Speaker Cross Over - UK Vintage Radio Repair and Restoration Discussion Forum

sp10mk11 · 10th May 2018, 4:46 pm

Hello sorry it is a bad drawing, but this circuit feeds the bass unit, does it look ok? sorry I cannot find the value of the Inductor.

I was wondering what the two 47uf caps do aswell the one with a 5.6 ohm resistor on it

Thanks
Gary

mark_in_manc · 10th May 2018, 5:01 pm

Looks OK to me.

The left-most inductor and capacitor form a second-order low-pass. (Note - 2nd order crossovers invert one of the driver polarities, because the phase between the hi-pass and low-pass sections is 180deg at cross-over. If you don't invert one of the drivers, they cancel at the x-over freq).

The right-most C and R in series are a 'Zobel network) which tries to make the driver look a bit more like it's nominal (resistive) impedance at x-over freq, by cancelling out the L_E (voice coil inductance) term in its electrical input impedance. R+jwL is a bit of a simplification of the electrical components of the driver electrical impedance (mechanical and acoustic terms have an influence too, but less so away from resonance, especially for mech terms) - but it will do.

Hope that helps.

vinrads · 10th May 2018, 5:50 pm

Phew I was just going to say that

It just shows you are never to old to learn .Mick.

sp10mk11 · 10th May 2018, 9:08 pm

Hello
Thanks very much for that explanation, so it is a sort of tone control/filter, does that mean by altering the values of the caps you would alter the cut off frequency? say it is 400hz it would perhaps go to 500hz with a pair of higher value caps.

Thanks
Gary

mark_in_manc · 10th May 2018, 10:41 pm

Yes, you could push the x-over freq up a bit by reducing the value of the capacitor which is on its own, in its own branch. Leave the one in series with the resistor alone. You'd ideally want to drop the inductance of the coil a little too.

There's a bit more background here

https://en.wikipedia.org/wiki/Linkwi...93Riley_filter

The equations relating x-over freq Fx, inductance L, capacitance C and driver nominal impedance R are pretty simple:

L=(2*R)/(2*pi*Fx)

C=1/(2*R*2*pi*Fx)

I used to be able to derive those but I've lost the knack - I'll dig out a proof if you really want one, but it will have 'jw's all over it

While looking for those equations I also stumbled on those for the Zobel network:

The resistor will be equal to the dc coil impedance R_E
The capacitor will have value L_E/(R_E^2) - where L_E is the voice coil inductance, a rough model of the rising impedance of the voice coil at HF.

cheers
Mark

10th May 2018, 4:46 pm	#1
sp10mk11 Octode Join Date: Feb 2009 Location: Cornwall, UK. Posts: 1,275	Speaker Cross Over Hello sorry it is a bad drawing, but this circuit feeds the bass unit, does it look ok? sorry I cannot find the value of the Inductor. I was wondering what the two 47uf caps do aswell the one with a 5.6 ohm resistor on it Thanks Gary Attached Thumbnails

10th May 2018, 5:01 pm	#2
mark_in_manc Octode Join Date: Nov 2011 Location: Manchester, UK. Posts: 1,875	Re: Speaker Cross Over Looks OK to me. The left-most inductor and capacitor form a second-order low-pass. (Note - 2nd order crossovers invert one of the driver polarities, because the phase between the hi-pass and low-pass sections is 180deg at cross-over. If you don't invert one of the drivers, they cancel at the x-over freq). The right-most C and R in series are a 'Zobel network) which tries to make the driver look a bit more like it's nominal (resistive) impedance at x-over freq, by cancelling out the L_E (voice coil inductance) term in its electrical input impedance. R+jwL is a bit of a simplification of the electrical components of the driver electrical impedance (mechanical and acoustic terms have an influence too, but less so away from resonance, especially for mech terms) - but it will do. Hope that helps.

10th May 2018, 5:50 pm	#3
vinrads Dekatron Join Date: Feb 2009 Location: Chesterfield, Derbyshire, UK. Posts: 3,766	Re: Speaker Cross Over Phew I was just going to say that It just shows you are never to old to learn .Mick.

10th May 2018, 9:08 pm	#4
sp10mk11 Octode Join Date: Feb 2009 Location: Cornwall, UK. Posts: 1,275	Re: Speaker Cross Over Hello Thanks very much for that explanation, so it is a sort of tone control/filter, does that mean by altering the values of the caps you would alter the cut off frequency? say it is 400hz it would perhaps go to 500hz with a pair of higher value caps. Thanks Gary

10th May 2018, 10:41 pm	#5
mark_in_manc Octode Join Date: Nov 2011 Location: Manchester, UK. Posts: 1,875	Re: Speaker Cross Over Yes, you could push the x-over freq up a bit by reducing the value of the capacitor which is on its own, in its own branch. Leave the one in series with the resistor alone. You'd ideally want to drop the inductance of the coil a little too. There's a bit more background here https://en.wikipedia.org/wiki/Linkwi...93Riley_filter The equations relating x-over freq Fx, inductance L, capacitance C and driver nominal impedance R are pretty simple: L=(2R)/(2piFx) C=1/(2R2pi*Fx) I used to be able to derive those but I've lost the knack - I'll dig out a proof if you really want one, but it will have 'jw's all over it While looking for those equations I also stumbled on those for the Zobel network: The resistor will be equal to the dc coil impedance R_E The capacitor will have value L_E/(R_E^2) - where L_E is the voice coil inductance, a rough model of the rising impedance of the voice coil at HF. cheers Mark