• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Simple crossover for 3-way

toxcrusadr said:
I think that's substantially what I sketched out but I've already lost the chit from yesterday.

I think kyle was suggesting to add + and - for each driver to the schematic. In any case that mid being out of phase is quite interesting, and was probably done to flatten the response at the crossover points.
Click to expand...

Here you go! I changed it to show the logical layout better (I hope) and I added the positive and negative markings. Yes, the midrange is flipped in terms of polarity, it would appear. I hope I did this correctly, I checked it several times as I drew it. How does it look? Make sense now?


2330k_crossover by Wigwam Jones, on Flickr
 
Excellent schematic!

I'll be darned, I shoulda recognized that woofer. I had a pair of HS1-C's at one time, they were a freebie - fixer upper - pass on speaker. Nice heavy cabinets. If the drivers had been more trashed they would have made a nice cabinet for a completely new speaker, but I found a replacment for the one bad woofer and kept them going. Not bad but nothing to write home about. I don't remember what the xover looked like.

BTW the fuzz on the outside of the pot is probably cadmium plating deteriorating. Wipe it off with a wet cloth and don't breathe the dust. I've seen worse that cleaned up fine inside with DeOxit and kept on playing.
 
toxcrusadr said:
Excellent schematic!

I'll be darned, I shoulda recognized that woofer. I had a pair of HS1-C's at one time, they were a freebie - fixer upper - pass on speaker. Nice heavy cabinets. If the drivers had been more trashed they would have made a nice cabinet for a completely new speaker, but I found a replacment for the one bad woofer and kept them going. Not bad but nothing to write home about. I don't remember what the xover looked like.

BTW the fuzz on the outside of the pot is probably cadmium plating deteriorating. Wipe it off with a wet cloth and don't breathe the dust. I've seen worse that cleaned up fine inside with DeOxit and kept on playing.
Click to expand...

Thanks! Can you help me understand how this crossover works, what the approximate crossover points are? I just find this one confusing. I get the flipped polarity on the midrange to put the overlap points out of phase and therefore cancel out the overlap, right?

At this point, I'm thinking just refinish the cabs, possibly replace the tweeter with a nice cheap horn, and recap; no need to redesign the crossover (although I will probably pull the drivers and measure them with WT3 just for fun). What do you think?

Thanks again!
 
First order is the very simplest type of crossover, working only on capacitance so it has very shallow rolloff from one driver to the next.

A more expensive speaker would have an inductor in line with the woofer to keep mids and highs out, IOW forming a sharper cutoff. This speaker depends only on the inability of the woofer to reproduce those frequencies very well.

As for the midrange, assuming the drivers are all 8 ohms, the crossover frequency between woofer and mid is about 2500 Hz. I like the table on this page:

http://www.colomar.com/Shavano/crossover6db.html

Look in the 3d column of the table for the cap value for a first order with 8 ohm drivers, and scroll down to 8 uF (they show 7.9).

Similarly the mid/tweeter crossover is at about 5000 Hz.

The rolloff on either side of the crossover points is 6 db/octave by definition for a first-order, capacitance type crossover. There are always tradeoffs with speaker design. The reason designers add inductors to make 12 db (2d order), 18 or even steeper rolloff curves is to more sharply divide the ranges amongst the drivers. This speaker with its shallow rolloff curves will play more of the midrange through the woofer and tweeter than a speaker with the same drivers and a higher order crossover. This may be why they reversed the phasing on the midrange, to tame it a bit.

Crossover design is not a simple science because of all the variables, as I've learned from reading Zilch's threads. I'd leave it the way it is, unless you're really in for a can of worms and have some test gear to plot experimental curves. The horn tweeter sounds interesting though and I'd be interested to hear how that comes out if you do it.
 
Actually a first order crossover has a cap in series with the upper driver and an inductor in series with the lower one. So if yours was a true first order it would have two caps and two inductors, i.e. one set for each crossover point. I'm not sure what you call a capacitance-only crossover, maybe this is 0-order with a 3 db slope. But everything else I talked about is correct regardless (I think!).
 
Here's another reference for capacitor-only crossover calculations.

http://www.carstereo.com/help/Articles.cfm?id=1

You can calculate the xover freq. as F = 159,000/(C x R) where F is the frequency, C is the capacitance in uF, and R is the driver impedance. For your speakers this comes out 2484 and 4968 Hz. Which is essentially identical to the first-order (cap + inductor) values I posted above, thus teaching both of us (:D) that the addition of an inductor, assuming the drivers are the same, increases the SLOPE of the cutoff but does not change the crossover frequency.
 
So what you're saying is this speaker has the mid and tweet with high pass only. The bass runs full range.
 
toxcrusadr said:
Here's another reference for capacitor-only crossover calculations.

http://www.carstereo.com/help/Articles.cfm?id=1

You can calculate the xover freq. as F = 159,000/(C x R) where F is the frequency, C is the capacitance in uF, and R is the driver impedance. For your speakers this comes out 2484 and 4968 Hz. Which is essentially identical to the first-order (cap + inductor) values I posted above, thus teaching both of us (:D) that the addition of an inductor, assuming the drivers are the same, increases the SLOPE of the cutoff but does not change the crossover frequency.
Click to expand...

