ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 [2] 3 4 ... 9   Go Down

Author Topic: Crossover, L-Pad, Impedence, Resistance question  (Read 10006 times)

Jeff Schoonover1

  • Full Member
  • ***
  • Offline Offline
  • Posts: 127
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #10 on: November 15, 2017, 09:51:40 am »

Has anyone messed around with Windows Passive Crossover Designer (WinPCD)?  On first look, it seems to have a solution for about everything we've talked about, as well as off-axis issues, phasing, etc.
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8867
  • Atlanta GA
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #11 on: November 15, 2017, 01:13:23 pm »

I think you're absolutely right.  Regarding crossovers, and for whatever reason, real-world doesn't work out like the math says it will in my experience.
Actually the math works totally fine.

The problem is that people don't use the right numbers in the right places.

For example, they "assume" the impedance is 8 ohms.  But what if it is actually 30 ohms at the particular xover freq? 

You MUST consider the curve-NOT a single simple number.

Then you come up with different numbers.  OH BTW, that 30 ohms is not constant, a half octave lower it is actually 7 ohms, so now the "math" is different.

I am working on a cabinet now that is "rated" at 8 ohms, but around xover the actual impedance is just over 100 ohms.

And what if the amplitude has a big hump in it around xover (like most do).  No you have to electrically cross it over higher, in order to get a good acoustical xover.

All the real crossover calculators (we use LSPCAD), simply use the math.  HOWEVER, they have all the numbers to put into the right places, so it works very well.

It is yet another, trying to put a simple single number on a complex problem.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Jeff Schoonover1

  • Full Member
  • ***
  • Offline Offline
  • Posts: 127
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #12 on: November 15, 2017, 06:40:52 pm »

... people don't use the right numbers in the right places.
Or omit needed values.  As in your example, not factoring impedance and frequency response of drivers, etc. could be a mistake.  All I ever did was use a simple calculator or chart, without regard to these other important elements.
Logged

Marc Sibilia

  • Jr. Member
  • **
  • Online Online
  • Posts: 68
  • Princeton, New Jersey
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #13 on: November 16, 2017, 12:17:12 pm »

Has anyone messed around with Windows Passive Crossover Designer (WinPCD)?  On first look, it seems to have a solution for about everything we've talked about, as well as off-axis issues, phasing, etc.

I have used WinPCD to design a three way for my living room and find it very useful and pretty easy to use.  I prototype my speaker by building the cabinet, and installing the drivers.

Then, I adjust DSP crossovers until I get the measurements looking good and do final voicing by ear in my listening room.  Then I measure each driver individually in the cabinet with the DSP crossover active and each driver individually with no crossover (at low level on the tweeter) without moving the microphone or the speaker.  Then I measure the total response with the crossovers and all drivers active, again without moving anything.  Finally, I do impedance measurements of each driver in the cabinet.

With these measurements, I then proceed to design a crossover that matches the predicted response of the speaker in room with passive crossover to the actual measured response in room of the speaker with DSP crossover.

The key is that the measurement and the predicted response may not look perfect because of room interactions, but it will match the sound of the DSP speaker that you have spent a week playing with to get the sound right.  It is so much easier to A/B different crossovers using DSP than passives that I wouldn't do it any other way.

It works very well.  I set up the two speakers adjacent and touching (one active with DSP and one passive) and listen.  The passive design sounds very nearly identical to the DSP speaker.  There was a bigger difference in voicing by exchanging the speakers left and right than between the passive and the active crossover.

If I were doing a PA speaker with a passive crossover (I'll be building some coax wedge monitors soon), I would do my measurements and voicing outdoors rather than in my listening room.

Marc
Logged

Jeff Schoonover1

  • Full Member
  • ***
  • Offline Offline
  • Posts: 127
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #14 on: November 19, 2017, 10:39:09 am »

The key is that the measurement and the predicted response may not look perfect because of room interactions ...
It is so much easier to A/B different crossovers using DSP than passives that I wouldn't do it any other way.
I have been using the box with a Crown XTi 4K. Using that active X-over at LR24 1.3K, it's sounding better than the passive one ever did. I also am able to run the woofers at 4-ohms, to help overcome the efficiency differences with the tweeter. Additionally, I'm able to use EQ settings via my analyzer software to flatten out the peaks quite a bit. I had wanted to have the flexibility of using the other channel for lows on a compact rig, which is why I was wanting to go passive.  But this is sounding too good to ignore.
I would do my measurements and voicing outdoors rather than in my listening room.
I have found outdoors and away from reflections to be hugely valuable as well.
Logged

Jeff Bankston

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2269
Logged

Jeff Schoonover1

  • Full Member
  • ***
  • Offline Offline
  • Posts: 127
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #16 on: November 19, 2017, 03:55:38 pm »

YES !
That's what I'm doing, and it sounds noticeably better than the passive one ever did.

So then, I had a thought.  What if I put TWO Speakon inputs on the back of my box instead of one.  Then somehow make a switch that would make one as a two-pole input which would go to the crossover.  Throwing the switch and connecting to the other would be a 4-pole which would disconnect the crossover and switch in wiring straight to the speakers.  Best of both worlds. 
Not sure they make a switch with that many poles on it though.
Logged

Paul G. OBrien

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 991
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #17 on: November 19, 2017, 04:18:44 pm »

You don't need 2 speakon connectors to do this but they can be useful other ways. This is what you need electrically.. 1 DPDT switch for the LF section and another for the HF.

« Last Edit: November 19, 2017, 04:20:54 pm by Paul G. OBrien »
Logged

Jeff Schoonover1

  • Full Member
  • ***
  • Offline Offline
  • Posts: 127
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #18 on: November 19, 2017, 08:53:41 pm »

You don't need 2 speakon connectors to do this but they can be useful other ways. This is what you need electrically.. 1 DPDT switch for the LF section and another for the HF.
Nicely done!

I think I see what you mean - this could be done with one 4-pole connector (?) And the same standard 4-conductor cable.   Position 'A' could use 1+ and 2+ only for when it's mono, lifting the negatives to prevent a short.  Position 'B' would switch in 1-/2- and re-route directly to drivers, bypassing the crossover.
Then I wouldn't even need to use a different cable (or adapter at the amp-side) for mono vs when it's bi-amped.

I wonder if it could be done with one switch though.  Seems like it might get confusing if someone other than me sets up the rig? Or stupid me would forget to switch both and think I'd blown a driver, LOL

Logged

Paul G. OBrien

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 991
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #19 on: November 19, 2017, 09:39:53 pm »


I wonder if it could be done with one switch though.

Yes, technically you really only need to switch the positive connections to take the crossover out of the circuit.
Logged

ProSoundWeb Community

Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #19 on: November 19, 2017, 09:39:53 pm »


Pages: 1 [2] 3 4 ... 9   Go Up
 



Page created in 0.128 seconds with 25 queries.