Axiom Home Page Forums General Discussion Technical Questions Crossover Calculator

http://ccs.exl.info/calc_cr.html

Link to a neat little sight to help you figure componnet values for that DIY crossover upgrade. The crossover is fundamentally one of the most important (often overlooked) elements in a speakers design. Most manufactueres (including Axiom) use cheap components in the crossover design and upgrading to better higher quality componnets of the same values (you'd be amazed) can drastically improve the sound of an already good sounding speaker. Keep in mind to never do anything you aren't comfortable with...its your investment!

So you're implying that expensive (i.e. not "cheap") crossover components always sound better?

No thanks. I think I'll leave the speaker design to the experts. I'm sure they've gone through many different tests with many different crossover designs and found one that performs well for the price.

I'm sure it's possible to design a better crossover than those included with the Axioms, but why mess with something that ain't broke?

I respect your position. By all means never do anything you don't feel comfortable with. And no, I'm not implying anything, its a scientific fact. When you choose compnents for their value to acheive a certian electrical effect, those with better tolerances will provide you with better performance. Period. Axiom uses an economy of scale to produce speakers of the quality they do for the price. They have to make concessions or the price would be passed on (to you). Yes, you could build an XO using the SAME values but use higher quality Indctures & Caps and get a different (better?) sound. Its all subjective anyway, so do whats best for yourself.

I have read quite a few threads here on the site from Axiom owners who tweaked their M22 XO's and made a great speaker spectacular. Neccesity is the mother of invention, but if you are happy with what you have, thats ok too. )

Clinton,Ian responded a long time ago to this capacitor and inductor myth. The only thing accomplished is wasting money. Engineers design good crossovers using components entirely adequate to meet performance requirements.

I would like to know where you get your information from.

You call it a scientific fact that replacing stock components in a crossover circuit with components of better tolerances (what is tolerance, anyway?) -- even though they have the same values of resistance, capacitance, and inductance -- results in better performance.

I beg to differ. Perhaps you should explain what you mean by tolerances. Do you mean that the given ratings of the components used in the Axiom XOs are off by a significant margin? Do you really think an expensive and overbuilt capacitor will woo the electrical current into sounding better than one engineered specifically to provide a certain Farad rating?

I understand you're just trying to give us some helpful information, but many of us here like to stick to our healthy sense of skepticism until such "tweaks" can indeed be considered scientific facts.

I'm sure you are familiar with the inescapable psychoacoustic side to audiophilia...

pmbuko:
Tolerance is the variable at which an electronic component meets its rated measure, and is identified as a + or - from the rating. If a capacitor has a 25% tolerance (purley for example) from its rated measure then what makes you think it will perform its function in an electrical circuit "as good as" a component that more closely meets its rating? Its not a difficult concept to grasp, and it is based in fact, otherwise the mathematical/physical principals behind all of it might as well be bunk!...

Axion does not "engineer" the individual components used in its XO's, they buy them from manufacturers (and in bulk), and they may or may not be the best component for the job at hand, but they get it done. For example: Iron core inductors are used in many XO's mainly because of their cost, but also because they have very low DC resistance and a small size for their given value. However, iron core inductors are known to introduce more distortion as apposed to foil coils...You can call it "woo" if you like, but some ears can hear the difference...

Its a free country, but the only way that you are ever going to know anything for yourself is to experiment for yourself, otherwise you are going to be taking somebody elses word for what you believe to be true anyway ... )

Where are you getting your parts and are you sure you're buying lower tollerance devices? Partsexpress, for example is selling normal components that any moron can buy from DigiKey or even get free if they call the disties for samples. There's nothing special about their components despite their claims of "audio-grade".

The frequency response of a simple RC lowpass filter is defined by:

1/(2*pi*R*C)

Precision refers to the manufacturing tolerance of the process. A 100 Ohm resistor with a 5% tolerance means that anything +- 5 Ohms from 100 is allowed to be sold and anything more is discarded (or more likely sold as a 10% piece). By swapping the components, you change the frequency response of the filter. Any difference in the sound you hear is due to the fact that the filter is responding to a different frequency.

If you want a steeper negative gain to your filter, you'll have to employ a different circuit, not merely change your frequency response.

If you use a use an inductor with a series resistance of less than 0.6 ohm (enabling more power to the driver because of a lower value of overall series resistance) how does this effect the "overall" sound of the driver?

