel for the same net value. Inductors and resisters add in series wiring and add in reciprocal with parallel wiring (meaning 1/Rt = 1/R1 + 1/R2 + 1/R3 etc.)

How components add is a basic part of electrical theory you should understand before you get started. Also realize that many high-quality crossovers will be the single most expensive part of a loudspeaker design. I have spent many times the cost of the raw speakers and the enclosure to create a good crossover. Do not cheap out on the crossover parts; they will do more to determine how the system sounds than anything else you use.

As far as component selection, there are as many conflicting ideas and opinions about this as there are about almost anything else in audio. You will go crazy trying to sort them all out and decide who’s “right.” In general caps are caps and resisters are resisters but let me give you a few of my basic rules of thumb developed over 30 years of doing this kind of work.

• First, component values do not have to be exactly what was modeled. Indeed, it is often impossible to find components that are precisely the computer modeled values. Generally using values within +/- 5% is close enough – and most components are no better than 5% accurate anyway. It will subtly change the performance of the network to not get it “perfect,” of course. If you want to see the effect of the change to decide if it’s acceptable to you, simply jump back onto your model software and enter the exact values you intend to use. That’s the beauty of using software.

• Second, always use non-inductive resisters of sufficient power rating. Generally resisters are used, if at all, as pads for tweeters, so their power rating should be at least that of the tweeter. In a loudspeaker system designed for, say, 100 watts total power handling the tweeter will generally never see more than 10 watts. If the resister is in series with the woofer, however, it can be subjected to the full 100 watts. Wiring 2 resisters in parallel effectively doubles their power handling capability (paying attention to what that also does to the net resistance.) Resisters used in Zobel networks can usually be any power rating that you have available.

• Third, pay attention to the wire gauge of your inductors. Be sure that any inductor used in series with a speaker can handle the power rating of the speaker, and also realize that the smaller wire gauge in an inductor the higher its series resistance will be. I generally recommend using no smaller than 18AWG inductors, and I also recommend generally not using multiple inductors to try and create the effect of a larger one. While this can be done and often is, I find it is less accurate and less predictable than using one inductor that is close to the target value. I also suggest using air-core inductors as much as possible. Sometimes in low-pass networks the inductance value required is so large that the cost of an air core is prohibitive, and iron core must be used. Minimize this, in my opinion, as the hysteresis in iron cores will introduce sometimes audible distortion. Use laminated cores instead of solid cores if this type of inductor must be utilized. If you have the right tools you can hand-wind your own inductors (or unwind commercially available ones) to get the right values. But don’t try this unless you do have the right inductance measurement tools or you can never be sure of the results.

• Fourth, capacitors. This is where most of the controversy and witchcraft exists, and I’ll likely add to it with these statements. I say never, ever use electrolytic caps in a loudspeaker crossover. Yes, they will work, yes many people suggest them, yes there are special non-polarized electrolytic caps for just this purpose, and yes many if not all commercial loudspeakers use them because they are cheap. I’ve used them in commercial designs many times. But since this is a DIY design and you are less restricted by things like target price points and corporate profit margins, don’t use them. Instead, use the best caps you can afford. Save up for them if you have to. Use metalized polypropylene in woofer circuits and film/foil caps in midrange and tweeter circuits. There is no need to buy esoteric gold foil caps that cost $100 each unless you want to, and if you do want to go for it! But don’t cheap out and use electrolytics, you will hear the difference.

As an aside, there is no such thing as polarity on a non-polarized capacitor. It is non-polarized. Duh. That being said, some suppliers of high-end caps – especially those intended as so-called by-pass caps - will mark them with some sort of polarity color code or marking. This usually corresponds to the inside and outside of the cap’s internal construction and is not really polarity at all. I have never been able to measure or hear any difference using these types of caps no matter how they are wired into a circuit, but if you use them then by all means follow the manufacturer’s directions, it can’t hurt.​

Once your components are selected and you’ve purchased them it’s time to begin the layout process. I suggest clearing a space on your table or workbench and physically laying out the components following your schematic as a guide. I like to lay out the components visualizing signal flow from left to right (although there is no valid electrical reason to do this. It’s just that’s how I learned to read, I’m right handed, and it makes sense to me.) I start with the amplifier input on the far left then lay out the high pass section from left to right ending with the tweeter output on the right. Then just below that I place the mid band-pass section if it’s a three-way, and then the low-pass section below that. This will take up far more physical space than the actual circuit will consume once it’s all wired together, but it lets me survey the components, look for anything that’s missing, and visualize the signal flow before I solder anything together. It also lets me decide if I have too many or too few components: I might decide to use a cluster of paralleled caps in place of a single large one or vice versa. I might discover I’ve forgotten to purchase something, or that I need to improvise something else. I might decide this circuit is complex enough that I need to incorporate barrier strip terminal boards, or that I can minimize the signal path wiring by moving things around. To me this is a very valuable exercise.