That is fascinating! Thanks!

I am also wondering if the two capacitors are somehow bracketing the top and bottom ranges of the midrange speaker, or is the midrange, like the tweeter, running essentially full range from the cutover frequency on up?

I hope I'm not being confusing. I mean, the woofer runs full range, I get that. But if it were a two-way speaker, we would then say that the tweeter runs from 5,000 Hz on up to infinity or wherever it naturally rolls off. Since there is a midrange in there and no inductor, it seems as though the same is true of the midrange - it runs from 2,500 Hz to infinity, only rolling off whenever it naturally runs out of steam. Does that sound right? If so, it's almost like it's two first-order, two-way speaker crossovers superimposed on top of each other.
 
No, the mid does not get the entire upper range. The 4 uf cap sees to that. Since caps are a high-pass device, higher frequencies go through the 4 uf cap to the tweeter. Above the mid/tweet crossover freq., 5000 Hz, the mid driver sees a diminishing amount of power at a rate of 3 db per octave as that material is directed into the tweeter. Similarly the tweeter plays some midrange below 5000 Hz, at a diminishing rate of 3 db per octave at frequencies below 5000 Hz.

In other words they overlap somewhat. A 5000 Hz tone would be played at equal volume through the tweeter and mid, and as you move off that frequency in either direction, the balance steadily shifts to the appropriate driver.
 
Those were made by Utah and the simple 1st Order crossover is typical for them.
Woofer runs full range....the midrange is tapped off the woofer with a single cap and the tweeter is a single cap with an L pad.
 
toxcrusadr said:
No, the mid does not get the entire upper range. The 4 uf cap sees to that. Since caps are a high-pass device, higher frequencies go through the 4 uf cap to the tweeter. Above the mid/tweet crossover freq., 5000 Hz, the mid driver sees a diminishing amount of power at a rate of 3 db per octave as that material is directed into the tweeter. Similarly the tweeter plays some midrange below 5000 Hz, at a diminishing rate of 3 db per octave at frequencies below 5000 Hz.

In other words they overlap somewhat. A 5000 Hz tone would be played at equal volume through the tweeter and mid, and as you move off that frequency in either direction, the balance steadily shifts to the appropriate driver.
Click to expand...

OK, got it. Thanks!
 
midrange could be wired out of phase to correct time alignment between the woofer and midrange. if the difference of the two cones acoustic centers (as measured from the baffle board) are half of the crossover frequencies physical length. does that make sense?
 
the two drivers must be in phase at the crossover point. otherwise the two drivers will not sum together to fill the dropoff. thats why time alignment matters. The midrange is already 90 degrees out of phase to begin with, seeing as there is a 1st order filter on it (the cap) compared to the woofer
 
Last edited:
toxcrusadr said:
No, the mid does not get the entire upper range. The 4 uf cap sees to that. Since caps are a high-pass device, higher frequencies go through the 4 uf cap to the tweeter. Above the mid/tweet crossover freq., 5000 Hz, the mid driver sees a diminishing amount of power at a rate of 3 db per octave as that material is directed into the tweeter. Similarly the tweeter plays some midrange below 5000 Hz, at a diminishing rate of 3 db per octave at frequencies below 5000 Hz.

In other words they overlap somewhat. A 5000 Hz tone would be played at equal volume through the tweeter and mid, and as you move off that frequency in either direction, the balance steadily shifts to the appropriate driver.
Click to expand...

Are you sure about this? Both the tweeter and the mid are connected to the input through their own caps so the caps are only attenuating the frequencies below their reactance. So the mid is playing from 2500Hz up and the tweeter is playing from 5000Hz up. I would think that all three drivers are running full above their xover point and rolling off mechanically. Or maybe I'm mistaken.
 
I agree with kyle above, the woofer and midrange are simply crossed over at their natural upper limits. Im not sure what toxccrusadr is talking about. A first order filter is 6db per octave. should be -3db at the crossover point* i think*
 
Last edited:
I may not have every detail correct here, I'll be the first to say that. Oh wait, I'm not the first. But anyway. I'm just giving it my best here because nobody else showed up to tell me how wrong I was. :D

I know that a first order Butterworth has a 6 db rolloff, but this isn't a Butterworth because it has no inductors, ain't that right? I think I posted somewhere above that I wasn't sure this was a 6 db crossover for that reason. I should have restated in my last post that 3 db was just an example.

How can the mid can play from 2500 all the way up if the signal is not being fed to it because it's passing through the 4 uf cap to the tweeter? Are you saying the 4 uf cap is ONLY there to keep midrange out of the tweeter, and the mid and tweeter are both playing 5000 hz and up equally?
 
but the two caps are individually wired in series to their respective drivers, they arent connected. the midrange driver im sure is naturally (mechanically) rolled off at the 5000hz freq. No inductors are necessary
 
Wiggy, let me dig about in the stash here and see if I have a couple of fresh caps for you to try in there once you get the cabs sorted out. I think I probably have some polys or worst case, fresh electrolytics you could swap in.
 
You must log in or register to reply here.
Back
Top Bottom