Planning 3 way SEAS monitor pair

I am planning a 3 way Studio monitor stereo setup using SEAS components and figured it would be wise to make a stop over here in case anyone can point out potential pitfalls.

Details

Components:
Tweeter: Seas Prestige 19TFF1 (H0737)
Mid-Range: SEAS Prestige MCA15RCY (H1262) 5.5" Coated Paper Midrange
Woofer: SEAS Prestige CA26RFX (H1305) 10" coated paper woofer

Port:
I am not sure what I want for a port setup, as I value time-domain response far more than frequency response so I suppose I would stick with a sealed cabinet. I did try to pick speakers that were matched in sensitivity and their respective bandwidths, though.

Goals:
However, in the lack of any actual available waterfall graphs on these speakers, I've just been planning around frequency response while taking stabs at preserving a reasonably dampened time domain characteristic.

For the woofer's frequency response, I am hoping to get 50-200 Hz +- 3db out of it.

For the midrange, 200-5000hz? Does that seem reasonable? It's just before the breakup on the graph. Maybe I should cross that over lower...

For tweeters, 5k-20khz. The tweeter's manufacture graph suggests it can go down to 2k easily which seems in line with Adam A7s I have on hand from what I can tell. Maybe I should really consider crossing over the tweeters lower.

Method:
To design the crossovers I was just going to use the XSim software after I get the FRD files generated with the cabinet built. The cabinet will most likely be a 3-way classic design...but it would be nice if there was another piece of software that could simulate the enclosure's transfer function combined with the speaker parameters as well? I can see how XSIM measures that but by then I've already built the cabinet haha.

The lynch pin of the build though is if the 10" woofer offers a slower transient response with no added frequency response compared with smaller drivers, then why use it? Should I just copy the 3-Way Classic entirely IOW just buy it lol?
 
Last edited:

OllBoll

Member
2010-11-22 11:24 pm
5 khz for a dust capped 5" midrange is probably pushing it a bit too high.

Keep in mind that you don't just want to match the response at 0 degrees but on the polars too so look at the 30 and 60 degree dispersion. This is the big reason why waveguides are so good as they let the tweeter crossover dispersion match the midwoofer / midrange driver. The 27TBCD/GB-DXT is a pretty good tweeter here as it has a smaller waveguide out of the box.

Also do you plan to build a passive crossover or active with DSP?
 
Last edited:
5 khz for a dust capped 5" midrange is probably pushing it a bit too high.

Keep in mind that you don't just want to match the response at 0 degrees but on the polars too so look at the 30 and 60 degree dispersion. This is the big reason why waveguides are so good as they let the tweeter crossover dispersion match the midwoofer / midrange driver. The 27TBCD/GB-DXT is a pretty good tweeter here as it has a smaller waveguide out of the box.

Also do you plan to build a passive crossover or active with DSP?
Passive, but the thought of digital filters is surely tempting. Thanks for the axis response suggestion.
 

OllBoll

Member
2010-11-22 11:24 pm
Passive, but the thought of digital filters is surely tempting. Thanks for the axis response suggestion.

I vote for active with DSP and a cheap measurement microphone like the UMIK. It makes it much easier to get a really good speaker in the end. Normally you'd have to build the woofer box to match Q of the driver and all that, then match the sensitivity on all drivers and on top of all that match the off axis dispersion.

With DSP you can pretty much build any box that is convenient and make the response what you want. Sensitivity on the drivers is a non issue so in the end you pretty much only have to match on and off axis response. And even here you can just move the crossover frequency until they match nicely.

Of course you don't want to strong-arm and use excessive correction but small adjustments are no issue and make everything much easier.
 

rectifryer

Member
2014-06-08 3:22 am

Yes, that helps immensely. Thanks a bunch!

I vote for active with DSP and a cheap measurement microphone like the UMIK. It makes it much easier to get a really good speaker in the end. Normally you'd have to build the woofer box to match Q of the driver and all that, then match the sensitivity on all drivers and on top of all that match the off axis dispersion.

With DSP you can pretty much build any box that is convenient and make the response what you want. Sensitivity on the drivers is a non issue so in the end you pretty much only have to match on and off axis response. And even here you can just move the crossover frequency until they match nicely.

Of course you don't want to strong-arm and use excessive correction but small adjustments are no issue and make everything much easier.

To be honest I simply have hesitation over another step of ADC/DAC conversion as I like the humble converters I have (just an audient ID22). It looks like the miniDSP board has a worse SNR, although that isn't everything.
 
Last edited:

rectifryer

Member
2014-06-08 3:22 am
I have misgivings about that tweeter here.

Poor Man'

Bit like a Poor Man's Strad isn't it? Or a Celestion Ditton 44.

I don't think the 88dB SEAS 19TFF1 is loud enough to work easily with a 10" bass. And you might want a lowish crossover with a 5" mid, and again it might struggle.

Somebody did a three way classic with a SEAS 10" bass not to long ago. It seemed to work.

3 ways are often difficult on impedance, especially where you push the mid and tweeter loud, so can't pad the filter with resistors.