Custom "Heritage" Center

[jwc]

Here is the plan with this center channel. This concept has been beaten to death for the last month. My original goal was to have a twin woofer with a K401/K55 and K77. I could then get the dB the same as my Cornscalas (or Cornwall). However, I have a limited space for the center and I dont want it to have a ton of weight.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Then I messed around with the isobarik idea to get the two woofers with ½ the volume. Everyone seemed to agree this was a bad idea for this project.

Then I thought I could use two 8 inch woofers. Now this will work well for what woofers I picked out. It would have matched great but I cant find a decent 8 inch woofer that is video shielded. Yes, I could do the bucking magnets and that whole bit but a went away from the idea.

I therefore made the simple decision and went with one 12 inch woofer that is alnicothe old alnico K22. I also ditched the idea of porting it to get a low bass response. This box would have been in my TV box and I didnt want a bunch of vibration around it from low signals. The K22 will do great performance ported theoretically. I plotted this design on WinISD. I will save the ported idea for another time.

The crossover needed shouldnt be too difficult. I will have the taps like a Heresy on the T2A. However, I will have to change the midcap to a lower value to cross the K401/K55 down to 400HZ. Right now I am planning on trying a 3.0 uF. K55 off tap #2 on T2A.

Tap 2 = 134.1 ohms

C = 1/ (2 x pi x f x z)

C = 1/ 2 x 3.14159 x 400 x 134.1

C = 1/337029.7752

C = 2.9670 x 10(-6) or 2.96uF

Now, for the low pass filter. The Heresy woofer is crossed at about 458hZ from my calculations. The inductor 2.5mH. Now this same inductor was used on the K33 with the Cornwall, La Scala.blah, blah. My calculation is K33 crossing over at 216hZ. The K22 is about 8 ohms and the K33 is about 4 ohms (actually more like around 7 for K22 and 3.4 for K33). With this in mind, I may need an inductor for the K22 in this project to be different. One possibility is using 5mH to cross the K22 at about 219Hz.

I = z/ (2 x pi x f)

I = 7/ (2 x 3.14259 x 219)

I = 0.00509 or 5.09mH

I could always leave the inductor at 2.5uF. One thing that could help is if someone has a graph/plot showing the low end run off of a K401/K55.

You crossover guys are welcome to eat my lunch on this. I am NOT the expert. In the past I have stuck my neck out making conclusions such as the above so that someone will give the answer I need. Then again, Im not an idiot either. I have no background with this stuff. Just a bunch of reading and hangin out with you guys here. LOL.

[jwc]

Take a look at the below and tell me what you think.

[Deang]

!

[jwc]

Thanks Dean.

I was looking for that on Al's site the other day and couldn't find it. I am gonna stare at that off and on today. Will try and figure out what the "values" for the individual lines are.

[jwc]

Dean,

I added some numbers to your image. On the bottom "x" axis, do I have the frequencies 200,300 and 400 labeled right?

What about the dB? Not sure this graph was meant to show the exact dB at certain frequencies. However, can I assume the K401/K55 at 300hZ is about 90db? I realize that there is no network or autoformer and the K55 is wide open.

By looking at your image, I am thinking crossing over the K22 lower for this project might be better (red line on my image).

[DrWho]

Take a look at the below and tell me what you think.

It looks like you got the model for the K-33 screwed up...or perhaps you're using really shallow crossovers?

[jwc]

Tell me how it's supposed to look? I thought it is strange how the curve wasn't flattened?

Answer this. Do you think it is crossed at 219Hz? That would be the one problem. When I "cross" it higher, the curve isn't much different. I just assumed that this was the way the Cornwall cabinet affects the response of the K33. The T/S parameters were provided by Bob Crites. Those numbers are very similar to what others have posted on the forum.

You have the software. I basically put in the following info. Vented K33 with cabinet volume of 6.19 cubic feet. The port is a "square" port with the dimensions 22.25 x 3.38. The depth of the port is 9 inches. I am confident those numbers are right. However, I could have entered the data wrong.

When using WinISD, I can come up with several ways to give a "better looking" response. This however, may not be better sounding.

[formica]

Interesting concept... what's your total height? I'm a little surprised you didn't go for the 15" like your Cornscalas... but you did shave off a couple of inches.

I like your ported idea, even if you're saving it for another phase. Ported woofers have less excursion (above and at tuning) than sealed enclosures which is a good thing given most (all?) alnico K22 had very little excursion capability. I believe it was increased only after the switch to the mud magnet in an unrelated modification.

