>....it's this bit I don't get :
""you'll get the 2nd order transient response of the box combined with the 2nd order electrical ""
Does that mean with a box of 0.5, and a hi pass of 0.7 I'd get a Q of 0.35 ?
====
Well, you'll get an 18dB/octave total roll off (Q = 0.35), but the transient response will be whatever the box says it is above Fb, IOW if the electrical filter begins at Fb, then Qtc = 0.5 in this case, but if it begins at F3, then Qtc = (0.35*0.5)^0.5 = ~0.418.
====
I actually want 'the roll of characteristics' of 24Db/oct with a Q of 0.49.
====
Then either use Qtc = 0.7 with a 0.7 HP or Qtc = 0.5 with a 1.0 HP.
GM
""you'll get the 2nd order transient response of the box combined with the 2nd order electrical ""
Does that mean with a box of 0.5, and a hi pass of 0.7 I'd get a Q of 0.35 ?
====
Well, you'll get an 18dB/octave total roll off (Q = 0.35), but the transient response will be whatever the box says it is above Fb, IOW if the electrical filter begins at Fb, then Qtc = 0.5 in this case, but if it begins at F3, then Qtc = (0.35*0.5)^0.5 = ~0.418.
====
I actually want 'the roll of characteristics' of 24Db/oct with a Q of 0.49.
====
Then either use Qtc = 0.7 with a 0.7 HP or Qtc = 0.5 with a 1.0 HP.
GM
RobWells said:
What ever the others are saying I can't agree with.
The class of 'critically damped' filters you are referring to
(no overshoot for a step response, phase response equivalent to a constant time delay)
Are known as Bessel or Thompson filters.
Cascading two 0.7 butterworths gives you the Linkwitz/Riley
aligment which is not by any means 'critically damped', and
does not have a Q of 0.5.
You cannot equate the Q of a 2nd order to the q of a 4th order.
For a 4th order Bessel you cascade Q=0.52 and Q=0.81.
http://www.rane.com/note147.html
🙂 sreten.
I'm starting to get even more confused now.😕
I've been looking through the 'active filter cookbook' and it seems quite easy to vary the Q of the filter by adjusting a pair of resistors. I've got to admit here that I am not confident on these circuits, so wish to stay as "cookbook' as possible.
Some background info may help.
The speaker is to be a 3 way, with the tweet crossed to the mid around 3Khz. (will be determined once the enlcosure is built and it's response measured) I want to use a 4th order LR x/o for this.
The bass units are finished, and I wish to cross to them as low as possible, hence trying to incorporate my mids natural roll off with an additional 2nd order slope added so that I can do a 4th order LR from bass to mid. This would be around the 85Hz mark (F-3 of the box.)
I wanted to build the mid box with a Q=0.5 as it is meant to give the 'tightest' sound. I'm still trying to work out what will happen to this 'tightness' if I add the 85Hz high pass x/o to it, and whether the 'Q' of this filter affects the 'tightness' of the mid, or just would produce a slight ripple in freq response at the x/o point. Some say it affects the mid/some say it doesn't. I haven't got a clue, and the books I have here (admittedly not many) do not discuss this either. - or if they do I just don't 'get iit'
😀
It's looking like I'll just do a 2nd order Q=0.7 filter and re-do my mid enclosure to a Q=0.7 volume.
Cheers
Rob
I've been looking through the 'active filter cookbook' and it seems quite easy to vary the Q of the filter by adjusting a pair of resistors. I've got to admit here that I am not confident on these circuits, so wish to stay as "cookbook' as possible.
Some background info may help.
The speaker is to be a 3 way, with the tweet crossed to the mid around 3Khz. (will be determined once the enlcosure is built and it's response measured) I want to use a 4th order LR x/o for this.
The bass units are finished, and I wish to cross to them as low as possible, hence trying to incorporate my mids natural roll off with an additional 2nd order slope added so that I can do a 4th order LR from bass to mid. This would be around the 85Hz mark (F-3 of the box.)
