P.E. crossovers any good?

droeses58

droeses58

Audioholic
The reason I ask, is I'm having a problem with my diy speakers which amount to a dayton reference 8" woofer with a reference dome tweeter [don't have the parts numbers in front of me at the moment] but the woofer is rated to 2000Hz and the tweeters fs are 600Hz

I didn't know squat about crossovers so I just bought there 2 way 2k pre-made crossver and it sounds decent until I turn it up to say about 90db then it starts to break up badly [sounds like crumpling cellophane] and just gets worse the louder I turn it up.

Now, I also built some smaller satellite speakers using the same tweeter and same reference style woofer but smaller 5". I built my own 2k crossovers and don't have the breakup problems that I have with the pre-made crossover and 8" woofer.

It sounds like to me that the tweeter is whats breaking up, but could it be the woofer that I'm hearing? That 2000Hz is just to much for it?

On a side note I would like to incorporate a bsc into the mix and at P.E. they give an example of a 1.0mh inductor and an 8ohm resister is this placed just in the woofer segment before the crossover or is it placed before the "whole" crossover?

Sorry about the long post, thanks for any replies.
 
Swerd

Swerd

Audioholic Warlord
Stock crossovers just won't do here

droeses58 said:
The reason I ask, is I'm having a problem with my diy speakers which amount to a dayton reference 8" woofer with a reference dome tweeter [don't have the parts numbers in front of me at the moment] but the woofer is rated to 2000Hz and the tweeters fs are 600Hz

I didn't know squat about crossovers so I just bought there 2 way 2k pre-made crossver and it sounds decent until I turn it up to say about 90db then it starts to break up badly [sounds like crumpling cellophane] and just gets worse the louder I turn it up.

Now, I also built some smaller satellite speakers using the same tweeter and same reference style woofer but smaller 5". I built my own 2k crossovers and don't have the breakup problems that I have with the pre-made crossover and 8" woofer.

It sounds like to me that the tweeter is whats breaking up, but could it be the woofer that I'm hearing? That 2000Hz is just to much for it?

On a side note I would like to incorporate a bsc into the mix and at P.E. they give an example of a 1.0mh inductor and an 8ohm resister is this placed just in the woofer segment before the crossover or is it placed before the "whole" crossover?
Are those the Dayton Reference Series aluminum drivers that you have? The tweeter is the RS28, the 8" is the RS225, and the 5" is the RS125. They are real good drivers.

That tweeter crossed at 2 kHz is probably OK and should not be the source of the noise you hear. The problem you've described sounds to me like it may be caused by using an 8" aluminum driver with a crossover not custom designed for it.

Go to the PE site and look at the sample response graph of the RS225. You'll see it looks fairly smooth up to about 2 kHz, with some real bad break-up noise beginning above 3 kHz and peaking in the 5-10 kHz range. This is typical of many aluminum drivers. Why should that be a problem with a 2 kHz crossover? I'm guessing that you bought a stock PE crossover with a 2nd order slope. Look at the magnitude of the breakup noise peaks. They are as much as 10 dB greater than the smooth response below 2 kHz. You will hear that despite the crossover.

Remember that in sound, one loudness level 3 dB higher than another is twice as loud, 6 dB is 4 times as loud, etc. A 2nd order crossover rolls-off the signal to the woofer gradually at 12 db per octave beginning at 2 kHz. So if the loudness is at about 88 db at 2 kHz, it should be at 76 dB at 4 kHz, 64 dB at 8 kHz, and so on. But if the breakup noise peaks are much higher than the smooth part of the curve below 2 kHz, you will hear it as high frequency noise. You should be able to tame that with a 4th order crossover, with a steeper roll-off slope of 24 dB/octave. If that still isn't good enough, a notch filter can be used to further quiet the worst peaks at 7-10 kHz. Stock crossovers just won't do for the 8" woofer.

If you look at the response graph of the 5" RS125, you'll see why you don't have a similar problem with a 2 kHz crossover. There are large breakup peaks, but the loudest one is well above 10 kHz. What kind of crossovers did you make for them - 2nd order?

