Crossover 3-way, working out the kinks- 1st timer

[Johnbob9]

Looking to design a 3-way crossover using the following:

Goldwood GT-400CD Bullet Piezo Horn Driver


* Similar to KSN1142A * Power handling: 100 watts RMS. * Frequency response: 1800-30000 Hz * SPL: 92 dB 1W/1m *
Mounts to any 1-3/8", 18TPI horn * Designed for use with our horn * Manufacturer model number: GT-400CD *
Dimensions: diameter: 2-7/16", depth: 1-1/2".

Pyle Pro PDMW6 6-1/2" Mid-Woofer

* Power handling: 125 watts RMS/250 watts max * Voice coil diameter: 1" * Impedance: 8 ohms *
Frequency response: 70-9,000 Hz * Magnet weight: 25 oz. * Fs: 67 Hz * SPL: 86.9 dB 1W/1m * Vas: 0.36 cu. ft. *
Qms: 4.99 * Qes: .95 * Qts: .80 * Xmax: 3.3 mm

Selenium 15PW3-SLF 15" Driver

Specifications: • Power handling: 250 watts RMS/350 watts max • Voice coil diameter: 2-1/2" •
Voice coil inductance: .70 mH • Nominal impedance: 8 ohms • DC resistance: 5.8 ohms • Frequency response: 40-4,000 Hz •
Magnet weight: 56 oz. • Fs: 37 Hz • SPL: 98 dB 1W/1m • Vas: 9.67 cu. ft. • Qms: 5.81 • Qes: 0.68 • Qts: 0.61 • Xmax: 3.0 mm •
Net weight: 9-1/2 lbs. • Dimensions: Overall Diameter: 15",
Cutout Diameter: 13-3/4", Mounting Depth: 6", Magnet Diameter: 6-3/4", Magnet Height: 1-3/8".


Trying to figure out the following

If I were to use this calculator
(3-Way Crossover Designer / Calculator)

assuming the following are the crossover points

low
70-4000
high
1800-4000

Do I enter theses as the crossover, or do I use the desired frequency response.

low
40-4000
high
70-30000

How do I calculate the Bandpass gain, as there seems no simple google answer, I have gathered that it isn't easy math to find this number.

I already asked a few questions on this thread:
( 2 way crossover build (diy first timer))

I mentioned using a Peavy ECS3X pre-built crossover. Would buying this and testing it in box for responses a viable way to solve my problem?
Or
Is an online calculator a better option.

I am new to this, looking for help to solve this problem.

Thank you.

John B.

[Jimbates]

I'm using That tweeter with a 400pb horn on my center speaker.
I found that placing a capacitor in series with it smoothes it response out.
It needs to be calculated as a 20 ohm capacitive load, which means you need .25mf to 1 mf as a crossover( mines 4-1mf caps in series,to make a high power .25 mf). other crossover components are likely to be unnecessary, Because your other drivers will be dong most of the work below 10k. If theres any good news the 1 mf caps are cheap so if they dont work for you theres minimal $ loss.for the other drivers a conventional 6 db/octave crossover should be close to what's desired.
Good luck with your project !

[sreten]

Hi,

FWIW you'd probably get better results leaving out the mid driver,
it is not very sensitive. Read the wiki here on piezo crossovers.

rgds, sreten.

[Jimbates]

Im using the tweeter on on a center speaker, powered by a crown xls202
It is in a 2 way configuration with a 12" generic woofer.

Sreten isn't far from the mark in thinking you can skip the 6 1/2 speaker. It will leave a small "hole" in the midrange response, but doesn't sound unappealing. In fact it works pretty well.
What I was thinking with the original question is that, since the woofer is more efficient than the other speakers, a run of the mill 800 hz first order filter will allow the woofer to still have output(-12db) up to about 3200 hz. crossing the 6 1/2 over at about 3200 hz, will cause the mid and woofer to have the same efficiency at that frequency, also your midrange will handle more power is it's not trying to produce bass. The reason to push the crossofer frequencies upward is a matter of power handling.
Goldwood claims this tweeter handles 100 watts, but the motorola version is rated at 200w. . the high crossover will help ensure the tweeter can absorb a 200 watt amp.. Mine does.
Again, I wish you luck with your project, Jim

[sreten]

Hi,

Whacking a big inductor on the 15" to implement BSC will bring down
the level to match the tweeter - though I've no idea of 15" response.

It might be that simple, it might not be, information required.

rgds, sreten.