No crossover on mid-woofers = better sound...what?

[Cheetah]

So I own a pair of Polk M40's that are hooked up to my PC. They've always sounded too "digital" to me(if that makes sense) or compressed, like a low quality mp3, despite using lossless source content. I read a forum post somewhere(that I lost long ago) where the guy did something with replacing the capacitors on the crossover, and he said it sounded way better.

So, in my boredom I decided to connect the midwoofers directly, bypassing the crossover. The result? Actually, it sounds a hell of a lot better The mids sound twice as clear, the "compression" sound is gone, and the mids seem to blend in more, rather than sounding like separate speakers from the tweeter, they just blend in.

Why is this? Does anyone have any more info on this crossover mod? I know running without a crossover is probably not the most optimal solution, I would think without it the mids would be trying to produce the higher pitched sounds they can't reproduce accurately, but that doesn't seem to be the case.

I trust my ears more than my vague understanding of audio systems, so, I come here requesting an explanation on this paradox

 

[DD66000]

Not all capacitors are created equal, when it comes to audio. And if caps start leaking then its really bad. Many like to use active crossovers, that go between the pre-amp and power amp, instead of using passive crossovers in the speakers.

The general thought is that polypropylene caps are the best type to use for crossovers. I've used Solen polypropylene cap for all my crossovers. Also using polystyrene by-pass caps for the tweeters will make the sound more open and airy.

Plus, there is an inhert situation that takes place when the audio signal flows into the cap and then reverses direction. It called phase shift. As the audio signal passes the zero dylectric point, which is when the flow reverses, the signal does a slight phase shift, which is distortion.
The way around that is to build Charged-Coupled (JBL TM) crossovers. Which are quite costly to build, as both the size and number of capacitors doubles. The last pair of Charged-Coupled crossovers I built for a pair of 3-way speakers ran $300.

 

[Cheetah]

Very descriptive answer

Would replacing the caps with polypropylene caps be fine since they're budget speakers? Or should I invest in an active crossover, since I do have extra amp channels and bi-wire capable speakers?

 

[WmAx]

Actually, there is no credible evidence to suggest that capacitor type has any audible influence so long as the actual measured value of the capacitors is matched up within tight tolerances. Not in theory correlated with known human auditory research nor in controlled tests. I had discussions with Bob Cordell in designing a double-blinded test to perform at several DIY events 2-3 years ago. DIY DC and another one(which I forget). He then partnered with Dennis Murphy and they tested many audio enthusiasts. And many of these tests were done in special isolated room from the rest of the events. No one demonstrated a repeatable statistically significant ability to detect common electrolytic from high end film capacitors.

Both in theory and in applied practical tests under bias controlled conditions is it suggested that crossover type has any correlation to audio quality.

Now, let's get past that....

The capacitors in the computer speakers could be defect or bad, though that seems unlikely as they are not old enough for age related failure. Or, they are just such low quality speakers that any change may improve them..... perhaps removing the crossover from them helped to add coloration that masked even worse problems....

-Chris

 

[TLS Guy]

So I own a pair of Polk M40's that are hooked up to my PC. They've always sounded too "digital" to me(if that makes sense) or compressed, like a low quality mp3, despite using lossless source content. I read a forum post somewhere(that I lost long ago) where the guy did something with replacing the capacitors on the crossover, and he said it sounded way better.

So, in my boredom I decided to connect the midwoofers directly, bypassing the crossover. The result? Actually, it sounds a hell of a lot better The mids sound twice as clear, the "compression" sound is gone, and the mids seem to blend in more, rather than sounding like separate speakers from the tweeter, they just blend in.

Why is this? Does anyone have any more info on this crossover mod? I know running without a crossover is probably not the most optimal solution, I would think without it the mids would be trying to produce the higher pitched sounds they can't reproduce accurately, but that doesn't seem to be the case.

I trust my ears more than my vague understanding of audio systems, so, I come here requesting an explanation on this paradox

The crossover is the heart of every speaker. That crossover is obviously not right, if there is not a good hand off from woofer to tweeter. The tweeters and woofers should not sound like different speakers.

I use polypropylene caps, for durability. I agree with WmAx, that your speakers are not old enough for the caps to have deteriorated.

The problem is that it is impossible to redesign a crossover without detailed Thiel/Small parameters and full acoustic data for the drivers involved.

Your best bet if you don't like those speakers is to sell them, and buy or build something else.

 

[Cheetah]

The crossover is the heart of every speaker. That crossover is obviously not right, if there is not a good hand off from woofer to tweeter. The tweeters and woofers should not sound like different speakers.

I use polypropylene caps, for durability. I agree with WmAx, that your speakers are not old enough for the caps to have deteriorated.

The problem is that it is impossible to redesign a crossover without detailed Thiel/Small parameters and full acoustic data for the drivers involved.

Your best bet if you don't like those speakers is to sell them, and buy or build something else.

I intend to buy something nicer in the future, but I'm in a tight financial situation for the next few months. I guess I'll just pick up an spl meter and try to equalize them the best I can.

 

[DD66000]

Actually, there is no credible evidence to suggest that capacitor type has any audible influence so long as the actual measured value of the capacitors is matched up within tight tolerances. Not in theory correlated with known human auditory research nor in controlled tests. I had discussions with Bob Cordell in designing a double-blinded test to perform at several DIY events 2-3 years ago. DIY DC and another one(which I forget). He then partnered with Dennis Murphy and they tested many audio enthusiasts. And many of these tests were done in special isolated room from the rest of the events. No one demonstrated a repeatable statistically significant ability to detect common electrolytic from high end film capacitors.

Both in theory and in applied practical tests under bias controlled conditions is it suggested that crossover type has any correlation to audio quality.


-Chris

Polypropylene do provide a slightly more dymanic sound and inserting by-pass polystyrene will give a more open sound, that has been proven.
And if you keep the audio signal above the zero volt line, by adding a positive charge between a pair of series caps you eliminate the phase shift that would occur at zero volts.
But I was in no way, suggesting the use of "high end" caps costing 2, 3, 4 times the price.

 

[mike c]

uh oh.