The crossover shown at the following URL was used in the 1970s in a
three way, eight ohms system. There are some design details I would
like to know the probable explanation for. One is the treble part which
is fourth order. Why would one want to do it this way, when the other
two channels are second order? And when I look at the formulas
recommended for calculating fourth order networks, they always have
different component values for the first and the second pair of
cap/coil. Here the cap and coil pairs have identical values. What could
the reason for this be?

Secondly, there is a kind of resonant circuit in the midrange part (C4
& L3). This seems to be a band reject filter, but how do I find out its
characteristics, and what could the reason be for including it?

Also, I seem to remember that second order networks shift phase 180
degrees. How about fourth order filters?

Maybe someone can explain this to me.

http://goto.glocalnet.net/bring/pics/FN12B.pdf

Sven

"sbring" wrote in message
oups.com...
The crossover shown at the following URL was used in the 1970s in a
three way, eight ohms system. There are some design details I would
like to know the probable explanation for. One is the treble part which
is fourth order. Why would one want to do it this way, when the other
two channels are second order? And when I look at the formulas
recommended for calculating fourth order networks, they always have
different component values for the first and the second pair of
cap/coil. Here the cap and coil pairs have identical values. What could
the reason for this be?

Secondly, there is a kind of resonant circuit in the midrange part (C4
& L3). This seems to be a band reject filter, but how do I find out its
characteristics, and what could the reason be for including it?

Also, I seem to remember that second order networks shift phase 180
degrees. How about fourth order filters?

Maybe someone can explain this to me.

http://goto.glocalnet.net/bring/pics/FN12B.pdf

Sven

I seem to think that the 4th order filter gives a flatter response than the
2nd order, and that this is good when used in the tweeter filter, but this
is a dim recollection from a long time ago, so may not be a valid reason.
Different values for L and C will give a different response to the filter,
broadening the rolloff. Using the same values will, I think, make the
rolloff steeper. The 4th order filter has a 360 deg phase shift, whereas the
2nd order has 180 deg. This is easily compensated for in mixed order
crossovers, by reversing the connections to the 2nd order filtered drivers,
with respect to the tweeter connections.

The parallel LC circuit in the midrange, has a resonant frequency of about
12.78 kHz. I suspect that it may be there to compensate for some unwanted '
hump ' in the midrange driver's characteristic, possibly due to natural
resonance of the driver, or an unwanted mechanical resonance in the
particular enclosure.

Bit vague I'm afraid - it's not really my area of expertise, just something
that I have had an interest in, in the past. Still, might get you started in
the right direction, and I'm sure that there is probably someone else out
there, who can tell you more than me.

Arfa

In article .com,
sbring
wrote:
The crossover shown at the following URL was used in the 1970s in a
three way, eight ohms system. There are some design details I would like
to know the probable explanation for. One is the treble part which is
fourth order. Why would one want to do it this way, when the other two
channels are second order?

Can't comment in detail as we don't know all the relevant details of the
speaker. However bear in mind that the behaviour is also affected by the
inherent response of the speaker units not being flat, and the units not
being the same as a resistive load. Thus there may be some additional
'orders' built into the behaviour which are not obvious from just looking
at the crossover diagram.


And when I look at the formulas recommended for calculating fourth order
networks, they always have different component values for the first and
the second pair of cap/coil. Here the cap and coil pairs have identical
values. What could the reason for this be?

Alignment. Details of loads/speakers as outlined above.

Secondly, there is a kind of resonant circuit in the midrange part (C4 &
L3). This seems to be a band reject filter, but how do I find out its
characteristics, and what could the reason be for including it?

Also, I seem to remember that second order networks shift phase 180
degrees. How about fourth order filters?

The phase shifts vary with frequency so I assume you are referring to the
nominal crossover or rollover point. The phase shift there will tend to
scale with order, but bear in mind that Linkwitz-Riley, etc, usage may also
adjust the relative turn-over points to achive the overall results desired.

Maybe someone can explain this to me.

http://goto.glocalnet.net/bring/pics/FN12B.pdf

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html