I appreciate the more scientific explanation, Semi. I was trying to say something similar, but didn't have the knowledge to fill in the gaps in my argument.

So again, slownlo, changing the components in the crossover only changes the frequency of the XO point. This point is readily measurable by feeding test signals through the XO circuit (not individual components).

To state it again, I trust that Axiom engineers have tested and measured their crossovers so that they conform to a level of quality control that allows their speakers to consistently achieve the expected performance.

Your entire argument hinges on the assumption that more precise parts result in a better sound -- when in fact all you're going to get is a different sound due to the fact that the crossover point has changed. Wouldn't you agree that the crossover point that comes with the speaker has been carefully chosen and consistent beyond a detectable level between one speaker and the next?

Loudness, rated in decibels, is a logarithmic scale. You have to double the power to create anything resembling an audible difference in volume. Eeking out an extra fraction of a watt isn't exactly a dramatic change in the face of a logarithmic scale.

Just in case anybody is still interested in upgrading the crossover parts, GR Research sells very high quality capacitors and inductors. WARNING: These tend to be physically much bigger than the "ordinary" factory parts. You will need to replace the circuit board altogether in order to accommodate them.

Incidentally, the calculator linked above is essentially useless. Modern crossover circuits are computer-designed by taking the dynamic impedance characteristics of the drivers per se into consideration. The "textbook" (Linkwitz or whatever) filters rarely work properly, without LOTS of trial-and-error. There are a number of Wintel-based crossover design software available for serious DIY'ers.

So, if anyone is interested in "upgrading" the crossover without wrecking havoc to the speaker, I would recommend simply substitute each L/C/R part with a "better" one with the identical value. WARNING: As people repeatedly stated above, this will unlikely "improve" the sonic quality of the speaker at all. If it were me, I would build the speakers from scratch rather than wasting my time and money on the dubious crossover upgrades.

I understand that, but lets broaden the focus a little.

How does this effect the damping factor? Its common knowledge that The better the amplifier can provide varying amounts of current without changing the applied voltage to the driver, the more accurately the speaker will reproduce what the amplifier is trying to tell it to do. Its not simply an issue of "eeking" out moe wattage, but improving the symbiotic relationship between the driver and the source...

Hi Sushi,

Your points are well taken, and it seems like your opininon is valued within this message board community. You make my point partially by describing the parts sold by GR as "high quality"....as opposed to what and to what effect in use? )

Whats useless to you may not be to someone else Sushi. The calculator is just another way of introducing the less experienced to how the values of a XO's components change a speakers characteristic response to its input. Some people may be intimidated by the GUI offered by some of the windows based programs...and again (4x), no one should do anything that they are not personally comfortable with.

Maybe building speakers from scratch, though personally gratifying, would also be a waste of time (and money) considering the value and sonic quality to be had in the purchase of speakers already built (think Axiom).

I have read a few posts from folks here that have successfully changed values of XO components to pleasant surprise, and have managed to alleviate a "personally" irritatiing aspect of the speakers translation of input. Its not for the "faint" of screwdriver to undertake, but rewards can sometimes outweigh the risk. Anyway, I'll not beat a dead horse. Thanks for your responeses and God bless you.

In reply to:

---

How does this effect the damping factor?

---

The damping factor is an easily calculable function of the resistive network used in the filter. Assuming you're even using an inductor with a lower internal resistance than the original design (which I've not seen you demonstrate is the case with these speakers), the effect will be a (very small) change in the total resistive network.

A damping factor of 2^.5 produces minimal overshoot. In the case of your upgrade, have you calculatedthe original circuit's damping factor? It may very well be pretty close to 1.414 (and I'd expect it to be) and you may very well be pushing it away from this critical value, thereby WORSENING your filter.

I wouldn't think there would be any debate as to whether a higher quality component could effect a change in the sound of a speaker, namely the crossover circuit. Just as the world's best amps and electronics do not skimp on .10 cent capacitors. It is relatively well known in the DIY community that the higher quality capacitors do change the quality of sound in the speaker. Just as you would use the highest quality foods in your meal you can definitely taste the difference. I know I can anyway. Now soldering and wiring is not an option for everybody, and in no way should someone change something they are not comfortable doing. But who ever said experimentation is a bad thing? Another great link is http://www.loudspeakers101.com.