Then again, there is more than one way to skin a cat... and we'd all come out with a slightly different design...

Looking forward to reading about your progress...

Rob

[jwc]

Interesting concept... what's your total height? I'm a little surprised you didn't go for the 15" like your Cornscalas... but you did shave off a couple of inches.

I like your ported idea, even if you're saving it for another phase. Ported woofers have less excursion (above and at tuning) than sealed enclosures which is a good thing given most (all?) alnico K22 had very little excursion capability. I believe it was increased only after the switch to the mud magnet in an unrelated modification.

Then again, there is more than one way to skin a cat... and we'd all come out with a slightly different design...

Looking forward to reading about your progress...

Rob

Total height is 25 inches w/o feet or a riser. Oh yes, just building another Cornscala would be great. This would involve a bass bin of over 6 feet; can't fit it here. Also, I didn't want to chance using a non video shielded woofer since this box will be right next to the TV. I was afraid the Bucking magnet would affect the T/S parameters of the driver. The guys at PartsExpress seem to think that was possible.

The current bass bin is about 2.5 cubic feet (Heresy 1.8). To port the K22 well, you will have to get up to 4-5 cubic feet. This is a great idea but I don't have the space. I am really not to worried about low end since I will use Subs in HT.

This center won't be use for music most of the time. Typically the 2 channel stuff (95%of the time) is through the CS-dbb's.

[formica]

Oh yes, just building another Cornscala would be great. This would involve a bass bin of over 6 feet; can't fit it here.

I was figureing single bass bin versus your doubles... with an obvious ±3db loss, but bigger all the same.

The current bass bin is about 2.5 cubic feet (Heresy 1.8). To port the K22 well, you will have to get up to 4-5 cubic feet. This is a great idea but I don't have the space.

If you ever decide to experiment with it, I'd seal in the space behind the mid and tweeter horns as this would bring you very close to that objective. Enclosure would remain about the same externally...

Rob

[DrWho]

What kind of TV do you have? I think I might have seen one picture and

wasn't it a slimline flat screen thing? like LCD or Plasma? If so the

magnetic shielding won't be an issue here...I suppose you could always

hold one of your woofers near the TV to see how much it distorts the

image and how far away you need to get.

As far as bucking magnets changing the T/S parameters of the

driver...IF it causes any difference it is going to be very small. The

important part of the magnetic field is inside the gap, which is

surrounded by the pole piece and the magnet. There is no way for the

B-Field of an external magnet to work it's way essentially inside

another magnet. The only possible change would involve slightly

weakening the overall force of the magnet....but you can model the

effects of this using WinISD, and it turns out to be very small.

==========

I actually figure the internal volume of the cornwall cabinet to be

7.75 cubic feet, which with the port dimensions you mentioned results

in a turning point of 40Hz and corresponds very well to the +-3dB spec

listed on the website. You mentioned that you entered about 6 cubic

feet for the volume of the cornwall....which corresponds to a tuning

point of 45Hz, a -3dB point at 43Hz and a huge peak centered around

70Hz (so your graph is right, but your internal volume for the cornwall

isn't). I hope you didn't make your DBB's 6 cubic feet per driver

because that will yield the very non flat response you modelled (though

it gets slightly less worse with two drivers)....and perhaps that's why

you notice a lot more bass. []

An extra 2dB at 80Hz and a higher cutoff point won't usually sound that

off considering that a lot of source material needs a little boost in

the low end anyway (usually because we're not listening at volumes as

loud as they were recorded in the studio...gotta love F-M curves)...If

you want to think of it another way, the smaller cabinet increases the

sensitivity by 2dB, raises the F3 to 47Hz and results in about 1dB less

output above 100Hz...many people also find small reductions in the

100-200Hz region as an improvement in sound because it "reduces

muddiness"

If you're going to be implementing the bigger squaker from the scala,

then I would suggest going the ported route...to save on space you

could put the ports to the sides of the tweeter and above the squaker

(two 6 x 2 inch ports each 8 inches long for a port tuning of 40Hz and

an F3 of 38Hz). Even though this increases the LF capabilities of the

driver, this isn't the main purpose. The port greatly reduces the cone

excursion of the woofer and in this case, the group delay doesn't

change at all (so you're keeping the same "fast bass" of the sealed

heresy). To control woofer excursion below the tuning point, then you

might consider throwing in a high pass around 35Hz, which will make it

as undestructable as the stock sealed heresy (This won't be an issue if

you set the center to small which you should do anyway). Btw, this is

all modelled using the oldschool K-22E so you would get significantly

better results going with the newer K-24.

