3 quick and simple crossover questions regarding the Altec 902...

[chops]

I dropped DeanG a PM about this as well, but figured it wouldn't hurt to post it here in the forum either, so here it goes...

I might be getting my hands on a pair of Altec 902-8B's soon. For a short time, I will be running them 3-way in my current setup, and later on down the road possibly switching over to 2-way. Now here's the questions...

Since I am running a 7uF cap bank on the squawker with the K-55 which is 16 ohms, I believe I was told it is closer to 350Hz than 400Hz.

1) NOW, if I were to drop one cap and run just the 3uF cap on the squawker, where would that put the crossover point for the K-55, around 700-800Hz?

2) If it is around 700-800Hz at 16 ohms, wouldn't putting in the 8 ohm 902 basically drop that crossover point back down between 350-400Hz, again at 8 ohms?

3) If that is correct, and I can run the 902s just like that, how could I, or would I have to attenuate the 902 by switching a tap?

So what do you all think? Is this doable?

Thanks in advance!

[djk]

Moving to an 8 ohm driver from 16 ohms and not changing the cap will double the frequency. It will also be too loud. Moving down one tap on the autoformer will match the output levels and give the same crossover point as before.

[chops]

From what I remember in the directions from some of the pre-built 3-way crossovers I have bought from Rat Shack and Parts Express years ago is that when you drop in impedence (16 --> 8 ohms), your crossover freq also drops by half (400Hz --> 200Hz), not the other way around (400Hz --> 800Hz). Although, I could be wrong, but I'm pretty sure that's what the directions said.

Of course, dropping from 16 to 8 ohms, the driver is going to receiver more power which in turn will make it louder.

Also, which tap would I move to? Right now, the squawker is on tap #5, the tweeter is on #4. Taps #1 and 2 are open.

[Deang]

It is as Dennis said.

Since I am running a 7uF cap bank on the squawker with the K-55 which is 16 ohms, I believe I was told it is closer to 350Hz than 400Hz.

The K-55 is actually more like 13 ohms. Once on the horn, it climbs up a bit and hovers around 15 ohms. With each step down on the autoformer, there is a doubling of impedance. Off of tap 3, the squawker has a reflected impedance back to the amp of 60 ohms. So, your primary capacitor value (between input and tap 5) is sized based on 60 ohms. However, between the autoformer output taps and the driver, the parts are sized based on the true impedance of the driver/horn combination (8 ohms for the tweeter, 15 ohms for the squawker). With 7uF as your high pass and 4uF as your tweeter cap -- you're sitting at 375/4900. The .50mH inductor you are using for the low pass starts rolling off the squawker at 4700.

1) NOW, if I were to drop one cap and run just the 3uF cap on the squawker, where would that put the crossover point for the K-55, around 700-800Hz?

Closer to 900Hz. It would be better to drop the 3 and use the 4 (about 650Hz).

2) If it is around 700-800Hz at 16 ohms, wouldn't putting in the 8 ohm 902 basically drop that crossover point back down between 350-400Hz, again at 8 ohms

When using the K-55, it is getting half the power that the 8 ohm tweeter above it is getting (when on the tap below it). So, like Dennis said, when you drop the 902's in, you're going to have a sensitivity issue.

3) If that is correct, and I can run the 902s just like that, how could I, or would I have to attenuate the 902 by switching a tap?

Dennis gives the fix. Dropping to tap 2 provides the proper level of attenuation, and since that tap once again doubles impedance, you can still use the 7uF value at the primary.

However, you will no longer be able to use the .50mH inductor for the squawker low pass. That value is based on 16 ohms, and you will be at 8. So, you need to change that to .25mH.

Personally, I think you should change the whole thing!

[Deang]

No, it's the other way around. If you double impedance, the crossover point shifts down.

Tap 5 is your input tap for the primary high pass. Your squawker is actually off of tap 3 (output tap) -- you will be moving it to tap 2.

Why not just buy the 16 ohm version of that driver?

[chops]

Hi Dean,

To start things off, I do plan on changing the whole thing out once I feel confident enough to build my own 2-way network. If possible, I'd like it to be a 3rd order design for both the K-33E and 902-8B.

So just for now, I'll be running the 902 on my current networks.

Please let me know if I have this straight.

1) Move from tap #5 to #2.

2) Leave 7uF cap bank as is.

3) Insert 902-8B in place of K-55V

4) Scrap .50mH inductor for .25mH inductor

5) Fire her up and let her rip.

