ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1] 2 3 ... 8   Go Down

Author Topic: Crossover, L-Pad, Impedence, Resistance question  (Read 1566 times)

Jeff Schoonover1

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
Crossover, L-Pad, Impedence, Resistance question
« on: November 14, 2017, 06:34:27 pm »

Can I use an 8-ohm L-Pad after a crossover for a 16-Ohm tweeter?  I've read one can only use an 8-ohm L-Pad with an 8-Ohm driver.  It seems to me that resistance is resistance and I'd just have to factor the different resistances (tweeter and L-Pad) when designing the X-Over.  What am I missing?

I don't suppose there's a nifty calculator for all of this somewhere?  I've found some web-based calcs, but none include the ability to add an L-Pad to the equation.
Logged

Paul G. OBrien

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 729
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #1 on: November 14, 2017, 06:49:36 pm »

Are you talking about one of those continuously variable potentiometer type L-pads? If so then what power levels are we talking about here? These rotary devices are really only good for very small power levels.. like 50w or less regardless what power rating they have, if you need more than that you should be building a fixed L-pad from power resistors. It is OK to use what you have to determine the required resistance for each leg of the L-pad.
Logged

Jeff Schoonover1

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #2 on: November 14, 2017, 07:12:33 pm »

Are you talking about one of those continuously variable potentiometer type L-pads?
Yes, an 8-ohm model from Eminence.
If so then what power levels are we talking about here?
The amp driving this unit (a mid-high box) will be 2-3KW, but I expect only a tiny fraction of that will be in use most of the time.  I have run the same speakers before with a crossover built with components made for less than 200VDC.  The tweeter itself is rated for 100 Watts RMS.  I used the same 100W L-Pad for a short time, which didn't blow up, but I took it out as I was paranoid.  I did this with slightly less power, but again, I don't think I will ever come close to the maximum available.
Someone else designed that X-over, but I think they did it wrong - it sounded poor, and the pad didn't fix it.  This time, I want to do it from scratch and be sure.

These rotary devices are really only good for very small power levels.. like 50w or less regardless what power rating they have, if you need more than that you should be building a fixed L-pad from power resistors. It is OK to use what you have to determine the required resistance for each leg of the L-pad.
  So use the L-pad, to find a good level, and then swap it out with resistors in series and parallel?  What wattage should they be able to withstand?
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8195
  • Atlanta GA
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #3 on: November 14, 2017, 07:51:29 pm »

Yes, an 8-ohm model from Eminence. The amp driving this unit (a mid-high box) will be 2-3KW, but I expect only a tiny fraction of that will be in use most of the time.  I have run the same speakers before with a crossover built with components made for less than 200VDC.  The tweeter itself is rated for 100 Watts RMS.  I used the same 100W L-Pad for a short time, which didn't blow up, but I took it out as I was paranoid.  I did this with slightly less power, but again, I don't think I will ever come close to the maximum available.
Someone else designed that X-over, but I think they did it wrong - it sounded poor, and the pad didn't fix it.  This time, I want to do it from scratch and be sure.
  So use the L-pad, to find a good level, and then swap it out with resistors in series and parallel?  What wattage should they be able to withstand?
The idea behind a L pad is to provide a constant impedance load to the xover as the levels are adjusted.

The "simple answer" is yes, you can use a 8 ohm lpad with a 16 ohm driver.

HOWEVER, you MUST take this mismatch into account in the crossover design.

It depends on how you are doing the "Design".

If you are "designing" using simple xover calculators, the results will be off.

If you take into account the load impedance by measuring the actual impedance and amplitude response of the lpad and the driver, then enter that information into a program that will calculate the response, it can work.

It might be easier to simply measure the response/impedance of the woofer and add that to the HF response, and adjust/design the xover for the total response the way you need it, and don't even worry about the lpad.

As usual, it depends, on how the "design" is being done.
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

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #4 on: November 14, 2017, 08:54:51 pm »

I knew this could get complex, LOL.  Maybe I should just say what I have, and what I'm thinking, and ask for suggestions? 
I have a JBL 2426 horn, with which I can use either an 8ohm or 16ohm diaphragm.  Both on paper (110-117db, depending on horn) and in real life, it's easily 10-15 (or more) dB more efficient than the woofers, which are -
Two Eminence Delta 10A 10" speakers at 8 ohms each (so, 4 or 16 ohms load) with a sensitivity of 98.8dB each, (about 102dB or so with both.)

 The woofer's response is flat-ish until it begins to rise at about 900Hz until their peak at 2.4K, where they drop off sharply.
The tweeter's response is dependent on the horn used. But basically, its curve begins at about 900Hz and rises gradually over about a 10dB spread to about 12.5K where it drops off.
I was looking at different alignments (BW, L-R, 2nd, 3rd, 4th order, etc.). Then I saw that I already have the components to make a simple low pass with a 1st order BW at 850Hz.  For the high pass, a 1st order BW at 2K.  Considering the way the components behave, I'm thinking the large gap will be compensated by the much louder tweeter, the efficiency bump of the woofers over 900Hz, and the very gradual 6dB slopes of 1st order filters. 

I could even just use equalization on the higher tweeter frequencies if need be, and not even bother with an L-pad... 

With what I had before using these speakers, there was a jump around about 1.5-2.5K anyway which was unpleasant.

Thoughts?
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8195
  • Atlanta GA
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #5 on: November 14, 2017, 09:25:47 pm »

I knew this could get complex, LOL.  Maybe I should just say what I have, and what I'm thinking, and ask for suggestions? 
I have a JBL 2426 horn, with which I can use either an 8ohm or 16ohm diaphragm.  Both on paper (110-117db, depending on horn) and in real life, it's easily 10-15 (or more) dB more efficient than the woofers, which are -
Two Eminence Delta 10A 10" speakers at 8 ohms each (so, 4 or 16 ohms load) with a sensitivity of 98.8dB each, (about 102dB or so with both.)

