I am working on a passive crossover of a 3-way. There is an active crossover in my preamp, dividing the bass and higher frequencies.
Then, there is a passive crossover between my mid and tweeter. This is the crossover I am working on.
What I have noticed is that, while trying different alterations of my crossover, the performance of the bass varies considerably.
I know that the bass unit (which has seen no changes) can play very very well, with deep and very dynamic accurate bass. However, with some iterations my xo the bass becomes thick and slow. I have looked, but cannot find anything that might explain what I hear. I looked at phase integration between the hf-drivers and the bass. I have looked at timing differences (time domain). But what I hear seems uncorrelated with anything I can analyze.
I am sure that experienced designers must be familiar with this, and I hope that in general terms I could get some pointers to look at.
What am I missing?
Then, there is a passive crossover between my mid and tweeter. This is the crossover I am working on.
What I have noticed is that, while trying different alterations of my crossover, the performance of the bass varies considerably.
I know that the bass unit (which has seen no changes) can play very very well, with deep and very dynamic accurate bass. However, with some iterations my xo the bass becomes thick and slow. I have looked, but cannot find anything that might explain what I hear. I looked at phase integration between the hf-drivers and the bass. I have looked at timing differences (time domain). But what I hear seems uncorrelated with anything I can analyze.
I am sure that experienced designers must be familiar with this, and I hope that in general terms I could get some pointers to look at.
What am I missing?
Could be a series crossover, trace out the crossover schematic.
Series Vs Parallel Crossovers
Series Vs Parallel Crossovers
Last edited:
Erik,
Could you elaborate? Should I be looking at the impedance as seen by the amp, or are you referring to the look-back impedance of the drivers?
I must admit I did not really address impedance as I know (read: have assumed) that the impedance seen by the amp would not reach a critical value.
Could you elaborate? Should I be looking at the impedance as seen by the amp, or are you referring to the look-back impedance of the drivers?
I must admit I did not really address impedance as I know (read: have assumed) that the impedance seen by the amp would not reach a critical value.
Since you've clearly mentioned that the bass and mid+treble are actively separated, I suspect pattern and lobing issues for the bass variations. Did you play around with the order of your HPF (mid+treble) filter by any chance?
With so little info, can only assume the woofer has a high inductance, so either XO as low as practical and/or reactance annul it: Zobel Calcs making me crazy!
The idea of lobing problems is interesting. Does the information below tell anything about possible pitfalls?
The woofer has an active 4th order Linkwitz-Riley (electrical and acoustical). The highs are filtered using a 2nd order filter (almost perfect 4th order LR acoustically). The passive crossover is LR2 (acoustical).
The best bass I got was when both active filters were 4th order. But ironically, the filter was way off on the acoustical side.
With that changed, the bass was sometimes quite good, but lacking impact (micro dynamics I would say). With other versions of the passive XO it was just very bad.
Each woofer is driven by its own 180W Hypex amp. I am not worried about its impedance at all.
Also, my 300W class AB mosfet amp I use to drive the mid and tweeter, should have no issues driving any impedance. I did not yet have time to look into the impedance of the xo though.
The woofer has an active 4th order Linkwitz-Riley (electrical and acoustical). The highs are filtered using a 2nd order filter (almost perfect 4th order LR acoustically). The passive crossover is LR2 (acoustical).
The best bass I got was when both active filters were 4th order. But ironically, the filter was way off on the acoustical side.
With that changed, the bass was sometimes quite good, but lacking impact (micro dynamics I would say). With other versions of the passive XO it was just very bad.
Each woofer is driven by its own 180W Hypex amp. I am not worried about its impedance at all.
Also, my 300W class AB mosfet amp I use to drive the mid and tweeter, should have no issues driving any impedance. I did not yet have time to look into the impedance of the xo though.
As a problem, the issue has resolved itself. With the latest, and close to final, iteration of the crossover, suddenly, the dynamics are back in full glory.
what it took was a small increase in capacitor value in the mid section. The increase seems minor (from 5.6 to 6.07 uF or 8%) but had a rather dramatic impact.
Still interested to understand how such a change could affect bass performance the way it does.
https://www.diyaudio.com/forums/attachment.php?attachmentid=938217&stc=1&d=1617357493
what it took was a small increase in capacitor value in the mid section. The increase seems minor (from 5.6 to 6.07 uF or 8%) but had a rather dramatic impact.
Still interested to understand how such a change could affect bass performance the way it does.
https://www.diyaudio.com/forums/attachment.php?attachmentid=938217&stc=1&d=1617357493
Attachments
It really doesn't take much once you begin to get close.
So what difference have you made, perhaps a very small increase at some point in the midrange (guessing). Maybe this corresponds to a harmonic of what sounded like a bass frequency issue...
So what difference have you made, perhaps a very small increase at some point in the midrange (guessing). Maybe this corresponds to a harmonic of what sounded like a bass frequency issue...
The difference, objectively, is small. The mid driver drops slightly steeper between 2-5 kHz (which is out of its passband) with less than 1 dB difference.
Phase alignment is unaffected (of course the graphs move slightly on my screen, but I can't say which is better).
What I've come to realize is that the Bass is the hardest range to get right (a lesson learned while designing amps and digital gear). It requires all drivers to deliver their part in perfect concert, including the tweeter.
So maybe, the better integration the xo provides between the mid and tweeter, pays off in the bass region too. It certainly seems that way.
The improvement is that everything has more impact. The small hammers on a grand piano, for instance, are really hard hitting, and I had to dial down the woofer with at least 1 dB after the change.
Phase alignment is unaffected (of course the graphs move slightly on my screen, but I can't say which is better).
What I've come to realize is that the Bass is the hardest range to get right (a lesson learned while designing amps and digital gear). It requires all drivers to deliver their part in perfect concert, including the tweeter.
So maybe, the better integration the xo provides between the mid and tweeter, pays off in the bass region too. It certainly seems that way.
The improvement is that everything has more impact. The small hammers on a grand piano, for instance, are really hard hitting, and I had to dial down the woofer with at least 1 dB after the change.
Bass tones from a musical instrument can have a very wide frequency range. Those finger plucks from a string bass and the hammers of a piano easily creep into the midrange and even the tweeters range. I have experienced this also, properly integrating the mid and high frequency driver usually makes the bass sound better. The opposite is also true.