If this is correct, I should still have a crossover point of roughly 375Hz/4900Hz for the bandpass?

[chops]

No, it's the other way around. If you double impedance, the crossover point shifts down.

Tap 5 is your input tap for the primary high pass. Your squawker is actually off of tap 3 (output tap) -- you will be moving it to tap 2.

Why not just buy the 16 ohm version of that driver?

Ok, got it!

And as for the taps, I got that too. Thanks. []

And the reason for going with the 8 ohm version is because I have 2 seperate opportunities to get a really nice pair of the 8Bs at a good price, and really can't afford the $450 that GPA wants for them.

It's kind of funny, because before I kept saying I wanted to buy the 16 ohm versions and everyone kept asking me "why not buy the 8 ohm ones"... Kinda damned if I do, damned if I don't sorta deal. []

[Guest " "]

It is as Dennis said.

Since I am running a 7uF cap bank on the squawker with the K-55 which is 16 ohms, I believe I was told it is closer to 350Hz than 400Hz.

The K-55 is actually more like 13 ohms. Once on the horn, it climbs up a bit and hovers around 15 ohms. With each step down on the autoformer, there is a doubling of impedance. Off of tap 3, the squawker has a reflected impedance back to the amp of 60 ohms. So, your primary capacitor value (between input and tap 5) is sized based on 60 ohms. However, between the autoformer output taps and the driver, the parts are sized based on the true impedance of the driver/horn combination (8 ohms for the tweeter, 15 ohms for the squawker). With 7uF as your high pass and 4uF as your tweeter cap -- you're sitting at 375/4900. The .50mH inductor you are using for the low pass starts rolling off the squawker at 4700.

1) NOW, if I were to drop one cap and run just the 3uF cap on the squawker, where would that put the crossover point for the K-55, around 700-800Hz?

Closer to 900Hz. It would be better to drop the 3 and use the 4 (about 650Hz).

2) If it is around 700-800Hz at 16 ohms, wouldn't putting in the 8 ohm 902 basically drop that crossover point back down between 350-400Hz, again at 8 ohms

When using the K-55, it is getting half the power that the 8 ohm tweeter above it is getting (when on the tap below it). So, like Dennis said, when you drop the 902's in, you're going to have a sensitivity issue.

3) If that is correct, and I can run the 902s just like that, how could I, or would I have to attenuate the 902 by switching a tap?

Dennis gives the fix. Dropping to tap 2 provides the proper level of attenuation, and since that tap once again doubles impedance, you can still use the 7uF value at the primary.

However, you will no longer be able to use the .50mH inductor for the squakwer low pass. That value is based on 16 ohms, and you will be at 8. So, you need to change that to .25mH.

Personally, I think you should change the whole thing!

Emphasis on the last line. Build from scratch or pay some one to do so. If you build from scratch, use the link to the cross over calculator I provided in a previous post to you. All you need to do is plug in your driver values, and the calculator will genrate a schematic with correct values. Be careful about using your existing xover network with mods...problem points are all the shunt devices that go to ground....if you cross over outside the operating specs of these parts...too much signal may be shunted to ground which could cause a short at lower frequencies. On the other hand...if the values of your inline parts are not correct...you could be causing holes of coverage due to gaps at the cross over points.

My recomendation is to leave your current xover as is so you have a fall back plan and some thing to compare to.

A quick observation for you so you have a model to go by. Since you basiclly replaced your 16 ohm atlas drivers (k-55's) with 8 ohm drivers, Your new model is a klipsch xover network deisgned for the K-55-M as a starting point. The k-55-m drivers were used with the AL line of klipsch crossovers in the LaScala. AL-2, AL-3, etc. Here is the schmatic for AL-3 which is favored by most over the AL-2.

[chops]

Also, would this .24mH 14AWG be sufficient enough? It has the lowest DCR of the bunch at .009 ohms.

http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=266-310

[Deang]

The K-55-M is also a 16 ohm driver. Also, he's using a modified Cornwall.

That part will do nicely.
1) Leave 7uF on tap 5
2) Remove .50mH from tap 3 to squawker +
3) Replace with .25mH, and place on tap 2 to squawker +
4) Trade out your drivers and then let her rip:)

I can't remember if you originally had Bob's A/AA's or not. If so, you can use the small inductor that used to be in the AA tweeter section for the new midrange connection.

