OK this may be a newbie question but I have been unable to find the answer, perhaps some-one can help steer me to the information on line or provide it here.
I have been playing with a third order crossover on a second rebuild of an experimental OB 3 way.
I built the cross-over according to normal usage placing the smaller capacitor first and the larger value after the parallel inductor, then I started thinking along mechanical filter line and wondered what would happen if I used the physical analogy and placed the larger capacitor first in line, so I did so and listened again.
By my ear the reverse of normal practice sounds much better smoother and with less ringing on certain cymbal sounds and bells ( I was listening to Mike Oldfeild and "Tubular Bells ) can some one try and explain to me in as symple as possible a manner??
Thanx in advance
Ted
I have been playing with a third order crossover on a second rebuild of an experimental OB 3 way.
I built the cross-over according to normal usage placing the smaller capacitor first and the larger value after the parallel inductor, then I started thinking along mechanical filter line and wondered what would happen if I used the physical analogy and placed the larger capacitor first in line, so I did so and listened again.
By my ear the reverse of normal practice sounds much better smoother and with less ringing on certain cymbal sounds and bells ( I was listening to Mike Oldfeild and "Tubular Bells ) can some one try and explain to me in as symple as possible a manner??
Thanx in advance
Ted
This sounds strange. Normally, what you are doing would create a serious electrical resonance in the filter.
What component values are you using? Any series (or parallel) resistors? Do you have an impedance curve for the tweeter?
Standard passive 3rd order high pass filters often need some trimming to avoid ringing, as component values are quite critical. The inductor should not be smaller than calculated - it is better to use a SLIGHTLY larger value. The first capacitor (see from the amplifier) should normally be a little less than 1/3 of the second capacitor.
What component values are you using? Any series (or parallel) resistors? Do you have an impedance curve for the tweeter?
Standard passive 3rd order high pass filters often need some trimming to avoid ringing, as component values are quite critical. The inductor should not be smaller than calculated - it is better to use a SLIGHTLY larger value. The first capacitor (see from the amplifier) should normally be a little less than 1/3 of the second capacitor.
Moondog55 said:
This does sound strange. I would see if you could find another enthusiest in your area with measurment capability... as previously stated you might find that you have lot of peaking in the crossover at the crossover frequency...
I've found calculated crossovers to be of zero value, the important acoustic slope is the combination of the electrical response of the crossover and the acoustic response of the driver...and calculated crossovers don't take into account the driver response....
That said...
If you like the way it sounds...and you don't think that you're causing a problem with impedance then it could be just what the Dr. ordered.
Scott
PaleRider said:The bigger cap might have a better power-handlig capasity. When placed after the big one the smaller will not see a full signal anylonger and given this easyer job it can handle more power without screaming
Doubtful. The current carrying capacity of the capacitor (say that 3 times fast!) is most assuredly sufficient for the job.
The dynamic or max power capasity in passive filters are underrated by many builders
I agree. Typically when using ferrite core inductors which saturate quite easily, and quite harsh sounding when they do. I have some 15mH ferrite cores that might be waiting a long time for use again....they sounded like hard clipping when saturating. Iron cores are better, and of course...air cores are best..but with copper pricess $$$$
Scott
In this case caps are NP 100 Volts, I chose 3rd order to stop driver bottoming out as its rated power handlind is 3watts when used full range.
And I experimented becauase driver was a dollar as part of a grab box.
I have to say that usually I only hear ringing in drivers when using bypass caps of different types; I have put this down to different grades of capacitor having different rise times thus giving resonance,
Still don't understand the why common/ standard method works, I find electronics counter intuitive at times
And I experimented becauase driver was a dollar as part of a grab box.
I have to say that usually I only hear ringing in drivers when using bypass caps of different types; I have put this down to different grades of capacitor having different rise times thus giving resonance,
Still don't understand the why common/ standard method works, I find electronics counter intuitive at times
Imagine a 12db filter...do you place the paralel inductor before the cap...no, you place the series cap first and THEN the "smaller" paralel inductor....and 18db filter is the same with just another smaller cap afterwards(last)...different values ofcourse
On low pass filters its the same...biggest inductors first
On low pass filters its the same...biggest inductors first
Another part of the same query
Are there any other described types of third order crossovers other than Butterworth??
For instance the third order in my little IMF's used equal values of capacitor both before and after the parallel inductor, what characteristics would this have electrically?? and if I remember correctly so did some KEF cross-overs
Are there any other described types of third order crossovers other than Butterworth??
For instance the third order in my little IMF's used equal values of capacitor both before and after the parallel inductor, what characteristics would this have electrically?? and if I remember correctly so did some KEF cross-overs
LaLena 3rd order description
3rd Order Butterworth
2500 Hertz
8 Ohm Tweeter / 8 Ohm Woofer Parts List
--------------------------------------------------------------------------------
Capacitors
C1 = 5.31 uF
C2 = 15.92 uF
C3 = 10.61 uF
Inductors
L1 = 0.38 mH
L2 = 0.76 mH
L3 = 0.25 mH
Smaller capacitor first in line/ larger second in line after parallel inductor
3rd Order Butterworth
2500 Hertz
8 Ohm Tweeter / 8 Ohm Woofer Parts List
--------------------------------------------------------------------------------
Capacitors
C1 = 5.31 uF
C2 = 15.92 uF
C3 = 10.61 uF
Inductors
L1 = 0.38 mH
L2 = 0.76 mH
L3 = 0.25 mH
Smaller capacitor first in line/ larger second in line after parallel inductor
Hi,
The inductor value depends on the first capacitor having significant
impedance. If you lower this impedance you end up with a very
low impedance, the first capacitor does not prevent the inductor
"shorting" the input to ground at the crossover frequencies.
There are a whole raft of 3rd alignments, though Butterworth
is by far the most commonly used, but note that 3 components
on a tweeter are used typically for a 4th order acoustic fiter.
Real electrical filters on real drivers genarally align to a known
filter alignment acoustically, the electrical values and the order
of the electrical filter is not the type of filter.
e.g. see http://www.zaphaudio.com/audio-speaker17.html
The crossover is 4th order L/R acoustic @ 2kHz.
Compare the actual crossover to the garbage lalena comes up with.
/sreten.
The inductor value depends on the first capacitor having significant
impedance. If you lower this impedance you end up with a very
low impedance, the first capacitor does not prevent the inductor
"shorting" the input to ground at the crossover frequencies.
There are a whole raft of 3rd alignments, though Butterworth
is by far the most commonly used, but note that 3 components
on a tweeter are used typically for a 4th order acoustic fiter.
Real electrical filters on real drivers genarally align to a known
filter alignment acoustically, the electrical values and the order
of the electrical filter is not the type of filter.
e.g. see http://www.zaphaudio.com/audio-speaker17.html
The crossover is 4th order L/R acoustic @ 2kHz.
Compare the actual crossover to the garbage lalena comes up with.
/sreten.
Moondog55 said:
Hi,
You can only decribe the electrical orders, not the function.
The example given acoustically is 4th order L/R.
Electrically :
On the bass section it is 3rd order Elliptic / Cauer type.
(1st order followed by a 2nd order notch filter)
On the treble it is 2nd order.
As a description of a crossover function it does not tell
you much. It does describe the crossover electrically.
/sreten.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.