Cornwall/Cornscala network round up

[jwc]

I have demoed Cornwall like designs over the last month. I have run 4 through my Cornscalas (not dbbs). The nice thing about this speaker is that the Cornwall crossovers work on them and there is no back cover to screw on and off. Yes this is a K400 and there is a difference/improvement over the Cornwall but works.

In the photo attached you will see 4 networks on top of my Cornscala. One is Deans ALK B which is a beauty (far right). He made this for me with the intention of putting in my Cornscala and you can SEE the network (as opposed to concealed). The B-3 is to the left and the B is between the ALK and B-3. There is a custom constant impedance design with a bandpass at the most front of the picture. Not made to be pretty. Not gonna go into details on that one.

Some conclusions. The ALK can handle the most volume w/o killing you. The tweeter is the best sounding on the ALK. More and more I am realizing that the K77 requires a more complex network to work "well". The imaging on the ALK is very good as well. Note: this ALK B doesn't have any yellow caps. Does have some Solen FF caps though. The design/build is what is coming through here. When I think of this network, I think of good tweeter and overall imaging.

The Type B had a very good bass to my liking. I think this is partly a room issue. It also may be more of what I like out of music. The bass isn't "snappy" but slightly "healthy" boomy. Hard to describe. Some hate this. I listen to a lot of funk music and this may be the reason. This network opens up very quickly. The tweeter sounds on the B fall behind of the group. I think this is a function of the tweeter personally. It starts to scream. Tried out Solen met caps, thetas and Bob's cans. The thetas are superior (of the three) for the first order tweeter filter. Bob's new Tweeter may be a good solution for the fix for the first order design. The mid cap to me doesn't seem to make as much of a difference with the B. I left the cans there, they sound fine.

The B-3 is well balanced and the bass is more "snappy" and quick. Most probably may like this better. I thought that the bass of the B-3 was similar to the ALK but slightly better. I posted this before in a thread. This is a good network. It will tolerate much higher volume than the B w/o getting harsh. If I had a pair of Corns again and had to choose between B or B-3, I would go with the B-3 if running off a high powered SS amp.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

All good networks.

jc

[BS Button]

To me the bass response is much more pleasing with the B-3 and that's without any loudness button or tone controls.

BS

[jwc]

I would think that most would like the B-3 bass better than the B. However, I think my room is a factor...just a guess. My Cornscalas are in my HT setup...smaller room and I play a bunch of funk, dance, metal and stuff. I also don't play them very loud. The B's open up quicker, more forward. I don't listen to the above mentioned music very loud that often. If I were to play some classic rock, jazz or easy listening...the bass would be more appropriate with the B-3.

I know this is ridiculous logic....

Todd, I play my dbb's with the "higher" quality recordings at a louder volume.

The B's in the Corns will have a different sound that is very classic in my opinion of the PWK sound. Probably a less flat response, opens quickly, and forward. The B-3's are IMOO more balanced and can jam! The B-3 is good.

Note that I haven't settled on a network for the Cornscalas. I am gonna listen to several over the months to come for a good trial. The B's I have in there are crossed at 400Hz right now. Later, I'm gonna put the ALK B's back in there...for awhile. I may also change the horns...just piddlin.

[BS Button]

I'd like to also toss a note to the Hepner driver K52H or the first K52K's.

[Deang]

I know this is ridiculous logic....

Must be why agreed with every single thing in the post -- nice job JC.

[BS Button]

I know this is ridiculous logic....

Must be why agreed with every single thing in the post -- nice job JC.

Dean,

If you'll recall, you and DJK jumped on me right good about changing from the B-3 back to a B. You also encouraged me to stick with the K52's. I listened not, so a year later I'm back. Should've taken your advice back then, but I was already too far along.

BS

[BiggerIsBetter]

Did you change the mid cap value on your B's to account for the 400 horn? If so, what value did you use?

I went with 7 for the mid and 2 on top as recommended. This resulted in a boost in the crossover region (5-7K), which further exacerbated the already bright highs. From simple math, it seems that the change to the mid cap has lowered to xover point to the highs, which seems to support what I measure and hear.

I wonder if a 1.5 or even 1 for the high cap is the way to go for the cornscala. I agree that the top is complex in that it is not very flat (for very long at least) so you either deal with a hump around the crossover region or a very early roll off.

Great comparison. It is just about time to build some ALKs....

[jwc]

Good point.

First of all, I just left the mid cap as 4uF just as in the Cornwall. The comments posted above reflect that cap.