If you go to the PE site's Tech Talk discussion board, and ask there if anyone has a 2-way crossover design for the RS225 and the RS28, you should get plenty of useful answers.
 
droeses58

droeses58

Audioholic
Thanks for your reply,

Yes those are the drivers I'm using, I guess I never thought that a drivers breakup 2 or so octaves above the crossover would be noticeable. But after your explanation it makes more sense.

Now is the breakup I'm hearing coming "out" of the woofer itself? or is it being directed through the tweeter?

I was thinking of lowering the crossover to 1700Hz, good idea or bad?

If I use a notch filter could you explain how and what size components?

Yes I built the satellite crossovers from scratch at 2k 2nd order, but a little better components then the pre made P.E. ones.

Can you recommend a Baffle Step Compensation network? Its a sealed design with a baffle about 12"w and 15"h

Thanks again for your response.

EDIT: I forgot to mention, I also have a zoebol network in the 8"
 
WmAx

WmAx

Audioholic Samurai
Swerd said:
Remember that in sound, one loudness level 3 dB higher than another is twice as loud, 6 dB is 4 times as loud, etc.
6dB is twice the actual amplitude/loudness increase, not 3dB. 10dB is the accepted standard level increase for percieved doubling of amplitude/ loudness according to human test subjects.

I think you may have confused the doubling of amplfier power to increase amplitude by 3dB with 3dB being twice as loud.

-Chris
 
Swerd

Swerd

Audioholic Warlord
WmAx said:
6dB is twice the actual amplitude/loudness increase, not 3dB. 10dB is the accepted standard level increase for percieved doubling of amplitude/ loudness according to human test subjects.

I think you may have confused the doubling of amplfier power to increase amplitude by 3dB with 3dB being twice as loud.

-Chris
I often get that mixed up, thanks for the correction.
 
Swerd

Swerd

Audioholic Warlord
droeses58 said:
Now is the breakup I'm hearing coming "out" of the woofer itself? or is it being directed through the tweeter?
If what I said is actually your problem, the woofer would make the noise, not the tweeter. Test it by listening after disconnecting the tweeter. If the tweeter is making the noise, something else quite different may be going on.
droeses58 said:
I was thinking of lowering the crossover to 1700 Hz, good idea or bad?
That might be a step in the right direction, but you would still have to take stronger steps to filter out the high frequency breakup noise. The RS28 tweeter is said to be able to be crossed as low as 1600 Hz without distorting. I would try a 4th order crossover. Go to http://www.mhsoft.nl/spk_calc.asp and click on the link for 24 dB Crossover filter. Note that the diagram just below the calculator box shows a 24 dB filter, but it is mislabled as 18 dB.
droeses58 said:
If I use a notch filter could you explain how and what size components?
Go to http://www.mhsoft.nl/spk_calc.asp and click on the link for Parallel Notch Filter. Fill in the values for F (center frequency of the range you want to filter, F1 (lower frequency of the range), and F2 (upper frequency of the range). Click on calculate to see the values for C (capacitor), L (inductor), and R (resistor), and assemble them as shown in the diagram in series with the woofer.
droeses58 said:
Can you recommend a Baffle Step Compensation network? Its a sealed design with a baffle about 12"w and 15"h
I thought I had a link for calculating BSC. Read this link http://www.quarter-wave.com/General/BSC_Sizing.pdf

You should still ask people at the PE Tech Talk board about this. Someone may have already made a design for these drivers that you can use. They may also know of other (perhaps better) website to calculate these filters and circuits.
 
droeses58

droeses58

Audioholic
Thanks again for the reply Swerd, I tried the 4th order on the woofer only but I had to disconnect the zoebel network so I could use the cap. in the crossover instead. I still get the distortion but I'm sure if I added the zoebel back in it would be less.

I'll check with the tech forum at P.E.

Thanks for the link on the bsc, explains it rather well.
 

Latest posts

newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top