Next is to determine how inductors need to physically lay out. You want to try and minimize the number of inductors that are on the same plane, meaning the windings should be laid out so they are not all oriented in the same direction. Set some parallel with the table, stand some up at 90-degrees, and lay others at various angles. You also want to physically separate as far apart as possible the low pass inductors from those used in the high pass section. Inductors work by creating magnetic fields, and the purpose of all this jockeying about is to minimize how much those fields interact with each other. Do not stack inductors on top of each other.

Now is time for moment of truth number one: test your crossover for signal flow. Use alligator clips or jumper wires and connect the components together to create your circuit. Check point-to-point continuity with an ohm meter to ensure the connections are intact. If you have access to some of the software tools mentioned earlier, then you may have the ability to test the crossover circuit functionally. I have spent several weeks at this point swapping components and making subtle changes in component values until I get exactly the output curve I am looking for. This is where the “art” comes into the science, and is where it is possible to build in many different schemes concerning “voicing” or EQ to create a particular output or to compensate for the specific characteristics of the drivers and/or enclosure type you are using.

On the other hand, if you don’t have these tools (although Speaker Workshop is a free download) or you simply don’t want to spend the time, at least connect a test speaker to each output and run some tones or frequency sweeps to make sure you are getting a signal through the circuit, and that you are not shorting out the amplifier.

Next determine how large a mounting board you can use and how large you need. I like to mount my crossovers using perforated fiber board, such as that used on shop walls to hang tools. It is cheap, easy to cut, and using the perorations I can tie-wrap my components in place (I use plastic cable ties) to make sure they do not rattle around and to keep all the leads separated from each other so they do not short out. You can also ensure the cut piece of perf board fits inside your enclosure (or inside the external enclosure – I love to mount my crossovers inside painted cigar boxes and keep them outside the actual loudspeaker) before you assemble it. Some people like to use printed circuit board blanks (also called bread boards) but I find these to be too limiting as far as component selection goes (some inductors and some caps are quite large) and that stuff costs more than fiber board and is harder to cut. It should also be noted that there really is no reason to use any sort of mounting board at all should you choose. You can simply solder the components together and toss them into the enclosure. The down side of this is several things, not the least of which is you’ll have to electrically insulate ever bit of exposed lead wire on the components to ensure they do not short out, but it could be done this way.

Mount your components on the board and then make all your electrical connections. I like to twist each connection together tightly by hand and then apply solder to secure them. After soldering I snip off the excess leads and then move on to the next connection. It is usually best to directly connect components together, but don’t forget that if you can’t make them reach a short piece of wire can be used to make the connection.

A few words about wire. This is as much a pit of vipers area as is capacitor selection. There are companies that make tons of money selling esoteric wire at sometimes quite high prices. There are people who claim they can hear the difference between one wire and another. If that is you and you like this idea and you want to spend the money or if you believe there might be some sort of wire effect that you want to try then by all means go for it. But in my opinion there is no scientific reason to use anything other than plain old copper hookup wire, or plain old speaker wire, in your crossover. Most commercial loudspeakers use copper-trace printed circuit boards in their crossovers and they sound just fine. Me, I use 16AWG 200-degree-C rated stranded hookup wire, which means it will be FEP Teflon insulated and the inner conductors are copper that have been silver plated. I do not do this for any sonic reasons; it’s just that I like working with it. The silver-plated conductors make it very easy to solder and the Teflon insulation offers a lower profile than comparable PVC or rubber. Plus Teflon looks cool. You can purchase this sort of wire from any electrical shop, and it’s usually less than 50-cents per foot (much less if you buy in bulk.) Wire color does not matter, although I do like to use different colors for the positive and the negative sides of the circuit just to make it easier to identify.

Once you have it all wired together it’s a good idea to again check point-to-point continuity with an ohm meter and to make sure you have not shorted out the positive to the negative input. Then mount it into your enclosure. I recommend that you do not glue it in. Use screws, or tape, or something removable just in case you decide either through testing or listening that something needs to be changed or adjusted. Similarly, mount your drivers temporarily in the enclosure using as few screws as possible and connect them to the crossover.