[jwc]

Doc, I do agree with most of what you are saying but you are off on the volume of a Cornwall. Bob Crites built his at about 6.2 cubic feet. Yes, there are ways I think the K33 could improve more. The video shielding thing I chose to do. I know there are alternatives. And yes, this center channel will be set to "small". So most of those issues you mentioned won't happen. Actually, the K22 driver will play very little in the over all sound IMO.

It is very important you read the below. I know this is concerning the K33 which is not the objective of this thread. Just to let you know, the bass from my Cornscalas and Cornscala dbb's are better than my recent late Cornwalls. This is tested with the same equipment in the same position. No real measurements....just my ears. There can't be any way the Cornwall is tuned lower than my Cornscalas. If anything, I cheated the Cornscalas to be slightly larger. So did Bob.

Cornscala external dimensions H x W x D

34.5 x 25.25 x 24.5. The height and depth are the same as the La Scala. The width is the same as the Cornwall. I kept the width the same so that the port would be identical to the Cornwall

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Cornwall 84 internal dimensions H x W x D

34.25 x 23.75 x 13.375 = 10879.7265625 cubic inches (6.296138057002315 cubic feet)

1. This doesnt account for any of the internal contents.

2. Volume K-77 = 18.31 cubic inches

3. Volume K-57 and K601 = 112.89 cubic inches

4. Volume of base of crossover, parts, excluding wires = 107.01005481 cubic inches

5. 10879.7265625 -18.31-112.89 107.01005481 = 10641.51650144 cubic inches. This is 6.158285012407408 cubic feet. This is the internal volume of the Cornwall minus the mid/high drivers and crossover.

Cornscala Internal dimensions H x W x D (bass bin only)

21.5 x 23.75 x 22.375 = 11425.23438 cubic inches

1. This doesnt account the volume of the bracing ¾ inch strips, woofer driver, dampening material, internal speaker wires, port shelf bracing and the port shelf. All of these items will essentially be the same volume as what was in the Cornwall. In other words, I didnt calculate the volume of these.

2. This cabinet will not have an internal tweeter, midrange with horn, and a crossover.

3. 11425.23438 - 10641.51650144 = 783.71787856 cubic inches. This is the volume to be made up with internal wood that can be used as reinforcement to improve the solidity of the Bass bin.

a. For example:

i. Back cover reinforcement 21.75 x 19.5 x .75 = 318.09375 cubic inches

ii. Top reinforcement 20.5 x 23 x. .75 = 353.625

iii. Bass bin bar bracing 1.5 x 1.5 x 23..875 = 53.71875

iv. 318.09375 + 353.625 + 53.71875= 725.4375 cubic inches

v. 783.71787856 671.71875 = 58.28037856 cubic inches more than original Cornwall internal dimensions or 0.033727070925925924 cubic feet. Make the subtle changes as you wish.

[DrWho]

Interesting...I'll run this by Michael Hurd. We spend a lot of time

designing stuff together through msn which is a lot easier than through

the forums. If you want we could chat too, I've got msn, aim, icq,

yahoo, and irc. I know there were three sets of T/S parameters floating

around for the K-33 so perhaps one of the others works better. Or

perhaps that 2dB difference is due to the baffle gain, which I believe

would cut off right around 120Hz or so...PWK did design by ear because

these fancy programs weren't around back in his day. One of these days

I'll have enough money for some of the bigger cooler programs, which

will include all sorts of other variables (including baffle gain and

driver orientation). On top of it I bet the cornwall specs were

measured in 1/8th space (a room corner), so there will be some room

gain as well which isn't accounted for in WinISD.

[jwc]

I 100% agree with your statement. By the way, I have done just what you said with three different K33. The older ones have a lower Fs. However, when graphed on WinISD...they are almost the same performance in the Cornwall cabinet. Just as you kinda said. This is a calculated performance and there may be other things that would account for the "good" sound bass of a Cornwall.

