Are my electronics correct?

[icor1031]

If I want to cross at 1900Hz, and I am using these drivers:

http://www.parts-express.com/pe/show...umber=264-1078
http://www.parts-express.com/pe/show...umber=264-1026


Did I put the specs in for my drivers correctly?




Also, if my tweeter is 5dB louder than the woofer, did I do this correctly?



Thanks!

[greywarden]

Whoa, whoa, whoa.

Why are you using a textbook crossover calc when you have measurements?

[icor1031]

..According to my measurements, 1,900-2,000 was the best place to cross??

Whoa, whoa, whoa.

Why are you using a textbook crossover calc when you have measurements?

[Chris Roemer]

2k will probably be OK, but even with 1/4" space between them, 2.5k would still be good. That would be more comfortable for the tweeter, and the woofer is small enough that it won't be beaming by then (and it looks so dang flat, that THAT won't be a problem).

I'd shoot for more like -8dB on the tweeter to give yourself about 3dB of baffle step compensation.

That should break 40Hz in a 0.87cf box with a 2"id x 4" vent.
Even in 0.57cf, with a 2"d x 5" vent, it should do mid 40's.

You've been here a while. What's up with the "textbook" calculator?

[icor1031]

I have been here a while, but I have barely touched on electronics.
Most of what I know is related to boxes and physics.

What should I be using for choosing electronics? Will it change any thing if I get an MS Office Trial and use PCD?

I'll probably check the BSC again, but in my measurement I didn't need any - even against the wall.

Last, in my measurements, 1.9k-2.0k was the best place to cross for distortion.

2k will probably be OK, but even with 1/4" space between them, 2.5k would still be good. That would be more comfortable for the tweeter, and the woofer is small enough that it won't be beaming by then (and it looks so dang flat, that THAT won't be a problem).

I'd shoot for more like -8dB on the tweeter to give yourself about 3dB of baffle step compensation.

That should break 40Hz in a 0.87cf box with a 2"id x 4" vent.
Even in 0.57cf, with a 2"d x 5" vent, it should do mid 40's.

You've been here a while. What's up with the "textbook" calculator?

[neildavis]

If I want to cross at 1900Hz, and I am using these drivers:

http://www.parts-express.com/pe/show...umber=264-1078
http://www.parts-express.com/pe/show...umber=264-1026


Did I put the specs in for my drivers correctly?

Take a look at the impedance curve for your woofer--it is in the spec sheet. It's hard to read the exact value, but it looks like at 1900Hz the impedance is about 12ohms. The tweeter graph is even harder to make out, but it looks like it is 5 or 6 ohms at 1900Hz. So why would you use 8 ohms and 4 ohms for your textbook filter--at your crossover frequency those aren't the correct values for the impedance. And at that frequency the response of the woofer is still extending well past the crossover point, whereas the tweeter drops sharply below that crossover point. So unless you use extremely sharp filters (like an 8-pole crossover), the sum of the drivers and crossover response probably isn't going to result in a flat response.

Just trying to help you visualize why you need to use real measurements and a modeling program to predict the response. If you don't have Excel, use the PSD_Lite program at this link: http://www.audiodevelopers.com/Software/PSD_Lite.zip. If you don't have the measurement files, you can usually get better results than a textbook filter by using posted response files for drivers that are "similar".

[icor1031]

Thanks for explaining how to set the impedance. I had assumed that it wanted the 'nominal.'

Also, I do have response measurements, which I took - but not impedance measurements.

Take a look at the impedance curve for your woofer--it is in the spec sheet. It's hard to read the exact value, but it looks like at 1900Hz the impedance is about 12ohms. The tweeter graph is even harder to make out, but it looks like it is 5 or 6 ohms at 1900Hz. So why would you use 8 ohms and 4 ohms for your textbook filter--at your crossover frequency those aren't the correct values for the impedance. And at that frequency the response of the woofer is still extending well past the crossover point, whereas the tweeter drops sharply below that crossover point. So unless you use extremely sharp filters (like an 8-pole crossover), the sum of the drivers and crossover response probably isn't going to result in a flat response.

Just trying to help you visualize why you need to use real measurements and a modeling program to predict the response. If you don't have Excel, use the PSD_Lite program at this link: http://www.audiodevelopers.com/Software/PSD_Lite.zip. If you don't have the measurement files, you can usually get better results than a textbook filter by using posted response files for drivers that are "similar".

[Chris Roemer]

Just trace the mfr impedance curve off the spec sheet using "SPLTrace" (used to be on the FRD Consortium site, not sure the site exists anymore).

[greywarden]

It is, it's linked from RJB audio site.

[icor1031]

Is there a program I can use, other than HolmImpulse - to get my frequency response and create an FRD file - so I can use PSD??

I ask, because I don't have excel - and I don't know what this is about creating a formula:

http://techtalk.parts-express.com/sh...es-from-Holm-s

[Chris Roemer]

I thought you said you HAD (frequency) response measurements? Those would be ....frd files.
Maybe you don't have the impedance files ...zma?
You can get them from a Woofer Tester (name changing to DATS - Dayton Audio Test System).

Alternately, if you can find a mfr. graph (with the response and impedance curves on it) you can "trace" the curves on your computer screen using "SPLTrace".

Chris

[greywarden]

Or SPL Copy if you use the FR graphs from Holm

[icor1031]

I do, holm created them as .txt.
Do I just rename it?

I thought you said you HAD (frequency) response measurements? Those would be ....frd files.
Maybe you don't have the impedance files ...zma?
You can get them from a Woofer Tester (name changing to DATS - Dayton Audio Test System).

Alternately, if you can find a mfr. graph (with the response and impedance curves on it) you can "trace" the curves on your computer screen using "SPLTrace".

Chris

[neildavis]

I do, holm created them as .txt.
Do I just rename it?

You have to do that for PCD. With PSD_Lite, you can select either FRD or TXT as the file extension. The parser for PSD_Lite is fairly robust--it will reject blank lines, accept any whitespace and non-numeric character as a delimiter, and reject garbage at the end of the line. If there are only two values in the line, it will set the phase to zero. It does some other error checking as well. The only limitation is that the file must be less than 64K lines (I just picked that as a practical limit).

[icor1031]

Excellent.

Thanks, gentlemen.

You have to do that for PCD. With PSD_Lite, you can select either FRD or TXT as the file extension. The parser for PSD_Lite is fairly robust--it will reject blank lines, accept any whitespace and non-numeric character as a delimiter, and reject garbage at the end of the line. If there are only two values in the line, it will set the phase to zero. It does some other error checking as well. The only limitation is that the file must be less than 64K lines (I just picked that as a practical limit).