Moment of truth number two: give it a test. Connect the speaker system to an amplifier and run some test tones, or some music, or whatever method you prefer to judge the performance. Take your measurement or simply listen to the results. If it’s what you intended, properly mount the drivers. And then sit back and enjoy.

But don’t be surprised if some time, in a day or a week or a year later that you decide to make a few adjustments or improvements. Tweaking a crossover design or trying new things is part of the joy of loudspeaker building. Have fun with it.

Cheers!



References:
Great Sound Stereo Speaker Manual by David Weems and G. R. Koonce
The Loudspeaker Design Cookbook by Vance Dickason
Testing Loudspeakers by Joseph D'Appolito
The Master Handbook of Acoustics by F. Alton Everest
 

·
Registered
Joined
·
311 Posts
Looks like a good read. I'm not prepared to get through it just yet at this time. I'm going to get at this later when I have the time. Can't wait to learn more about crossover design!
 

·
Elite Shackster
Joined
·
7,142 Posts
I'm gonna try get stuck into this and have a shot at building some speakers. I'll try report on my experience from this guide. If I can follow it, anyone can :T
 

·
Registered
Joined
·
1,001 Posts
All your information is still current and usable. Excellent article.
Why the film/foil caps in midrange and tweeters?
Do all "2 way" speakers need:
1) X-over circuit
2) Baffle step compensation
3) Notch filter
4) Zobel
 

·
Registered
Joined
·
704 Posts
Discussion Starter #6
All your information is still current and usable. Excellent article.
Why the film/foil caps in midrange and tweeters?
Do all "2 way" speakers need:
1) X-over circuit
2) Baffle step compensation
3) Notch filter
4) Zobel
Thanks for the comments. I suggest using the best quality caps you can afford in series with drivers, and film/foil tend to be faster and have tighter tolerances (and to sound better to my ears) but certainly a good sounding crossover can be made using poly caps. As to your questions:

1) Yes. At the very least the tweeter must be protected from too much power at fs.
2-4) No. These components are only necessary when they are, well, necessary. BSC can be avoided with proper front baffle design, notch filters are only required to tame resonances that might be present in some drivers, and Zobels are only necessary when a woofer's impedance rises so much that it affect the crossover performance, which may or may not be the case for every driver. You can test this in any good modelling program.

Cheers.
 

·
Registered
Joined
·
1,001 Posts
Is the impedance value of a driver an "exact" number, or is it a number representing an average value?
The free X-over calculators on-line will post much different cap and inductor values if you change the impedance value of the drivers even 2 ohms.
 

·
Registered
Joined
·
704 Posts
Discussion Starter #8
Loudspeaker impedance varies continuously with input frequency. The values listed by manufacturers represent the average impedance of the driver over its operating band. Use these numbers in calculators (and realize that on-line calculators result in typical "textbook values" and will never produce the best results. You really need to model your crossover in a simulation program such as Speaker Workshop or Passive Crossover Designer.)
 

·
Registered
Joined
·
1,001 Posts
Just out of curiosity, if a low freq driver's impedance curve measures high (32 ohms) at my desired X-over point (1200Hz) is using THAT impedance likely to produce a more desired cap/inductor value relationship?
Or does the impedance @ the driver's Fs have more to do with correct X-over values?
 

·
Registered
Joined
·
704 Posts
Discussion Starter #10
If the impedance of a nominal 8-ohm driver is up to 32 Ohms at the desired cut frequency, then perhaps you are trying to operate that driver at too high a frequency. If you look at the impedance curve of a woofer, you will see a relatively flat band between fs and where it rises significantly, and the nominal impedance is in this band. If you need to operate a woofer in the band after it rises to more than twice nominal, then my advice is to use a Zobel circuit and build the Xover for nominal impedance.
 

·
Registered
Joined
·
4,418 Posts
Thanks for the comments. I suggest using the best quality caps you can afford in series with drivers, and film/foil tend to be faster and have tighter tolerances (and to sound better to my ears) but certainly a good sounding crossover can be made using poly caps. As to your questions:

1) Yes. At the very least the tweeter must be protected from too much power at fs.
2-4) No. These components are only necessary when they are, well, necessary. BSC can be avoided with proper front baffle design, notch filters are only required to tame resonances that might be present in some drivers, and Zobels are only necessary when a woofer's impedance rises so much that it affect the crossover performance, which may or may not be the case for every driver. You can test this in any good modelling program.