I wanted to build the mid box with a Q=0.5 as it is meant to give the 'tightest' sound. I'm still trying to work out what will happen to this 'tightness' if I add the 85Hz high pass x/o to it, and whether the 'Q' of this filter affects the 'tightness' of the mid, or just would produce a slight ripple in freq response at the x/o point. Some say it affects the mid/some say it doesn't. I haven't got a clue, and the books I have here (admittedly not many) do not discuss this either. - or if they do I just don't 'get iit'
😀 It's looking like I'll just do a 2nd order Q=0.7 filter and re-do my mid enclosure to a Q=0.7 volume.
Cheers
Rob
While the cascaded butterworth 0.707*0.707=0.5, this does not imply that the transient response is the same as a 2nd order box with Q=0.5. Q isn't really defined as such for a 4th order equation.
The box response is not important in this case. The difference between a sealed box Q of 0.5 and one of 0.7 is hardly significant. Going for a Qtc of 0.5 on a driver that is highpassed is a waste of box volume.
The box response is not important in this case. The difference between a sealed box Q of 0.5 and one of 0.7 is hardly significant. Going for a Qtc of 0.5 on a driver that is highpassed is a waste of box volume.
Rob,RobWells said:
I wanted to build the mid box with a Q=0.5 as it is meant to give the 'tightest' sound. I'm still trying to work out what will happen to this 'tightness' if I add the 85Hz high pass x/o to it, and whether the 'Q' of this filter affects the 'tightness' of the mid, or just would produce a slight ripple in freq response at the x/o point. Some say it affects the mid/some say it doesn't. I haven't got a clue, and the books I have here (admittedly not many) do not discuss this either. - or if they do I just don't 'get iit'
It's looking like I'll just do a 2nd order Q=0.7 filter and re-do my mid enclosure to a Q=0.7 volume.
Cheers
Rob
Put the Q of the filter aside for a moment. Build your mid-box with a Q of .5 to get the transient response you desire. Note the (theoretical) FR of this alignment.
The electrical filter will alter the overall FR so you will have to adjust the cutoff frequency to get the desired overall FR, but your transient response will be that of the box alignment.
In the end you will need to adjust/tweak the electrical filter to accommodate for all of the system/room variables, but the mids will be tight.
Ron E said:
Hi Ron,
You see originally I planned to just use the natural 2nd order of the box to cross to my bass speakers, but decided to change it in light of the vd of the mid - I listen loud, and want to keep the mid 'clean' at higher volumes. As to a waste of box volume, we're only talking a difference of about 6L - a total box vol of 11L - it's not huge!(and it's almost finished too😉 )
Thanks for the input🙂
Rob
This is a portion of filter info from Snippets:
Note that you can also cascade *acoustic* filters with the active electronic filters. For example, lets say I had a sealed midrange that I wanted to cross to my woofer with a 4th L-R. I could start by designing a Butterworth sealed cabinet. This is the most standard sealed cabinet alignment generally discussed, which results in an *acoustic* second-order lowcut with a Q of .707. Cascading this with a 2cd order Butterworth active filter lowcut, yields a 4th L-R lowcut total acoustical response. This would mate nicely with an active 4th L-R highcut on your woofer. This is a very elegant and simple solution for mid to woofer crosses.
The resultant Q would be .5.
If you stayed with a Q of .5 on the box and used a Q=.7 on the filter you wind up with a Q of .35. If you want a final Q of .5 and want to keep the box at .5, the filter Q needs to be 1.
If you want to raise the Q of the box to .7 without rebuilding it, simply put something solid inside the box to reduce it's internal volume.
Note that you can also cascade *acoustic* filters with the active electronic filters. For example, lets say I had a sealed midrange that I wanted to cross to my woofer with a 4th L-R. I could start by designing a Butterworth sealed cabinet. This is the most standard sealed cabinet alignment generally discussed, which results in an *acoustic* second-order lowcut with a Q of .707. Cascading this with a 2cd order Butterworth active filter lowcut, yields a 4th L-R lowcut total acoustical response. This would mate nicely with an active 4th L-R highcut on your woofer. This is a very elegant and simple solution for mid to woofer crosses.
