ScottG said:
...which means you are really talking about the mid/high performance. Once you feed a starting/stopping lf wave through the xover, you will see that the transient parts go to the mid/high range, and that the wave that gets to the woofer has nicely rounded start- and stop parts; the sharp edgy parts are no longer there. Those are filled in by the mid/high. There is no need for the woofer to start and stop abrubtly. It just gets up to speed leisurely (so to speak) and stops equally leisurely. It's the mid/high that takes care of the fast parts.
Which brings me back to my earlier question: Does it make sense to speak of 'fast' woofers? I seems that it does not.
Jan Didden
Jan,
What you say makes sense, but it does appear that some drivers go about the process a little too leisurely and end up falling behind, so the individual bass notes get lost in a mush of bass. This is what comes to my mind for a term like "slow bass". The opposite would be detailed, accurate, precise, or "fast", with all having the same meaning.
What you say makes sense, but it does appear that some drivers go about the process a little too leisurely and end up falling behind, so the individual bass notes get lost in a mush of bass. This is what comes to my mind for a term like "slow bass". The opposite would be detailed, accurate, precise, or "fast", with all having the same meaning.
So, just to clarify ...janneman said:
Suppose two bass drivers (one with a heavy cone and one with a light cone) are both outputting a pure sine wave at 40 Hz. They each work alone, without a crossover.
If the signal is suddenly cut completely, will they both immediately stop and be completely inaudible from that instant?
If the signal is switched on again, will both cones instantly be at the full 40Hz output?
Jan, I believe that this is the concept that some forum members have difficulty understanding. Please could you explain further?
janneman said:
I don't think anyone has tried to say any woofer is fast compared to a higher frequency driver, just compared with other woofers. It's no different than calling a female track star fast yet compared to a male track star she isn't. Of course, compared to me she would be, unless I sat on her.
Another example: Let's say I just developed a new driver using the magnet from a 3" driver and a 10 pound metal cone. Wouldn't you consider it too slow to play any music and any real woofer would be fast in comparison. I'm sure there must be some kind of motor to weight ratio and ratios of other specifications which would indicate how "fast" a driver is.
7V said:
Of course, in the situation you describe, neither driver will stop dead in its tracks. The vibration will take some time to die out. How long that takes depends on the moving mass, the mechanical damping and the electrical damping (amp damping factor, leaving the xover out of the discussion for this moment). Same thing with switch on suddenly, it will take some time for the moving mass to get 'up to speed' so to say. Otherwise, they would require infinite acceleration (and deceleration) which would require infinite power.
Same thing wit a car, just the numbers are different, the principle isn't.
Exploring this a bit further, are there any musical instruments that generate vibrations that suddenly start and stop? I am no expert, but knowing that instruments (a piano string, a violin string, the instruments' bodies that generate a large part of the output) are all mechanical systems, so I would think not. Presumably, electronic music *can* generate fast start/stop transients.
At any rate, the signal is 20 kHz bandlimited anyway, so the signal risetime is limited too, excluding real fast start/stop stuff. And the xovers will direct whatever high freq components are in the signal (which represent the sharp parts) to the mid/tweeter.
Can someone do a simple sim with a 40Hz toneburst of a couple of cycles into a hi and lo pass (say 2nd order) and look at the two outputs? That would be instructive.
So, knowing that a reasonable woofer can work up to a several hundred Hz or even a kHz, and crossing it over say at 100 Hz, I don't think such a woofer would be challenged for speed. IMHO the differences heard (clean lows, coupled with sharp transients are much more related to the correct integration of the freq bands wrt amplitude vs freq. Unless you know what's going on in your system in this respect, changing the woofer can change this for the better or worse, but it remains a bit rolling dice and hoping for the best.
Otherwise, one man's 'fast' woofer can be another man's snail.
Jan Didden
johninCR,
Rember that driver's have intrinisic THD that has little to do with excrusion. (i.e. it behaves more like IMD.) Of course decreasing power to a driver will decrease it, but your still dealing with a summed output. In otherwords despite best intentions you may (and perhaps will) experience doubling.
janneman,
Actually, no. Remember that we are dealing with music - not a sweeped signal (like pink noise). Many have postulated that music can be regarded as a sum of sinusoidal waveforms, but they fail to truely understand the time based element that occurs with regard to changes in SPL. (i.e. transients are fundamentally TIME BASED, just look at how wrong Adire's info. was on this.)
A woofer will of course not start and stop abruptly with most music (my mention of "starting" and "stoping" was contextual). The key then is hearing those dynamic contrasts AND when they occur. This transient information is most certainly within lower freq.s - looking and the sum suggests a smoother "lossy" characteristic, but its there. More importantly its there in TIME - which is why I mentioned that the initial pulse for the particular transient was more important (i.e. the rise-time), but as you get better drivers for a particular bandwidth you'll notice that the decay of those transients is also very important. Would you like a visual representation? - download a screensaver like geiss and play an MP3 on your computer. You'll essentially see a summed sinusoidal waveform (though the shape may be "funky"), and you'll also notice the spl behaviour "punctuated" in TIME with low and high freq.s alike.
You know, I could debate this this until "the *dogs* come home" (woofers that is), and still not alter your perspctive with this theory based reasoning. Instead, if you are actually interested in seeing if your theory is correct, try doing a little empirical research comparing drivers with the physical characterstics I mentioned vs. drivers that don't have these characteristics, (excursion alone is rather easy to do). Utilize whatever crossover strikes your fancy.
