otto88 said:
I meant just add together any L and R channels there might be and certainly not creating new channel divergences by imprecise filters. 'Fraid I am mired in the old world of acoustical music reproduction (using the term loosely). In acoustic music, while you can put mics anywhere you want in a hall and pick up very LF stuff differing in phase, etc., it really isn't meaningful differences and choosing just one of those mics is probably better than feeding both to a CD.
I suppose there are serious electronic composers (Vangelis' thundering Mission to Mars) as well as pop producers of disaster movies who diddle with the knobs to make synthetic special effects based on differing waves in two LF channels.
But all things considered, I'd just wrap them together anyway. By "but all things..." I am also thinking of the cost of making two seriously profound subwoofers, not just intra-music issues. Not sure I wouldn't extend that argument to favouring one fine speaker instead of two so-so stereo speakers. Or even one fine narrow-band full-range driver instead of....
The effect of a 12dB 15hz Q=.9 filter on an arbitrary speaker design ported to 20hz.
Note that the combined response of the system with and without the filter is essentially the same to about 15hz~17hz, and then begins to reduce both response amplitude and cone excursion. By 13hz the cone excursion is reduced by 30%, by 10hz it is reduced by 50%, by 5hz it is reduced by over 85%.
Note that the combined response of the system with and without the filter is essentially the same to about 15hz~17hz, and then begins to reduce both response amplitude and cone excursion. By 13hz the cone excursion is reduced by 30%, by 10hz it is reduced by 50%, by 5hz it is reduced by over 85%.
Thanks for the curves, theoretical though they may be.
If I am reading the colours correctly, I'd say they are a good argument for not using anti-resonant boxes for low frequencies because they go wonky in the subsonic range... dealing with which is what this thread is about.
You get wild, undesirable, uncontrollable cone flaps but none of the sound down there. Unlike sealed boxes or Klipschorns (that also have sealed boxes deep inside them).
If I am reading the colours correctly, I'd say they are a good argument for not using anti-resonant boxes for low frequencies because they go wonky in the subsonic range... dealing with which is what this thread is about.
You get wild, undesirable, uncontrollable cone flaps but none of the sound down there. Unlike sealed boxes or Klipschorns (that also have sealed boxes deep inside them).
K horns aren't going to be any better at cone control at 10 hz. Sealed boxes might have more control, but not much spl. Pick your poison.
>By 13 hz the cone excursion is reduced by 30%, by 10hz it is reduced by 50%
> Thanks for the curves, theoretical though they may be
A theory that needs to be applied for EBS
> a good argument for not using anti-resonant boxes for low frequencies because they go wonky in the subsonic range
wonky?
You get wild, undesirable, uncontrollable cone flaps but none of the sound down there
They wont be uncontrollable with an appropriate filter. And why would they *not reproduce the sound down there - soundtracks?
> Thanks for the curves, theoretical though they may be
A theory that needs to be applied for EBS
> a good argument for not using anti-resonant boxes for low frequencies because they go wonky in the subsonic range
wonky?
You get wild, undesirable, uncontrollable cone flaps but none of the sound down there
They wont be uncontrollable with an appropriate filter. And why would they *not reproduce the sound down there - soundtracks?
Same as before, but upped the power 3dB and added a sealed box for grins and giggles.
(note: if you apply EQ to the sealed to get some bass, the excursion will climb through the roof)
(note: if you apply EQ to the sealed to get some bass, the excursion will climb through the roof)
I'm over my head here, so I hope a tech-inclined person will correct me (well, correct all of us).
The Klipschorn cone is well controlled at all frequencies by the small sealed box behind the cone and even at frequencies below which the front horn-air-load is no longer much.
Watching cones flap about, that's what I mean by wonky behaviour. You don't want the cone to be loose in light of many mechanical and electrical reasons, and sonic reasons include Doppler (I'm not sure the status of Doppler effects esp. now-a-days that small woofer cones are in style). Adding electric LF filtering helps only to a point because you really need the acoustic control of flapping downstream (except for motional feedback, ahem ahem, sadly there's no feedback from the cone... and none ever possible from resonant boxes). Flapping bad, even if some deranged mixer added it to your movie soundtrack. Best to use a very sharp LF cutoff that removes as much signal as you can below the useful range of the woofer.
When djk cheerily points out that sealed boxes with dramatic amounts of LF EQ have big excursions "that will climb through the roof" - a very comic image indeed seeing a large woofer enclosure climbing my cluttered walls to reach the roof - it is just a matter of what EQ he chooses to plot (esp. if that's a 5 inch woofer plotted there). Big and growing EQ leads to big and growing excursions. It would be nice to see the curves/assumptions described and the colours identified too.
The Klipschorn cone is well controlled at all frequencies by the small sealed box behind the cone and even at frequencies below which the front horn-air-load is no longer much.
Watching cones flap about, that's what I mean by wonky behaviour. You don't want the cone to be loose in light of many mechanical and electrical reasons, and sonic reasons include Doppler (I'm not sure the status of Doppler effects esp. now-a-days that small woofer cones are in style). Adding electric LF filtering helps only to a point because you really need the acoustic control of flapping downstream (except for motional feedback, ahem ahem, sadly there's no feedback from the cone... and none ever possible from resonant boxes). Flapping bad, even if some deranged mixer added it to your movie soundtrack. Best to use a very sharp LF cutoff that removes as much signal as you can below the useful range of the woofer.
When djk cheerily points out that sealed boxes with dramatic amounts of LF EQ have big excursions "that will climb through the roof" - a very comic image indeed seeing a large woofer enclosure climbing my cluttered walls to reach the roof - it is just a matter of what EQ he chooses to plot (esp. if that's a 5 inch woofer plotted there). Big and growing EQ leads to big and growing excursions. It would be nice to see the curves/assumptions described and the colours identified too.