The resultant Q would be .5.
If you stayed with a Q of .5 on the box and used a Q=.7 on the filter you wind up with a Q of .35. If you want a final Q of .5 and want to keep the box at .5, the filter Q needs to be 1.
If you want to raise the Q of the box to .7 without rebuilding it, simply put something solid inside the box to reduce it's internal volume.
roddyama said:
Whilst I agree with paragraphs 1 and 3, paragraph 2 is incorrect.
To significantly affect the power handing of the midrange unit
the transient response will be affected.
The missing link here is the baffle step compensation of the mid
unit which will throw most of this discussion out of the window.
🙂 sreten.
sreten said:
On the Linkwitz site he applies the baffle step (in his case floor mounted woofer to mid 100-200Hz eq to the full range signal before it is fed to the x/o sections. I was planning to do likewise. How does this throw the discussion out of the window? - I'm trying to learn here, not waste peoples time 😀
Cheers
Rob
http://www.diyaudio.com/forums/attachment.php?postid=287151
Damn, Rob, you do that layered wood bit better than anybody!
Damn, Rob, you do that layered wood bit better than anybody!
Thanks Kelticwizard,😱
It's having access to the cad/laser cutter at work that takes all the effort away..
Rob
It's having access to the cad/laser cutter at work that takes all the effort away..
Rob
RobWells said:
Well I was taking note of the first point when referring to the second.
Sorry for not explaining myself, I was getting frustrated after spending half an hour
trying to find the Fo and Q requirements for a cascaded 2nd order 4th order Bessel filter.
If you are a volume fiend running your mids down to below 100Hz is not good idea.
You will need 6dB boost by 200Hz for BSC on the mids for a start.
Bass heavy music excepted, peak excursion requirements are often in the 100 to 200Hz range.
Running such a low c/o frequency will likely severely impair maximum levels.
At high levels distortion will be poor compared to c/o at say 200 to 250Hz.
Depending on the amplifier ratings for each unit a higher c/o frequency
is very likely to yield a significantly higher maximum SPL capability.
🙂 sreten.
Hi Sreten,
The 'mid' unit is actually a 'mid-bass' unit, and should do approx 103Db at 80Hz (linear) Taking into account that it will be 6Db down at 83ish, this means that the final speaker will acheive 109Db linear, and between 103 and 106 depending on the amount of baffle step correction. I tend to do my 'loud' assements using the thx reference level for theaters method - ie: At cinema reference the speaker should do 105Db at peaks(with average levels@ 80Db). Most people I know watch films at about 10Db below reference, and that is still loud🙂
WRT sensitivity, that is more of an issue - The speaker is 85Db/watt, so it would be 82/79Db including baffle step comp. This means my amp would need to hit 128W/256W for 103 Db.
Of course this ignores boundary gain etc, and that can only be determined once the speakers are finished and in place.(and measured)
As an aside, I originally bought a pair of focal 6wm mids mainly for their sensitivity (95Db), and chose again due to their break up at approx 1500Hz, pushing towards a lower x/o point. I can always resurrect these if I need more sensitivity😀
Cheers
Rob
The 'mid' unit is actually a 'mid-bass' unit, and should do approx 103Db at 80Hz (linear) Taking into account that it will be 6Db down at 83ish, this means that the final speaker will acheive 109Db linear, and between 103 and 106 depending on the amount of baffle step correction. I tend to do my 'loud' assements using the thx reference level for theaters method - ie: At cinema reference the speaker should do 105Db at peaks(with average levels@ 80Db). Most people I know watch films at about 10Db below reference, and that is still loud🙂
WRT sensitivity, that is more of an issue - The speaker is 85Db/watt, so it would be 82/79Db including baffle step comp. This means my amp would need to hit 128W/256W for 103 Db.
