First attempt

[daveh]

OK, so I have built numerous designs by other folks, and am trying to do a little something on my own.

The wife needs new computer speakers and asked me to build some for her. The tough part is the small space that is available for them. I can basically fit about a .2 cu. ft. speaker on each side of her LCD monitor.

So I started searching for drivers that might work in that small an enclosure. I like a few different woofers for this application but believe I am going to go with one of the TBs. My first thought is the W4-1052SD. It looks pretty good modeled in WinISD. Not spectacular on the low end but should be fine for the wife.

So, I picked another TB tweeter to go with woofer. I looked at Zaph's site and figured with the 24-1414SE on sale right now I would give it a shot.

Here is what I came up with. Modeled the whole thing in PCD and Speaker Workshop and it looked pretty good.

I would like some input from those of you who know a lot more about this than I do.

Your help is greatly appreciated.

Dave

[philiparcario]

looks like the f3 is too high. is it 200hz? or 150hz

[daveh]

When I model it in WinISD with a 1.375x6.5 in. port I get an F3 somewhere around 66 Hz. Not sure why it does not come out the same when I model in PCD (most likely I am not using the tool correctly.)

I used Unibox, FRC, then PCD so I could have goofed up somewhere in the process.

Thanks,

Dave

[philiparcario]

i get good bass with it ported. the graph you posted must be wrong. phil

[Wolf]

OK, so I have built numerous designs by other folks, and am trying to do a little something on my own.

The wife needs new computer speakers and asked me to build some for her. The tough part is the small space that is available for them. I can basically fit about a .2 cu. ft. speaker on each side of her LCD monitor.

So I started searching for drivers that might work in that small an enclosure. I like a few different woofers for this application but believe I am going to go with one of the TBs. My first thought is the W4-1052SD. It looks pretty good modeled in WinISD. Not spectacular on the low end but should be fine for the wife.

So, I picked another TB tweeter to go with woofer. I looked at Zaph's site and figured with the 24-1414SE on sale right now I would give it a shot.

Here is what I came up with. Modeled the whole thing in PCD and Speaker Workshop and it looked pretty good.

I would like some input from those of you who know a lot more about this than I do.

Your help is greatly appreciated.

Dave

I have used the -1414, and it's splendid, once you filter the top-end. It requires a 0.44uF cap, and a 0.1mH coil in parallel placed in series with the driver. You can see that in the circuit where I used it.



The rest should be a little different, as you won't need that complex of a filter on the -1052, or on the tweeter, IMO. I had a oal to meet, and met it with the circuit shown.

It's best to use it above 3kHz if possible, and the -1052 will get that far with aplomb. LouC did a project called "Amethyst" with the W4-1052 and the Dayton Neo tweeter. His profile should refer you to his website.

If you want, I could take a crack at the simulation sometime tomorrow, as I have the files on-hand. I just need box params, physical offsets, and intended xover point.

I know your rolloff is not good looking in your simulation. I think it has to do with how you did it. If you do not have the PCD 6.12 or higher, or the Response Modeler program, you should have both of these. (Jeff B's profile should get you those.) I've not used the EDGE, FRC, or the diffraction simulator, other than the RM 2.0. The RM has all of this in one program. Since your setup is for the PC, you would most likely have it against a wall, correct? I would not go more than 2dB if that for the BSC. The RM 2.0 also splices in the box alignment you modeled from Unibox. Just copy the data into the RM from Unibox, and splice the curve.

RM:
Import .zma's, and .frds
emulate box model
splice bass-curve
model baffle size and edge type with driver directionalities and X/Y offsets
Invert and remove some applied BSC if you want less than 6dB
extract minumum phase
save as your modified .frd file

Then just import into the PCD with the new files, and enter the X/Y/Z offsets in relative to the tweeter position (0/0/0).

Model to your heart's content! I hope this helps,
Wolf

[gmilitano]

daveh,

What software package is that?

[Wolf]

daveh,

What software package is that?

It's the Passive Crossover Designer, by Jeff Bagby. Look up Jeff B, and he has a link to his software on the internet. It's freeware, and a very powerful program. You'll need to have teh Response Modeler 2.0 as well, as it melds very well with the PCD program.
Regards,
Wolf

[daveh]

Wolf,

Thanks for your help and the links to LouC's site. I had looked at his site quite some time ago and now remember where I got the idea to use the W4-1052 in the first place. Thanks Lou!

Using your instructions for RM and PCD I think I have come up with something a little more respectable. The front baffle size is 6.5" x 9.5".

Anyone who would like to give me some criticism, constructive or not please feel free.

Thanks!
Dave

[Wolf]

Wolf,

Thanks for your help and the links to LouC's site. I had looked at his site quite some time ago and now remember where I got the idea to use the W4-1052 in the first place. Thanks Lou!

Using your instructions for RM and PCD I think I have come up with something a little more respectable. The front baffle size is 6.5" x 9.5".

Anyone who would like to give me some criticism, constructive or not please feel free.

Thanks!
Dave

That looks a LOT better, but I'll still see what I can create. There is still room for improvement. I'll let you know!
Later,
Wolf

[Wolf]

Wolf,

Thanks for your help and the links to LouC's site. I had looked at his site quite some time ago and now remember where I got the idea to use the W4-1052 in the first place. Thanks Lou!

Using your instructions for RM and PCD I think I have come up with something a little more respectable. The front baffle size is 6.5" x 9.5".

Anyone who would like to give me some criticism, constructive or not please feel free.

Thanks!
Dave

If you send me an E-mail, I'll send you the .csp as I have it currently, and the files I used. I am having issues blending them together. I either get the inherent dip in the -1052's response around 1.5kHz, or a huge trough in the BBC region. When they align flat-ish, the phase alignment sucks. When I get the power response flat, and the FR flat-ish, the phase is ignored and the tweeter isn't protected enough, because the circuit is 2nd order with a few notches. If the reverse null is decent, I get the BBC trough. I spent about an hour on this, but you can fiddle from here.
Hope this helps,
Wolf

[gmilitano]

It's the Passive Crossover Designer, by Jeff Bagby. Look up Jeff B, and he has a link to his software on the internet. It's freeware, and a very powerful program. You'll need to have teh Response Modeler 2.0 as well, as it melds very well with the PCD program.
Regards,
Wolf

Thank you, I will give it a try.