(sorry if any of this seems rude, it not intended to be so. I am however to the point where I feel like I'm beating my head against the wall reiterating points I've already mentioned, but that have been perceived by others not fully within the context presented.)
Rember that driver's have intrinisic THD that has little to do with excrusion. (i.e. it behaves more like IMD.) Of course decreasing power to a driver will decrease it, but your still dealing with a summed output. In otherwords despite best intentions you may (and perhaps will) experience doubling.
janneman,
Actually, no. Remember that we are dealing with music - not a sweeped signal (like pink noise). Many have postulated that music can be regarded as a sum of sinusoidal waveforms, but they fail to truely understand the time based element that occurs with regard to changes in SPL. (i.e. transients are fundamentally TIME BASED, just look at how wrong Adire's info. was on this.)
A woofer will of course not start and stop abruptly with most music (my mention of "starting" and "stoping" was contextual). The key then is hearing those dynamic contrasts AND when they occur. This transient information is most certainly within lower freq.s - looking and the sum suggests a smoother "lossy" characteristic, but its there. More importantly its there in TIME - which is why I mentioned that the initial pulse for the particular transient was more important (i.e. the rise-time), but as you get better drivers for a particular bandwidth you'll notice that the decay of those transients is also very important. Would you like a visual representation? - download a screensaver like geiss and play an MP3 on your computer. You'll essentially see a summed sinusoidal waveform (though the shape may be "funky"), and you'll also notice the spl behaviour "punctuated" in TIME with low and high freq.s alike.
You know, I could debate this this until "the *dogs* come home" (woofers that is), and still not alter your perspctive with this theory based reasoning. Instead, if you are actually interested in seeing if your theory is correct, try doing a little empirical research comparing drivers with the physical characterstics I mentioned vs. drivers that don't have these characteristics, (excursion alone is rather easy to do). Utilize whatever crossover strikes your fancy.
(sorry if any of this seems rude, it not intended to be so. I am however to the point where I feel like I'm beating my head against the wall reiterating points I've already mentioned, but that have been perceived by others not fully within the context presented.)
Scott,
I for one appreciate your input. Regarding the fast/slow issue I was hoping for some rules of thumb, ratios, etc. that would help guide us in choosing drivers, but i guess they don't exist.
Regarding driver distortion: My thinking is that it would come from 2 sources, non-linear movement and distortion in cone shape. I thought that distortion from non-linear movement was easily audible in the higher frequencies and was what causes most of the cheap drivers that I hear to sound so very bad even at low levels. I also thought that distortion from the cone itself comes 3 sources, asking it to play higher frequencies than it is capable, stresses on the cone from higher excursion, and third asking it to play too loud (this probably includes non-linearities too.). This lead me to thinking that lots of cheap but decent sounding drivers was the most economical way to build a subwoofer that excels in low end response, spl and accuracy that will work as well for music as it does for HT.
I understand that the crossover will have to take into account that driver is being asked to operate in a range where it's frequency response is falling off rapidly and in all likelyhood the crossover point will need to be somewhere about where response starts to achieve a flat enough response.
Have I missed something? I there some source of distortion below Fs that I can't hear in low level listening tests which I should consider?
I for one appreciate your input. Regarding the fast/slow issue I was hoping for some rules of thumb, ratios, etc. that would help guide us in choosing drivers, but i guess they don't exist.
Regarding driver distortion: My thinking is that it would come from 2 sources, non-linear movement and distortion in cone shape. I thought that distortion from non-linear movement was easily audible in the higher frequencies and was what causes most of the cheap drivers that I hear to sound so very bad even at low levels. I also thought that distortion from the cone itself comes 3 sources, asking it to play higher frequencies than it is capable, stresses on the cone from higher excursion, and third asking it to play too loud (this probably includes non-linearities too.). This lead me to thinking that lots of cheap but decent sounding drivers was the most economical way to build a subwoofer that excels in low end response, spl and accuracy that will work as well for music as it does for HT.
I understand that the crossover will have to take into account that driver is being asked to operate in a range where it's frequency response is falling off rapidly and in all likelyhood the crossover point will need to be somewhere about where response starts to achieve a flat enough response.
Have I missed something? I there some source of distortion below Fs that I can't hear in low level listening tests which I should consider?
I'm certainly no expert on distortion - I just know its there below fs with most drivers (and that it increases substantially nearing fs). Even worse, its been my experience that cheap drivers display this characterstic in abundance. A google search should bring up a list of some sort (..voice coil heating is should certainly be on the list).
As for hearing distortion - well you can hear a decrease in linear distortion at low fundamentals, but if it doesn't bother you then just do it and enjoy. If you have access to measuring equipment that does distortion analyis for a given input (i.e. adjustable) - then go ahead and see what you come up with.
As for a rule of thumb: well the BL/Mms (high/low) does seem to come up with "faster" drivers more often than not - so why ditch something that seems to work even if the underlying theory contains flaws?
As for hearing distortion - well you can hear a decrease in linear distortion at low fundamentals, but if it doesn't bother you then just do it and enjoy. If you have access to measuring equipment that does distortion analyis for a given input (i.e. adjustable) - then go ahead and see what you come up with.
As for a rule of thumb: well the BL/Mms (high/low) does seem to come up with "faster" drivers more often than not - so why ditch something that seems to work even if the underlying theory contains flaws?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.