Of course this ignores boundary gain etc, and that can only be determined once the speakers are finished and in place.(and measured)
As an aside, I originally bought a pair of focal 6wm mids mainly for their sensitivity (95Db), and chose again due to their break up at approx 1500Hz, pushing towards a lower x/o point. I can always resurrect these if I need more sensitivity😀
Cheers
Rob
well I'm not going to argue with a man who seems to know what he is doing.
The mids do look very small from your picture of your excellent cabinets.
I remember reading a conclusion from Martin Colloms that there was a
strong correlation between drive power and distortion, and that highish
efficiency was the simplest route to low distortion at a particular level.
Still, for a Bessel 4th order alignment you need two second order filters,
one at Q=0.5 the other at Q=0.8. I still can't find the related Fo
requirements to tell you what the actual frequency is.
But your mid cabinet will certainly do as the 0.5 section.
I'll try again tommorrow. there seems to be suggestions the filter should
be misaligned - this would relate to the bass unit filter - to give a flatter
overall summed power response.
🙂 sreten.
P.S. 103Db at 80Hz (linear) on a flat baffle = 97dB freefield in my book.
(BSC again !)
The mids do look very small from your picture of your excellent cabinets.
I remember reading a conclusion from Martin Colloms that there was a
strong correlation between drive power and distortion, and that highish
efficiency was the simplest route to low distortion at a particular level.
Still, for a Bessel 4th order alignment you need two second order filters,
one at Q=0.5 the other at Q=0.8. I still can't find the related Fo
requirements to tell you what the actual frequency is.
But your mid cabinet will certainly do as the 0.5 section.
I'll try again tommorrow. there seems to be suggestions the filter should
be misaligned - this would relate to the bass unit filter - to give a flatter
overall summed power response.
🙂 sreten.
P.S. 103Db at 80Hz (linear) on a flat baffle = 97dB freefield in my book.
(BSC again !)
Alternatively I could cross at 200Hz, however, this would mean a non perfect x/o - the mid will be dropping about 2-3Db at the -24 point - I doubt it would be heard in room, what with the room causing way more ups n' downs than any filter could🙂
I'd rather keep the mid as low as I can though..
Cheers
rob
I'd rather keep the mid as low as I can though..
Cheers
rob
Attachments
""well I'm not going to argue with a man who seems to know what he is doing.""
If I was that sure I wouldn't be asking here😉
""The mids do look very small from your picture of your excellent cabinets.""
They're 5.5" units - I'm using the s-speak revelator tweeter which has a large faceplate, making the mid look tiny.
http://www.d-s-t.com/scs/data/15w_8530k00a.htm
http://www.d-s-t.com/scs/data/d2905_990000a.htm
Cheers
Rob
If I was that sure I wouldn't be asking here😉
""The mids do look very small from your picture of your excellent cabinets.""
They're 5.5" units - I'm using the s-speak revelator tweeter which has a large faceplate, making the mid look tiny.
http://www.d-s-t.com/scs/data/15w_8530k00a.htm
http://www.d-s-t.com/scs/data/d2905_990000a.htm
Cheers
Rob
""P.S. 103Db at 80Hz (linear) on a flat baffle = 97dB freefield in my book.
(BSC again !)""
Yes but at 80Hz the speaker will be -6Db to the rest of the output for the LR x/o🙂 (the bass drivers will be doing half the work...)
I used the chart kelticwizard posted here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=5668&highlight=SPL+Volume+Chart
Cheers
to determine the linear output (my vd is 61.75cc's - .65cm(lin xmax ) * 95cm^2 sd )
Rob
(BSC again !)""
Yes but at 80Hz the speaker will be -6Db to the rest of the output for the LR x/o🙂 (the bass drivers will be doing half the work...)
I used the chart kelticwizard posted here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=5668&highlight=SPL+Volume+Chart
Cheers
to determine the linear output (my vd is 61.75cc's - .65cm(lin xmax ) * 95cm^2 sd )
Rob
- Status
- Not open for further replies.