Ben
> now-a-days that small woofer cones are in style
Im intending to use a driver a got a coupla years ago, the Peerless XLS, 12” with 12 mm Xmax. Though much bigger Xmax drivers are available now.
> Best to use a very sharp LF cutoff that removes as much signal as you can below the useful range of the woofer.
Yes, that’s my intent.
What do you mean by “the *acoustic control* of flapping downstream”?
> now-a-days that small woofer cones are in style
Im intending to use a driver a got a coupla years ago, the Peerless XLS, 12” with 12 mm Xmax. Though much bigger Xmax drivers are available now.
> Best to use a very sharp LF cutoff that removes as much signal as you can below the useful range of the woofer.
Yes, that’s my intent.
What do you mean by “the *acoustic control* of flapping downstream”?
otto88 said:
You suggested (electronic) filtering to control flapping. That only controls the electric input and only stuff upstream of the filter, not in the amp, turn-on transients, etc. Moreover, the frequency range where flapping gets going falls right square in the music range. In other words, you can cut the real low freqencies but not higher.
For resonant boxes, TLs, and baffles (and unlike sealed boxes), as I poorly understand these things, there is no brake on the cone motion below the anti-resonant or unloading frequency (often the same as the speaker resonance) and the cone is just a floppy mass except as limited by the restoring mechanisms (spider and maybe the surround). So any nudge, electric or mechanical, can set the cone in motion at those low frequencies.
Aside from Doppler, you just don't want your cone shifting about in the magnetic gap, reacting to a rhythms besides your source, or otherwise misbehaving.
You mentioned “wild, undesirable, uncontrollable cone flaps”
I thought by cone flaps, you were referring to over excursion. I’d suggested an electronic filter to control *over excursion below c 13 Hz.
You’re right, a filter won’t prevent turn-on transients. I thought a reasonable amp should have a circuit to prevent them. I should check that with the one I have in mind.
Might cone flapping have other triggers?
I thought by cone flaps, you were referring to over excursion. I’d suggested an electronic filter to control *over excursion below c 13 Hz.
You’re right, a filter won’t prevent turn-on transients. I thought a reasonable amp should have a circuit to prevent them. I should check that with the one I have in mind.
Might cone flapping have other triggers?
Otto,
DC offsets in the cone position can be caused by a number of nonlinearities in the woofer itself. Since these are dynamic and depend on frequency and drive level, they can end up as cone flapping. Basically, asymettries in the suspension stiffness, motor force (BL linearity) or inductance linearity become rectified by the driver and cause the DC offsets. Since they are the result of design choices made in the driver, there is no way to filter them out. DC offsets can also be caused by asymmetric port behavior. A flare on the outside end, but not on the inside for example.
If you can post a schematic to the plate amplifier you are considering, I'm sure there are a number of people here who can point out exactly what resistor values to change to set the rumble filter to whatever Q and corner frequency you need.
DC offsets in the cone position can be caused by a number of nonlinearities in the woofer itself. Since these are dynamic and depend on frequency and drive level, they can end up as cone flapping. Basically, asymettries in the suspension stiffness, motor force (BL linearity) or inductance linearity become rectified by the driver and cause the DC offsets. Since they are the result of design choices made in the driver, there is no way to filter them out. DC offsets can also be caused by asymmetric port behavior. A flare on the outside end, but not on the inside for example.
If you can post a schematic to the plate amplifier you are considering, I'm sure there are a number of people here who can point out exactly what resistor values to change to set the rumble filter to whatever Q and corner frequency you need.
Jack Hidley said:
Hi,
I am also going to use the Jaycar AA0508 sub plate (already have) and thinking of using it with a peerless 830500 and a passive resonator (still trying to decide between a "criticlal Q" and a resonator design). I have found the schematics for the AA0508 if anyone could suggest modifications or improvements:
http://www.customanalogue.com/files/Sub-Amp.zip
Thanks.
That same site has the mods for the AA-0508 amp....
http://www.customanalogue.com/critical_q_sub_av_v.htm
http://www.customanalogue.com/critical_q_sub_av_v.htm
Collo said:
Are these mods appropriate if I am not making a "critical Q" Sub?
He explains what some of the mods do but not all of them.
For example:
change R1,R2 from 6K8 to 1K = controls frequency range.
But no more info is given on what that mod does to the frequency range.
Thanks.
It seems the instructions I linked to are not as useful as the ones that come with the Parts Express amps (which are slightly different to the Jaycar ones). For example the PE 240w amp has a pdf containing a table to allow you to change the centre frequency and gain of the boost circuit
The changes outlined in the Joe Rasmussen Jaycar AA-0508 instructions will give a flatter amp, as shown in the technical notes for the changes
Unfortunately it doesn't add in any changes for a high-pass filter, which is what the OP was asking for.
For that, a separate HP filter would be needed.
Here's my effort, a 24 dB/octave Linkwitz-Riley HP filter. The page also links on to Rod Elliot's site where a crossover board can be pressed into service if you would rather not work with veroboard..
The changes outlined in the Joe Rasmussen Jaycar AA-0508 instructions will give a flatter amp, as shown in the technical notes for the changes
Unfortunately it doesn't add in any changes for a high-pass filter, which is what the OP was asking for.
For that, a separate HP filter would be needed.
Here's my effort, a 24 dB/octave Linkwitz-Riley HP filter. The page also links on to Rod Elliot's site where a crossover board can be pressed into service if you would rather not work with veroboard..
- Status
- Not open for further replies.