Currently....yes. I have changed the mid cap to raise the mid crossover frequency....right now at 6uF. This does make a slight diffence in sound...can't really comment on that point as I don't remember A/B this cap change.

You could lower the tweeter cap. If your into experimenting, I would move more toward bandpassing before changing the crossover frequency of the tweeter. I think to get away from what you are describing, you will need to go to second order (B-3) or ALK or come up with a custom deal like I was doing.

Now you notice I didn't comment much on the mids specifically on the ALK B. This network has great imaging and great tweeter. I'm not 100% sure the squawker is balanced with the woofer and tweeter for my setup. I really didn't want to elaborate on this because the K401 vs the K600 may be a different bird and the ALK is a "complex" design that is specific for driver/horns. They sound great but at the volume I listen to, they don't shine as much. They definately will tolerate the most volume put to em. They become more balance for me as the volume goes up. Again the HT room with the Cornscalas is small.

[BiggerIsBetter]

Yep, I definitely want to go to a bandpass on the mid at some point soon. In the immediate term I am concerned that the change of the mid cap from 4-7 has significantly lower the high crossover point. Although toasting the tweeter would give me a good excuse to ditch those dogs, it has made the 4-10K boost a lot worse (up 5-10 db through that region) and it sounds it with the "wrong" music.

[Deang]

I am concerned that the change of the mid cap from 4-7 has significantly lower the high crossover point.

How? You moved from 600Hz to 400Hz with that change -- the tweeter section is the same. There is the reactive component, and the top might shift a bit -- but I don't see how you could possibly get a 5-10dB increase in output where you're saying it is.

[BiggerIsBetter]

Much of the 5-10 db bump of the top was there already. I believe that it was just made worse by changing the mid cap, which added more enery around 4-6K . What ever ratio of the mid cap you consider (i.e. the AT), how can a 7+2 not be a lower crossover point to the tweeter than a 4+2?

[jwc]

When I changed the mid cap from 4 to 6uF, I didn't notice what you are describing. Not sure how that would happen. The network is already top end heavy and comes more apparent at higher volumes. Great for lower volume listening.

I know the big thrill of the Cornscala idea was to bring that Crossover point to 400Hz but I think the better improvement just came from the use of the K401 vs k600.

[BiggerIsBetter]

Ignoring the ratios of the AT for a moment, which is not correct I know, and just assuming the caps are in series at their full values, my math has the crossover point to the tweeter being lowered by almost 1000hz with the change to the mid cap. I know that "cross" over is a loss term when we are talking 1st orders with these drivers, but I did hear and measure (at the listening position) a difference that is consistent with this math.

A 1.5uF for the tweeter should maintain about the same crossover point as with the stock B. Does anyone elses math provide a different answer?

[Deang]

"There is a custom constant impedance design with a bandpass at the most front of the picture. Not made to be pretty. Not gonna go into details on that one."

Unfortunate.

"Some conclusions. The ALK can handle the most volume w/o killing you. The tweeter is the best sounding on the ALK. More and more I am realizing that the K77 requires a more complex network to work "well". The imaging on the ALK is very good as well. Note: this ALK B doesn't have any yellow caps. Does have some Solen FF caps though. The design/build is what is coming through here. When I think of this network, I think of good tweeter and overall imaging."

The ALKs defintely shine at the higher sound pressure levels. However, at lower volumes there is a loss of dynamics and the soundfield collapses. In the Universal Type A, where the squawker level can be adjusted -- I invariably always end up back on taps 4 and 0. Too hot by the numbers, but it's the only way to bring the snap and sense of immediacy back into the midrange at the volume levels I usually listen at. I think those with smallish rooms or who sit right on top of their speakers will struggle getting balance with the ALKs. I just don't think it's a very good filter for low volume/low power listening.

"The Type B had a very good bass to my liking. I think this is partly a room issue. It also may be more of what I like out of music. The bass isn't "snappy" but slightly "healthy" boomy. Hard to describe. Some hate this."

I'm one that doesn't care for any 'bloat' or 'boom' in the bottom. I did own some very nice Cornwalls for a short period of time -- and I thought the bass was fabulous. I think maybe in your case it very well may be room related. The Cornwall is a speaker that I personally think sounds better a bit off the walls.

"This network opens up very quickly."

The Type A has similiar behavior. It's no surprise that these networks are preferred by SET/low power users. Something to think about when you hear a SET user say, "Man, 1/2 watt and it's driving me out of the room!" lot of power is dumped into the tweeter right quick with these networks:

"A tweeter with a third order high pass filter with a crossover frequency of 5000 Hz driven by the 100 watt amplifier used in the power chart above, will receive about 1.6 watts at 2500 Hz versus 25 watts with a first order filter at full output."