If you are going to buy some parts -- then I would take advantage of the extended response of the new driver. Bob's mod is nice if you are using the old horns with his tweeter, but my opinion is that only the old horns in combination with the K-55 really benefit from it. You now have better horns, and are moving to drivers with some serious extension. If I were sitting where you are, I would:

1) Change 7uF to 4uF at tap 5
2) Remove .50mH from tap 3 to squawker + and replace with .13mH
3) Replace 4uF from tap 4 to tweeter + and replace with 2uF

You will do it your way first. Curiosity will finally get the best of you, and then you will try it my way. My way always sounds better.

[chops]

Sorry for the late reply guys. I was a little busy on eBay... [:$] To make a long story short, I now have a need to put these questions I'm asking into action in the next few days! YES, I got the 902-8B drivers! [<)]

Anyway Dean, I think if I try these 3-way, I'll just go ahead and go your way first off. Remember, my 4uF to the tweeter is (2) 2uF caps in parallel, as is the 7uF for the squawker being a 4uF + 3uF in parallel. So it would be very simple for me to just drop on 4uF and 2uF cap per driver to get the desired changes. Then all I have to buy are (2) .13mH inductors.

Speaking of the inductors, I guess I would go with the Jantzen 18AWG .13mH, but it has a DCR of .16 ohm. Would I be better off going with the Erse 14AWG .15mH with a DCR of .07 ohm?

Or better yet, couldn't I just leave the inductor out altogether and let it run all the way out to 20kHz?

I was thinking, I remember reading in another thread that you cannot just take the tweeter out of the network without it automatically shifting all the other values around and messing up the crossover points and such.

If that's the case, then what would happen if I completely disconnected the caps going to the tweeter? Wouldn't the network then essentially be a true 2-way? Wouldn't the crossover points for the K-33 and 902 stay the same?

"You will do it your way first. Curiosity will finally get the best of

you, and then you will try it my way. My way always sounds better.:)" <<-- LOL! I'm sure it probably is! []

[Deang]

My first choice would be part # L16.13 from www.solen.ca/ , my second choice would be the 14AWG Erse, and then use a meter to peel down to the right value. My third choice would be the Jantzen.

Like you said, another option is just to go to a two-way now and run the new driver full out. You have the right idea -- just disconnnect the tweeter section (remove connection from tap 4). It'll be interesting to see what you think.

[chops]

My first choice would be part # L16.13 from www.solen.ca/ , my second choice would be the 14AWG Erse, and then use a meter to peel down to the right value. My third choice would be the Jantzen.

Like you said, another option is just to go to a two-way now and run the new driver full out. You have the right idea -- just disconnnect the tweeter section (remove connection from tap 4). It'll be interesting to see what you think.

Dean,

I noticed that the Erse 18AWG .13mH has a DRC of .15 ohms. It's the same as the Solen, but the Solen is 16AWG. Should I even be concerned with the gauge difference? Is there a considerable "sound" difference or improvement of the Solen?

The only reason I ask is because I already have an account with Parts Express and all I would have to do is hit the "GO" button. []

But for now, it looks like I'll immediately be going the 2-way route, disconnecting the entire tweeter section from the network.

So completely remove (disconnect) tweeter section, leave 7uF cap bank at tap #5, remove inductor completely, and run squawker (+) to tap #2, connect 902 drivers and go, right?

Boy, that sounds simple as PIE! [H]

[IB Slammin]

Chops,

At this point, the cat is out of the bag, so speak to Dr. Cullison about the "Alcorn" crossovers. They were designed for my CW bins, 511b/902 mids, and the K77 or CT125 tweet. Very sweet mids & top, and very tight LF.

tc

[chops]

Chops,

At this point, the cat is out of the bag, so speak to Dr. Cullison about the "Alcorn" crossovers. They were designed for my CW bins, 511b/902 mids, and the K77 or CT125 tweet. Very sweet mids & top, and very tight LF.

tc

Hey Terry!

I was wondering if you were ever going to add anything to this post. We 511B/902 guys need to stick together. LOL

Ah, so jc is the one behind the Alcorn networks you were "beating around the bush" about, huh? Interesting. Yeah, I guess I'll have to drop him a line and see what he has to say. Every little bit of info helps.

Thanks!

Charles

[Deang]

JC's already admitted he can't hear worth a crap, wait -- maybe what he said was he likes things that sound like crap -- and Terry can't hear anything until it gets above 110dB.

Stay the course Charles. Work with the simple filters/mods and get your ears dialed in to what they do to the sound. If you move too fast, you'll have to go back and do it all over again! I'd like to see you try some different caps too -- before you get brainwashed into believing they all sound the same.

