Checking Crossover before buying components

[Sdiver2489]

I finally got around to designing a full crossover for my M6a and Seas TDFC/TV project. Because I was unsure about a few things I was wondering if someone could look at this to make sure that it is done correctly(obviously there is more than one way to do this, but I am just talking about obvious errors)

I crossed at 1600 Hz, with a 4th order linkwitz riley and a second order on the tweeter to compliment the natural 2 order HP filter for the tweeter.

I attenuated 2 DB to accomadate for the difference in sensitivity.

When entering this into SW I took into account the series resistance when entering the second order Filter(not sure if I was suppose to do that)

There is also a filter to counteract the inductive rise, I will be remeasuring this outside of SW since Hi-Vi did not provide the voice coil inductance of this speaker.

I know that I should use a lot of trial and error and such but I don't have the time to do that before I go back to college, so I am just trying my best with this and might fiddle with it later on. (I don't have the mics and such necessary to model this with SW)

Also, I want to thank all of you who have helped me out with this, it has all been very helpful in getting me this far. And JonMarsh.... ;x(

I am trying to keep this fairly low budget which is why I skipped out on a higher order crossover, maybe someday I will go back and redesign it once I have money

[Davey]

Hmmmm.

That's a high-pass filter you have on your M6 driver. I assume you want that to be a low-pass, correct? You need to swap the positions of the inductors and capacitors and start again.
The tweeter filter doesn't look too bad. However, L1 seems like a larger value than it should be. Rp2 is also rather large and could probably be eliminated.

Cheers,

Davey.

[GrahamT]

I'm no expert, but I'll try to help. You get what you pay for though.

One thing I see right off the bat is that it looks like you created a 4th order high pass on the M6a by accident. Edit: Davey beat me to it.

I crossed at 1600 Hz, with a 4th order linkwitz riley and a second order on the tweeter to compliment the natural 2 order HP filter for the tweeter.

I'm not sure what you mean by "natural 2nd order HP filter for the tweeter", I'm guessing that you mean 2nd order rolloff? That crossover point seems low to me, and the 2nd order filter is probably not steep enough.

I attenuated 2 DB to accomadate for the difference in sensitivity.

Did you make any adjustments for baffle step compensation though?

You say you dont have the equipment for measurement. Do you have a RS SPL meter? That would work. It seems to me that you are doing this the hard way. At least ask around for SW files for the M6a and Seas. I dont want you to waste your money or be dissapointed.

I know what you mean about the deadline for college. I'm scrambling to get my speakers done too.Good luck.

[Sdiver2489]

Whoops how embarassing. Here I fixed that slight problem. Thats what I get for working so late...Its been fixed in the orignal post so see the new attachment

I also meant to say that the crossover freq. is 1800 Hz. Not 1600.

Well, about the "natural second order" as you can see I don't know my terms yet, but the spec sheet says that the tweeter "rolls up"(my words) to 3500 Hz with a 2nd order filter. So if you combine that with my second order filter, it would be a combined 4th order no?

I have a very simple SPL meter that only reads to the decible(no 85.2 in other words). By asking for files are you saying to see if people have measured the acoustic response of the driver before? I suppose I can look around but BFC would be dependent on my baffle.

What type of circuit helps with BFC? As the frequency goes higher the SPL gets higher so what would I use...

I gotta head off to work, but thanks for the input already guys, I do want to do this the best I can with the time I have, although I think it would be hard to be disappointed because of what type of speakers I am coming from(provided these work as intended).

[GrahamT]

Well, about the "natural second order" as you can see I don't know my terms yet, but the spec sheet says that the tweeter "rolls up"(my words) to 3500 Hz with a 2nd order filter.

You could say that the low end response rolls off at 12 dB/octave, or second order acoustic attenuation, there is no filter.

So if you combine that with my second order filter, it would be a combined 4th order no?

Sort of. It is VERY tricky to do that without modeling it first. But if done right, the acoustic and electric attenuation would combine.

I have a very simple SPL meter that only reads to the decible(no 85.2 in other words).

Good enough, or at least better than nothing. I use the digital RS meter with a calibration file. You can see the results here: http://htguide.com/forum/showthread....eaker+workshop
Now it is not super accurate, but it will work. It is probably the best option for us poor students. The results were intentionally similar.

I used the meter with the Claudio cable jig ($10).

By asking for files are you saying to see if people have measured the acoustic response of the driver before? I suppose I can look around but BFC would be dependent on my baffle.

Yes, acoustic response and impedance. It would give you a better idea of what the filters are doing. Not perfect of course but maybe someone has them.

What type of circuit helps with BFC? As the frequency goes higher the SPL gets higher so what would I use...

Well there are a few ways, do some searches. But the easiest is probably to adjust the value of the first inductor on the low pass. At least that is what I do, but I'm a noob.

Hope this helps.

[Davey]

Here's what I came up with real quick for a simple high-pass crossover. I used Bjorn Magne-Idlands published SPL and Z values for the 27TDFC/TV driver. A simple second-order filter that will yield an approximate LR24 slope at 1800Hz with a few db attenuation:

http://home.comcast.net/~dreite/Temp/27TDFC.jpg

I'm sorry I can't help you with the M6a driver, but without accurate Z and SPL files it's pretty difficult.

Cheers,

Davey.

[Sdiver2489]

why are the resistors incorperated into the filter as they are? I thought the "L pad" attenuation was seperate from the 2nd order filter...is this another way to do this? Is there anything wrong with the way my filter was put together, I realize the numbers may not be the best, but just wondering if I at least put it together right.

[Teran]

One of the resisters isn't there

Try this in speaker workshop:

Create a network with your tweeter. Add a resister before the driver. Its value should be whatever it takes to pad down the response to your target.

Next, add a stock crossver, but when it asks for the resistance to match to, enter Driver Re + Resister value. So, for a RS28, enter something like 5.5 or 6 ohms.

How does that work for you?

Notice how you don't have to do quite as much Optimization to meet your goal?

[Davey]

Sdiver,

Yes, it's just another way to do it. A resistor could be placed after the filter as well. With just a couple of db attenuation required a single resistor is usually fine. Many systems have a higher sensitivity difference between drivers and may require more baffle-step correction. In that case a much higher reduction is required on the tweeter and a two resistor pad would probably work better. Generally.

I wouldn't order any parts until you get this sorted out better.

Cheers,

Davey.

[Sdiver2489]

so the R1011 is the resistor that creates the attenuation. What is the second resistor there for?

[AndrewM]

I believe that's the resistance of the inductor (LSPCad quirk IIRC)

Andrew

[Davey]

Yes, the resistor in series with the inductor represents its DC resistance. It's not a separate component.
0.3 ohms is probably a better estimate for the DCR of an appropriate inductor, but it doesn't make any difference for a quick estimation like this.

R1011 provides the attenuation.

Cheers,

Davey.

[Sdiver2489]

Hey, what do you guys think about this. I own a HK 435. Can I use the Supplied Mic that measures near field and far field speaker performance to record into SW? Would it work well enough to get an idea of what the crossover would do?

[GrahamT]

I have a very simple SPL meter that only reads to the decible

You can use any condenser mic and pre amp. The Radio Shack SPL meter would work. Many people have used it. Just use the RCA output. Already pre amped