Design me a crossover and I'll build it

[CharlieLaub]

Need a crossover for a center channel I want to build this summer.

See attached for a pic of the front baffle. Volume is likely to be 1-1.25 ft^3, sealed. Front baffle dimension is 8"x20" WxL.

Drivers are:

Vifa DQ25SC16-04
Fountek FR88EX
2x Silver Flute W17RC38-08

The unit will be below ear level but can be tilted up to have its forward axis meet the ears.

Not looking to spend too much, as you can see from the driver selection. I am OK with steel-core inductors but would prefer poly caps. I expect the crossover points should be somewhere around 2 kHz and 400 Hz. Slopes are up to you.

Isn't the center channel mostly for voice? The bandwidth on that is something roughly like 100Hz - 3000Hz. You could keep it simple and use a single 4" or 5" extended range driver with no crossover. Tang Band makes some reasonably priced full range drivers that would work well. IMO I don't see a need to make a complicated 3-way for the center channel...

-Charlie

 

[454Casull]

Isn't the center channel mostly for voice? The bandwidth on that is something roughly like 100Hz - 3000Hz. You could keep it simple and use a single 4" or 5" extended range driver with no crossover. Tang Band makes some reasonably priced full range drivers that would work well. IMO I don't see a need to make a complicated 3-way for the center channel...

-Charlie

SPL and off-axis response.

 

[454Casull]

Mid and tweet are from zaph, woofer from manufacturer.

The W17RC38-04 is on Zaph's page. Would it make any sense to use this instead of the -08 version?

Are you able to generate polars (horizontal and vert) with your software?

Thanks

 

[Jay1111]

If you wanted to use the 04 version I could modify the design for it, it would be more accurate. The sim will be very close as is though.

No I cant generate polars.

 

[454Casull]

If you wanted to use the 04 version I could modify the design for it, it would be more accurate. The sim will be very close as is though.

No I cant generate polars.

Will it affect the crossover components or the system sensitivity?

Is the tweeter HP at 3700 Hz because the crossover becomes simpler? Or is it because the response is easier to keep smooth?

Sorry for all the questions.

 

[Jay1111]

Will it affect the crossover components or the system sensitivity?

Is the tweeter HP at 3700 Hz because the crossover becomes simpler? Or is it because the response is easier to keep smooth?

Sorry for all the questions.

Sensitivity might be a challenge, might lead to crossing the woofers over a lot higher.

The mid/tweet crossover was where phase, smooth response, and acceptable impedance all lined up the best. Trying to force a lower xover with a smooth response caused the impedance to drop very low. This is the benefit to having a very small/clean mid with tight CTC spacing, there isnt a negative to crossing this high.

 

[454Casull]

I'm sold.

Will be ordering parts in a week or so. Let me know if you come up with any improvements or catch any errors in the parts list.

Which components require tight values? I'd like to replace the 68 uF with a 69 uF.

 

[454Casull]

How much does DCR matter on the inductors?

Also, L9 0.1 mH is shown as 20 ga on the schematic but 16 ga on the parts list. Which one to use?

 

[Jay1111]

The 69 uf cap wont make a difference. DCR is fairly important, depending on the location. Everything is spec'd for the DCR of 20ga air cores except for the woofer. You could use a lower DCR part for the 3.9mh inductor on the mid without much change. I wouldnt mess with any of the other inductors. Here is the response exactly as spec'd (with the 69uf cap).

An 18ga iron core on the woofer will raise its level another db.

 

[454Casull]

Can you look over this when you have a chance?

 

[Jay1111]

That will work. Are you combining smaller values for cost saving?

Make sure you follow Zaph's advice for mounting the driver

This is without a doubt the highest performing 3" I've come across, at least on a few fronts. The motor is an underhung neodymium design and the cone is thin (and fragile) formed aluminum. The only negative is a slightly ragged top end that may take some filter work if used full range, but I'd generally recommend a cheap and small neo tweeter placed close and crossed over high. There's a lot on the positive side, like high sensitivity, a clean low end and smooth midrange response. The frame is cast with ventilation under the spider, but users will have to pay close attention to airflow behind the cone - It's even more important to chamfer the inside baffle opening and in this case likely use a thinner baffle material as well.

 

[454Casull]

Yes, somewhat. Solen's polyester series is cheaper than the series you picked (probably MKP) but they didn't have the exact values required, so I needed to parallel smaller caps to get the right values.

I am going to chamfer away as much as I can. I needed to do the same for some Hi-Vi B3S drivers I had, too.

 

[454Casull]

There are 5 inductors in this crossover. I'm worried I may not have enough space to isolate them properly. Should I put in some steel sheet between the inductors to form maze-like walls?

 

[Jay1111]

Never thought about that, not sure if it would work? For my center build I put the woofer/tweeter crossover behind one of the woofers, and the mid crossover behind the other woofer.

 

[454Casull]

I believe it would shield the fields just fine. I just don't know if it would affect the inductance (being that the permeability is so high).

 

[454Casull]

The dip at 2.3 kHz is coming from baffle diffraction? The Fountek should be smooth there. I know there is a dip at 1.2 kHz from the driver itself.

I'm wondering how much I can fix with liberal application of wool felt.

 

[Jay1111]

Yes, the dip/peak/dip in the mids response is all related to its placement on the baffle.

 

[dumptruck]

The response graph you're looking at is what, +/- 1.5dB? There's little point in analyzing the little bumps in that. The real question is how close Jay's estimates on the acoustic offsets end up being and (relatedly) whether the overall power/polar response ends up being a voicing you like in-room.

 

[454Casull]

Won't know until I try. It's a hell of a lot better than I would be able to do on my own anyway.

I think I will be adding some F-13 felt on the baffle.

 

[454Casull]

Er, it looks like the width of the baffle is actually going to be 22", not 20". Does this affect the crossover?