[firefighter]

Please pardon my intrusion but you guys are all over what I'm trying to accomplish. A friend and I split a pair of 1981 AA LaScalas to use as a center, there is no way I can use the full cabinet in the center position, the intentions are to use the HF guts in a custom cabinet with dual 8" woofers. I can understand some of the issues that are associated with the designs but most of what I read might as well be Greek. I can't begin to understand the technical aspects of your conversations but enjoy the knowledge I am exposed to. If I may, my questions are simple. Does there need to be any changes to the crossover to run 2 woofers? Which woofers are the best match? Magnetic shielding is not a problem, it will sit on top of an RPTV where I currently have a Heresy. I know there is a lot that goes into answering these question but I need it simplified such as, you can do this, or, you can't do that. Not trying to hijack the thread but you are the ones that can help me. Thanks for any input.

Jack.

[DrWho]

Hi Jack, first of all you will need to do some crossover adjustments whenever you change the impedance

of the drivers. I believe the woofer in the lascala is an 8 ohm driver

so theoretically you could wire two 16 ohm drivers in parallel or two 4

ohm drivers in series in order to maintain the same 8ohm

configuration...which would then allow you to avoid doing any crossover

mods. Otherwise you will need to do some modifications, in which case

we can catch the attention of the crossover gurus and get them to chime

in.

As always, the first two constraints when building a speaker are: how

big can it be and how much are you willing to spend? After that it's

pretty much a hunt to find the most applicable driver. Is there any

reason you want to go with a dual 8" driver configuration? What do you

find lacking with your current heresy that is prompting you to upgrade?

(we don't want to build something that's not going to improve the

situation).

A few years ago I was in the same situation where none of this made any

sense...but if you stick around and ask questions you'll find that most

everyone is willing to explain everything and provide links to other

sources. And don't worry about going off topic because that is what

keeps threads alive and interesting, not to mention there are things to

be learned from slightly different design approaches too (so you're not

even that far off topic anyway).

[jwc]

Please pardon my intrusion but you guys are all over what I'm trying to accomplish. A friend and I split a pair of 1981 AA LaScalas to use as a center, there is no way I can use the full cabinet in the center position, the intentions are to use the HF guts in a custom cabinet with dual 8" woofers. I can understand some of the issues that are associated with the designs but most of what I read might as well be Greek. I can't begin to understand the technical aspects of your conversations but enjoy the knowledge I am exposed to. If I may, my questions are simple. Does there need to be any changes to the crossover to run 2 woofers? Which woofers are the best match? Magnetic shielding is not a problem, it will sit on top of an RPTV where I currently have a Heresy. I know there is a lot that goes into answering these question but I need it simplified such as, you can do this, or, you can't do that. Not trying to hijack the thread but you are the ones that can help me. Thanks for any input.

Jack.

The way to go is with Eminence beta 8 inch woofers dual.

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=290-404

If you run these parallel, then your impedance for the low pass would be about 3.15. Your current inductor for the low pass is 2.5 mH. You could keep the inductor the same, but you will have to change the taps on the T2A and change the cap on the midrange. Now this isn't too hard and cheap.

Now those woofers listed weren't designed for low end, but in parallel in a 2.5 cubic foot ported cabinet will go as "low" or lower than the Heresy. If you were to set the HT setting for "small", it wouldn't matter. I have this plotted and it looks nice. The plot basically runs "flat" btwn 97 and 98 dB and drops at 60hZ. This is w/o a low pass filter.

Doc, that La Scala should have a 4 ohm woofer...the K33. Even if it were the pro version the K43...I still think it is 4 ohm.

[jwc]

I don't mind a hijack either, but:

GUYS, GUYS, GUYS....I NEED INPUT FOR THE ORIGINAL QUESION AT HAND.

Here is a plot of the K22 in my current cabinet. 4 options here. Ported vs sealed and crossing at 458 vs 216Hz. Which would you use?

Doc, I used to use mIRC a lot in the past for chating online. Is this what you are referring to? If so, how do I find you?

[jwc]

Doc, this is an FYI on 3 different K33's. I would love for someone to figure out why these curves look the way they do.

[DrWho]

I'd go with the 400Hz crossover...I think it'd sound more cohesive to have the woofer playing for more than a single octave

Just curious, is there any specific reason why you sealed off the LF

cabinet from the upper cabinet? Seems like a good 1.5 cubic feet there

you could use...

still dunno what is up with that K33 plot...I've always modelled using

8 cubic foot cabinets which go relativtely flat beyond 40Hz. I wonder

if there isn't something totally wrong with the T/S parameters we

have...I know D-Man said they don't match up with the mouth and

expansion rate of the khorn either.