[chops]

JC's already admitted he can't hear worth a crap, wait -- maybe what he said was he likes things that sound like crap -- and Terry can't hear anything until it gets above 110dB.

Stay the course Charles. Work with the simple filters/mods and get your ears dialed in to what they do to the sound. If you move too fast, you'll have to go back and do it all over again! I'd like to see you try some different caps too -- before you get brainwashed into believing they all sound the same.

Don't worry Dean. I was referring more towards later courses. I plan on doing what we spoke about. And I do want to try some different caps, it's just hard tyring to figure which ones are the best for my application, (5% or 10%, Solen, Kimber, Hovland, Mundorf, etc, etc). There are too many choices and I'm obviously not up to full speed yet on what's what and why.

If I ever completely figure out the whole idea of building crossovers, I would like to evenually have a 2-way (most likely - hopefully) network with a -48dB per octave slope on the Altec 902 at least, if not the K33 also.

No brainwashing going on here! [8-|]

[Duke Spinner]

, I would like to evenually have a 2-way (most likely - hopefully) network with a -48dB per octave slope on the Altec 902 at least, if not the K33 also.

Chopsy .....

I think you will find the high roll -off x-overs work best on speakers with multiple overlapping drivers

although , i admit, say a 12/dB/oct on a k-33, and say, a 2nd order filter on a 902/908 set lower than one would normally choose..... might improve power handling

at what levels do you commonly listen ..??

i'd say, tho ...

i suspect you won't need much power handling in the 902,with say, an 600-800 hz x-overed K-33 therefore, no need for higher order x-over

and ....

a 12/dB/oct ... may sound more open, and musical in a 2-way

a lot will depend on the horn you choose, eh ....???

[jwc]

JC's already admitted he can't hear worth a crap, wait -- maybe what he said was he likes things that sound like crap -- and Terry can't hear anything until it gets above 110dB.

Stay the course Charles. Work with the simple filters/mods and get your ears dialed in to what they do to the sound. If you move too fast, you'll have to go back and do it all over again! I'd like to see you try some different caps too -- before you get brainwashed into believing they all sound the same.

My hearing is in question but I love stuff that sounds like crap. Infact, I am partial to the smell of it.

Chops, I am no expert. I have piddled a bit with this CW K33 bass bins. The 511/902 combo is very good.

I would stay the course of what was mentioned above. The one strong point I will add is to consider getting away from crossing the tweeter at 4500Hz. The 902 does great out there and there is no good reason I can think of to cut the 902 off at 4500Hz.

The second strong point is to not rush into a new change quickly. Dean is right on this one (he has said it to me before too). I used to get too excited about a particular mod or change and then make a decision. It is quick and easy to mod/upgrade but its evaluation takes awhile.

You are on the right track. The 902/511 combo is a winner in my book.

jc

[chops]

, I would like to evenually have a 2-way (most likely - hopefully) network with a -48dB per octave slope on the Altec 902 at least, if not the K33 also.

Chopsy .....

I think you will find the high roll -off x-overs work best on speakers with multiple overlapping drivers

although , i admit, say a 12/dB/oct on a k-33, and say, a 2nd order filter on a 902/908 set lower than one would normally choose..... might improve power handling

at what levels do you commonly listen ..??

i'd say, tho ...

i suspect you won't need much power handling in the 902,with say, an 600-800 hz x-overed K-33 therefore, no need for higher order x-over

and ....

a 12/dB/oct ... may sound more open, and musical in a 2-way

a lot will depend on the horn you choose, eh ....???

Dukey....

I'm using the Altec 511B horn, and my current crossover point is 375Hz on the K-55V driver. I do not want to run the K-33 any higher than where it's at now, which is about 437Hz according to BEC.

If I could somehow get the 902 driver right at 400Hz with something more substantail than a 1st, 2nd or 3rd order network, then I think I'll be fine. That's why I would like something like -36 or -48dB per octave. And if I could integrate constant impedence into it, all the better.

Granted, I don't play my system as loud as Terry does (IB Slammin) at 110-115dB, but I'm not too far behind him either at about a constant 100dB with peaks anywhere between 103-107dB.

Chops

Dean,

Am I right in assuming that adding a Mills 10 ohm / 12 watt resister across taps #0 & 5 will pretty much make the network a constant impedence design like you and Al keep mentioning about?

http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=005-10

Thanks