Cheers.
Thanks dyohn, that is very good information to know.:T
 

·
Registered
Joined
·
1,001 Posts
In this particular driver the nominal impedance is 8 ohms. The impedance IS 32 ohms at the X-over point of 1200 Hz. I had intended to use a Zobel on this driver.
Does the harmonic distortion of the driver plus this added circuitry of X-over and Zobel change for the worse? How about any of the other driver qualities ie: dispersion, freq res, delay, etc.?
 

·
Registered
Joined
·
5,771 Posts
In this particular driver the nominal impedance is 8 ohms. The impedance IS 32 ohms at the X-over point of 1200 Hz. I had intended to use a Zobel on this driver.
Does the harmonic distortion of the driver plus this added circuitry of X-over and Zobel change for the worse? How about any of the other driver qualities ie: dispersion, freq res, delay, etc.?
That is a hard question to answer, it depends on the driver and it's parameters.
A) the harmonic distortion of the driver can not be changed.
B) the added circuitry of X-over and Zobel are purely to send the correct signals to the driver and present a uniform load to your amplifier.
C) all these other factors should be analysed and compared before you commit to a design.

The manufacturers data sheet should give recommended settings for maximum and minimum crossover frequencies, you should err on the safe side when designing a passive crossover.

Hope this helps.
 

·
Registered
Joined
·
1,001 Posts
The driver I'm working with is a:

Woofer, 6.5 inch Peerless 830874 with the PPB cone in the HDS series.

The X-over treatment questions arise over the change in manufacturer's T/S data (significant) and my own testing results using WT 2 woofer tester.
I have minimal testing equipment and can't test harmonics, decay, freq res, etc.
I'm mating the Peerless woofer with a textile dome tweeter:

Seas Prestige 27TDF.

First initial modeling with 1st order X-over showed both speakers with exceptional detail but had some audible peaks in the cross over region and the spl difference of +3db indicates an L-pad for the tweeter.
I'm currently in the 2nd order X-over testing stage and now believe a Zobel and notch filter may be required to achieve a good response curve.
 

·
Registered
Joined
·
350 Posts
Don't forget that the roll off of the driver can and often does contribute to the db/octave crossover slope so that an electrical crossover of 12db can become 24db with resulting phase changes. Also, if one knows how, a crossover can take into account the inductance change and do without the extra circuitry the zobel adds. The only way "proper" baffle design can eliminate the "baffle step" is by using a large baffle which effects imaging.
 

·
Registered
Joined
·
1,001 Posts
Thanks Theresa.
I'm in the listening stage and settled temporarily on 2nd order crossovers for both woofer and tweeter but separated the frequencies an additional 200 Hz to flatten a hump @ crossover. I placed a zobel on the woofer and THAT seems to clear some woofer irregularities in the mids and into the X-over region. I also reversed wiring on the tweeter to phase match with woofer.
I have designed a minimal baffle in an attempt to achieve soundstage and imaging.
Right now I'm pleasantly surprised with their overall performance.
 

·
Registered
Joined
·
350 Posts
Isn't the gratification of building and designing one's own wonderful? I lived with some Vifa's for years always amazed that I put them together. Now I'm using Eton and ScanSpeak drivers and am even more amazed. I'm going to replace the Eton midwoofers with ScanSpeak Revelators to match my center as soon as Meniscus has the baffles cut and ships them and the drivers to me. I use active crossovers with independent amplification of each driver for the front's and sub's. I'm in high end DIY bliss!
 

·
Registered
Joined
·
1,001 Posts
LOL yup, you're high end all right. Any time you use an amp per driver you got yourself a high high end system.
I have no experience with the "Revelator" series of drivers but hear great things about them and observed they are used by high end manufacturers. Cost per unit has a limit for me.
I have a Pioneer AVR and use the MCACC circuits to further adjust response curves beyond the X-over.
I like to think I've got the poor man's "Theta".
 

·
Registered
Joined
·
14 Posts
In My last speaker proekts I used condensatorless ( Acoustic Reality crosower) filters.
Work perfektly in My TL proekt Scan Speak 18W8531G + twit9900O-Revelator.
My inductors-Mundorf copper foil 1.5mH and 0.12mH
Resistor-Mundorf M-resist.
 
1 - 20 of 22 Posts
Top