 The woofer's response is flat-ish until it begins to rise at about 900Hz until their peak at 2.4K, where they drop off sharply.
The tweeter's response is dependent on the horn used. But basically, its curve begins at about 900Hz and rises gradually over about a 10dB spread to about 12.5K where it drops off.
I was looking at different alignments (BW, L-R, 2nd, 3rd, 4th order, etc.). Then I saw that I already have the components to make a simple low pass with a 1st order BW at 850Hz.  For the high pass, a 1st order BW at 2K.  Considering the way the components behave, I'm thinking the large gap will be compensated by the much louder tweeter, the efficiency bump of the woofers over 900Hz, and the very gradual 6dB slopes of 1st order filters. 

I could even just use equalization on the higher tweeter frequencies if need be, and not even bother with an L-pad... 

With what I had before using these speakers, there was a jump around about 1.5-2.5K anyway which was unpleasant.

Thoughts?
Here is the basic problem with using "off the shelf" passive crossovers.

They DO NOT take into account the ACTUAL impedance curve, or the freq response.

They assume both are flat and constant.

It would be VERY VERY rare that any designed (based on actual impedance and amplitude) would meet any standard slope or type (butterworth etc) curve.

I am not sure how you would design a passive 1st order xover of any particular type, as you say BW.  A 1st order rolloff via a capacitor is what it is, it has no particular shape.  If it did, what would be different if you choose to make it a 1st order LR or bessel?  With one part, the only thing you can do is change freq.

But be ready for the math not to work, unless the impedance and amplitude are flat.  In both cases, that would be rare around xover freq for a HF driver.

Often the math is actually a octave or two off to make it work properly

Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Paul G. OBrien

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 729
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #6 on: November 14, 2017, 09:29:23 pm »

Thoughts?

Biamp it.

1st order crossovers are not an option in Pro audio IMO, one of the worst things you can do is cross a CD too low or not steep enough, they just get ugly sounding and you risk blowing diaphrams.

So back to the biamping thing. Do you have an active crossover and a couple amp channels you could use to at least find out what minimum crossover points and slopes you need to use to keep everything alive and then sounding the best it can? With that info you could then build a passive crossover to match if  it is financially fesable... as in not more expensive than buying a second hand analog crossover and a small amp to drive the horns directly.

« Last Edit: November 14, 2017, 09:32:45 pm by Paul G. OBrien »
Logged

Stephen Kirby

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2498
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #7 on: November 15, 2017, 12:27:51 am »

As folks have said, the L-Pad itself is not the problem.  Most crossovers use fixed attenuation on the HF drivers although there are a few with switchable HF levels and things like bass guitar cabinets and very old "hi-fi" speakers that use variable attenuation.

Passive crossover design is a very complex art.  As Ivan said, the reactance (not resistance because it isn't static with different frequencies) of a speaker system (note I said system as the cabinet and other things hooked to the crossover play a part) is a complex curve.  Creating filters that balance things out is complex.  Then if you really want it to be good, you have to balance out the phase shift of those filters (or create additional phase compensation with other out of band filters) against the reactance of the drivers in the cabinet to try and get reasonable phase response.

For an amateur it's probably easier to get a simple DSP and start learning about this stuff when there are less levers to push or at least the DSP can compensate for some things while you push some levers.
Logged

Chris Grimshaw

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 401
    • Grimshaw Audio
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #8 on: November 15, 2017, 03:26:43 am »

Biamp it.

1st order crossovers are not an option in Pro audio IMO

+1.

For what it's worth, a second-order highpass only gives constant excursion as frequency decreases. For high-power PA use, you want excursion to decrease quickly as frequency goes down, so compression drivers should get a 4th order crossover IMO. 3rd order minimum.

Chris
Logged
Sheffield-based sound engineering.
www.grimshawaudio.com

Jeff Schoonover1

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
Re: Crossover, L-Pad, Impedence, Resistance question
« Reply #9 on: November 15, 2017, 09:49:39 am »

Here is the basic problem...
be ready for the math not to work...
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.
Biamp it. 
Do you have an active crossover and a couple amp channels you could use to at least find out what minimum crossover points and slopes you need to use to keep everything alive and then sounding the best it can? With that info you could then build a passive crossover to match...
I do, and thought of just that when I woke up this morning.  I have a Crown with active processing which I can set up on the box and use it's active X-over.  Figure out what sounds best, and then try to build a passive once I find a good set-up?
As folks have said, the L-Pad itself is not the problem.  ...Passive crossover design is a very complex art.  As Ivan said, the reactance (not resistance because it isn't static with different frequencies) of a speaker system (note I said system as the cabinet and other things hooked to the crossover play a part) is a complex curve.  Creating filters that balance things out is complex.  Then if you really want it to be good, you have to balance out the phase shift of those filters (or create additional phase compensation with other out of band filters) against the reactance of the drivers in the cabinet to try and get reasonable phase response.
This is another worry of mine. That the active X-over I use from my amp will not match the passive set-up I try to create based upon it.  If I understand correctly, shifting impedance, reactance, and sensitivity will have much more of an effect when using a passive unit than with an active one?
[/quote]
Also, yes, I can use an amp in two channel with active X-over, but there will be times where I want to use a more compact rig and use just one side of an amp to drive this entire cabinet.
Logged
Pages: [1] 2 3 ... 8   Go Up
 


Page created in 0.066 seconds with 17 queries.