Altering Crossover for retrofit

[Autochthon]

Greetings, I'm rather new to all this so I'm looking for some advice.

I have recently purchase some replacement midrange and tweeters destroyed by the curious probings of a 2 years old. They were purchased as much for their grill coverings as their sound quality.

Unfortunately I have a frequency graph but no other figures.
Tweeter
Midrange
the mid-woofer is a bit for problematic I can't find a data sheet anywhere but I believe it is a 6.5 inch poly with a 20 ounce magnet model number MG 6020 and as I recall it stats where similar to this Woofer. I may be able to track down the data sheet I'll keep looking.

I had planned on using the existing cabinet and crossover circuits.

Cabinet ~ .56 cubic ft with 1.5" x 5" vent and was basically sized to the woofer.

I cross over circuit looks a bit simple to my eye but I don't know anything. No circuit software but here is the layout

Woofer has inductor of indeterminate value.
Mid has 16uf 50V electrolytic Cap
Tweeter has 8.1uf 50V electrolytic Cap followed by a 50ohm Pot.

With this limited information my best guess was that the low crossover is 1240 Hz and the high crossover is 2450 Hz.

question. Will this circuit work? Why no inductor in the high crossover?

I was going to test the sound "as-is" but from looking at the diagrams it occurred to me that it might help to boost the upper Xover a bit and perhaps attenuate the mid and put a notch filter at about 700hz as both mid and tweeter appear to have a spike there.

Can I get by with a first order Xover or will I need a second order?

Thank you

[infinia]

Originally posted by Autochthon

Woofer has inductor of indeterminate value.
Mid has 16uf 50V electrolytic Cap
Tweeter has 8.1uf 50V electrolytic Cap followed by a 50ohm Pot.

I was going to test the sound "as-is" but from looking at the diagrams it occurred to me that it might help to boost the upper Xover a bit and perhaps attenuate the mid and put a notch filter at about 700hz as both mid and tweeter appear to have a spike there.

Can I get by with a first order Xover or will I need a second order?

Thank you

You could get by with first order filters, but maybe not for tweet filter. All depends on the tweeter resonance, mid roll off responses and both showing smooth wideband behavour beyond the cross points. From your crossover circuit description seems like you can make quite a bit of improvement ditch the 50 ohm pot here.

What I would try, as always YMMV-
For a start add an inductor and a series resistor to mid, smaller cap 3 uF - 5 uF on tweeter with 8 ohm L pad? Possibly a known bigger inductor on woofer. When working blind all depends on...

[Autochthon]

Originally posted by infinia


What I would try, as always YMMV-
For a start add an inductor and a series resistor to mid, smaller cap 3 uF - 5 uF on tweeter with 8 ohm L pad? Possibly a known bigger inductor on woofer. When working blind all depends on...

Well my seat of the pants guess from the graphs is that the tweeter resonance is about 700 Hz and the mid resonance is about 750 Hz with the Mid having the stronger spike.

My understanding is that a a 3 - 5 uF cap would raise high xover to 4K to 6.5 K with a first order butterworth.

The LPad would be on the mid to lower the DB a little and the series resitor would be on the mid as a notch filter. Do I understand that right?

[Autochthon]

Originally posted by Autochthon


Well my seat of the pants guess from the graphs is that the tweeter resonance is about 700 Hz and the mid resonance is about 750 Hz with the Mid having the stronger spike.

oops, I meant 420 Hz on the Mids which likely would put it to low to matter. The Fs spike on the tweeter looks pretty small so perhaps a notch Filter is unnecessary.

[infinia]

Hi
With 1st order you have to spread the crossover points (-3dB) further away from each other than higher orders. So for example -mid LP could be set at 2KHz then tweeter HP sb about 1.7*2KHz. Most tweeters resonance is around 1KHz so this sb fine. IMO would not concern peaks in response with 1st order filters unless you want to get fancy and spend alot on Xover parts, when you could of had smoother drivers instead.

Simple example

L = Xm/(2*PI*fxo)
where: Xm = mids impedance at fxo, fxo = 2KHz

L= 0.6 mH using 8 ohm at 2KHz mid impedance.


resistor in series with mid, atten by formula
atten = 20*LOG( Xm/(R+Xm))
atten = - 6dB with R=Xm



C = 1/(2*PI*fxo*Xc)
where: Xc = speakers impedance at fxo, fxo = 1.7*2KHz

C= 6 uF using Xc of 8 ohms (Lpad)

[infinia]

example of three way Xover schematic, which can simplified in the tweeter part to first order series cap > Lpad >tweeter,
Mid as shown

[infinia]

Just noticed you are using a small mid dome (& metal yikes) for the mid. for some reason I thought you had a small cone there. my bad.
Anyway those small mid domes want to be crossed high esp with 1st order. Probably best cross at least 3 (IMO would pick 3 times with a second order) times freq of resonance. so 3*440Hz = 1.3KHz;
Conversly the small mid domes want to go higher than a 6 in cone so you can pick the crossover to the tweeter pretty much where you want.

[Autochthon]

Thanks for suggestions

Ah ... found a datasheet on the Quart 2" mid
Fs appears to be about 500 Hz and can be lowered to about 225 Hz with the back removed.

If I understand you correctly this suggests a low cross of 1500 Hz and might require a 2nd order crossover.

I'm still looking for more detailed tweeter information but I take it you think any cross over from 2500hz up to 5000hz will work?

I think I deciphered most of your sample diagram but on the woofer circuit with its 50uF cap and high value on the inductor puzzles me as it is outside the range of most frequency I played with.

[infinia]

Originally posted by Autochthon
Thanks for suggestions
Ah ... found a datasheet on the Quart 2" mid
Fs appears to be about 500 Hz and can be lowered to about 225 Hz with the back removed.

If you remove the back you need to block woofer pressure from getting in thru the back. Seperate enclosure or ?

Originally posted by Autochthon
If I understand you correctly this suggests a low cross of 1500 Hz and might require a 2nd order crossover.

Why dont you shoot for 1500 1st order and see how it sounds to you.

Originally posted by Autochthon
I'm still looking for more detailed tweeter information but I take it you think any cross over from 2500hz up to 5000hz will work?

Yes

Originally posted by Autochthon
I think I deciphered most of your sample diagram but on the woofer circuit with its 50uF cap and high value on the inductor puzzles me as it is outside the range of most frequency I played with.

They use 5mH and 50uF to cross at f = 500Hz/1.7
you want to cross here at f = 1500Hz/1.7 using fxo=1/(2*pi*(LC)^0.5)

[infinia]

Oh forgot the link where the example 3-way xover schematic comes from,
http://www.partsexpress.com/projects...crossover.html