"The tweeter sounds on the B fall behind of the group. I think this is a function of the tweeter personally. It starts to scream..."

It's a decent tweeter for where it's being used at -- 6kHz and up. The problem isn't the K-77 so much as it is the effect a 1st order filter has on it when the volume is dialed up. I use a very simple rule here: the louder you play the higher order filter you need. Also keep in the mind that the lack of true bandpass means the upper end of the squawker's response is stomping all over the tweeter. Now, by your own admission the ALK sounds very good up loud -- but you're using the same tweeter!

"...Bob's new Tweeter may be a good solution for the fix for the first order design. The mid cap to me doesn't seem to make as much of a difference with the B. I left the cans there, they sound fine."

My personal opinion is that regardless of the quality of tweeter, a first order tweeter section is best reserved for those who enjoy modest volume levels.

"The B-3 is well balanced and the bass is more "snappy" and quick. Most probably may like this better. I thought that the bass of the B-3 was similar to the ALK but slightly better. I posted this before in a thread. This is a good network. It will tolerate much higher volume than the B w/o getting harsh. If I had a pair of Corns again and had to choose between B or B-3, I would go with the B-3 if running off a high powered SS amp."

JC, did I build your Cornwall ALKs before or after Al redesigned the low pass section. IOWs, did I build yours using the new design or the original?

The B3, like the AL/AK-3 has some additional attenuation on the squawker -- this probably has a much to do with changing the character of the bass as the 2nd order low pass section.

[Deang]

"If you'll recall, you and DJK jumped on me right good about changing from the B-3 back to a B. You also encouraged me to stick with the K52's. I listened not, so a year later I'm back. Should've taken your advice back then, but I was already too far along."

Todd -- my concern was that we didn't have any evidence that the B3 was ever used with the K-55-V, and it also appeared it was designed to work with the drivers being used during the transition to the Cornwall II. Since Klipsch refuses to weigh in on the matter, I guess we'll never know.

I can't remember what Dennis' reason was -- I'm sure it was better than mine.

[Deang]

"...and just assuming the caps are in series at their full values, my math has the crossover point to the tweeter being lowered by almost 1000hz with the change to the mid cap. I know that "cross" over is a loss term when we are talking 1st orders with these drivers, but I did hear and measure (at the listening position) a difference that is consistent with this math."

On tap 3 (60 ohms), 4uF puts you in the vicintiy of 600Hz, and 6.5uF around 400Hz. There is nothing here as far as know that would have any impact on the tweeter -- where the transition behavior is being dictated by the 2uF. I'm not a math wizard by any stretch, but I am curious what math you are using. Now, since there is nothing electrically speaking to deal with the squawker's upper end, what you might be hearing may actually be coming from the squawker. Get an inductor in there for the squawker -- it's a big improvement.

[jwc]

Dean. My God. Great post.

Your built ALK B's for me is of the new design.

I think I posted that constant impedance design somewhere for the Cornscala. Let me look for it.

jc

[jwc]

Agreed. Just leave that XO the way it is and add a 1.5mHy inductor in series somewhere after the 7uF cap. Don't put it in front of the tweeter filter. Do on one channel and compare! Would love to hear your results!!!

Try

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=255-426

[Deang]

1.6mH is based on the 60 ohm impedance on the input side of the autoformer. If you move it downstream to the ouput side of the autoformer -- you can size that inductor based on 15 ohms. IOWs, .40mH between tap 3 and the squakwer. If you do it this way, you'll have a 'BECwork'. I would try it both ways. The higher resistance of the larger coil does sound more attenuated, and some might prefer it that way.

Found your CornScala schematic: In your tweeter section you either need to change the .15mH coil to .20mH, or change the 7uF to 4uF. Currently, you have .15mH value from the Super AA coupled with the 7uF value from the Universal Type A.

[jwc]

Hey Dean,

I noticed that difference between those two designs. Here is how I came up with that tweeter filter.

When I assume an 8 ohm tweeter crossed at 6000Hz...I get 3.3uF for first order. When I go third order, I get the 2.2uF, .15mHy, and 6.6uF.

Why is that? That is currently the way I have it on my dbb's as well.

I also have a .5mHy I bought for the purpose of trying out a bandpass on a Type A the way you described it above....As Bob did too. I guess if you put the inductor "after" the autoformer, it will always be calculated as about 15 ohms regardless of the tap. Sound right?

jc