Dereck,
Speakers usually have a higher amount of distortion than amplifiers. It's funny that everyone quotes amplifier distortion when speakers introduce more distortion into the final sound. If good low distortion speakers were used, I could tell the difference in 1% increments in distortion with a 1Khz sine wave. I agree with music it becomes much more difficult.

For everyone else,
I should clarify that it takes very astute hearing to be able to identify 1% distortion introduced in the signal path. The difference between 1% and 5% distortion also deppends on frequency. A tweeter reproducing a signal with 5% distortion introduced will be much more pronounced than a subwoofer with 5% distortion. As a matter of fact, most subwoofers routinely distort between 5% and 20% but people can't hear it.

Where most people introduce distortion into their system is driving a component into clipping. It can range from the Head Unit output to the power amplifier output, and is a very powerful case for level setting.

For the average Joe, distortion specs are irrelivant. Most people drive their systems well into clipping before hearing anything "bad". At that point you've got 50% to 90% distortion that's comfortably masking the 0.01% of THD.

Diabolik,
For the money you'd be better off with a XXK 4150. You'd save by only having to run power and signal wire to one amp. And, the active cross-over built into the amp can do everything you'd need for a 2 way setup. 1 pair of channels can be configured in bandpass (for the mids) and one pair of channels can be configured for high-pass. It's a very powerful sounding amp, there is very little difference in output between 85 watts and 100 watts. (we are talking a fraction of 1db)

EDIT: I just saw your other post. Since you already own an XXK 2100, I agree an XXK 2050 makes sense for the tweeters. But personally I'd run another 2100, I'm a sucker for power.

Adam

[ April 21, 2004, 09:04 PM: Message edited by: PEI330Ci ]

 

Yes, I got a great deal on the 2100 XXK I think a 2050 will be plenty for tweeters myself... 70x2 should be great. Unless I can get another sweet deal on a 2100 XXK.

-D

 

Dereck, I think you've illustrated the point about THD very well. I should get that disk and tell you what I hear as well. I think it's a very good idea for everyone to be aware of this.

Single ended tube amps have TONS of distortion, and they sound pretty darn good to my ears. Maybe there is more to clear sound than distortion #s. Eh? [img]smile.gif[/img]

My Audison VRx amps have a level limiting circuit that limits output when 2% distortion at higher frequencies is present and 5% at lower (read below 100hz) frequencies. It compares the input stage signal to the output stage signal and limits the output when the above limits are reached. I've had the pleasure of illuminating the limit lights on both of my amps. For the sub amp, I could not tell that there was 5% distortion. For the higher frequency amp, I could tell there was distortion, but it was at such a high level that I'm unsure if it was from the speakers or the amp. I'd have to hook the amp up to a monster P.A. speaker to test that one I think.

Thanks for zoning in on the subject Dereck, I'll have to do some more listening.

Adam

Edit: I mispelled your name Dereck, but I corrected it if you read before.

[ April 22, 2004, 09:22 AM: Message edited by: PEI330Ci ]

 

Adam,

I was reading the bi-amplification white paper listed on your other thread...

It seems to indicate that most active crossovers are not phase-coherent, where in my case, the passive crossover is designed to be as such.

In your case, I assume you are running an all active system. What type of active crossovers are you running and are they phase coherent? If not, IYHO, do you believe that to be a critical feature of active crossovers?

 

Phase coherence in electronics is hard to pin down. The general reason is that aside from digital filters, phase will shift slightly across the bandwidth of the circuit used. Many people don't realize that active X.O.s in general shift phase a bit, and that is good that you picked up on this. But, as I'm finding out in my current installation, phase coherance is a mythical proposition.

You can get close to having good phase response from your electronics, but most speaker locations wreck this. My point is that you can get equipment that theoretically can deliver what is needed for good sound, but the "in car" response will be different. This is why having as much flexibility as you can in your electronics is benificial in a car.

To my ears, active sounds better than passive. I'm using and Audio Control DQX, which is an all digital cross-over/ EQ.

I feel that working on the quality of the speaker mounting, location, and direction is the most important part of system tuning. Digital time alignment is at the bottom of my list right now, because it will be the last thing that I tune in the car.

At one point, I poorly mounted my current drivers, then EQed and time aligned the heck out of them and it sounded like crap.

On a final note, the best music that I've heard in my car to date was through a pair of 4" midranges. These were running full range in the kick area with NO EQ, and NO X.O.

Adam

 

..."On a final note, the best music that I've heard in my car to date was through a pair of 4" midranges. These were running full range in the kick area with NO EQ, and NO X.O.

Adam "

Is this because the drivers own phase shift
is directly related to driver roll off?
I wonder what difference that makes to your ears.

I was messing around with the xover point and slope on my mids. The setting I have stuck with for the time being is -6db @ 200hz. I wonder if the above has anything to do with that?

Comments?

 

Adam,

Very interesting insights. This thread has generated excellent conversation...

 

Seahag, phase shift in a driver's response is static. The driver's phase response does not change regardless of the input. At some point, I'll dig up some driver phase response graphs and go over the subject more in depth. But, different drivers have different phase response. It is however, not varied enought to be audible to the average listener. The difference in 2 different drivers phase response at let's say 3Khz may only be a few degrees.

Electronics, are a whole other topic.


The electrical signal feed through different slope analogue active X.O.s WILL vary. That is why playing with X.O. slope affects both time and frequency domain. The fact of the matter, is that applying a high pass filter in the analogue domain requires a capacitor. There is just no way around it. (aside from stuffing ports with a sock) Weather this capacitor comes before or after the amplification stage is what defines an active or passive X.O. arrangement.

Capacitors shift phase. The more capacitors in the signal path, the greater the phase shift. So going from a 6db/oct X.O. to an 18db/oct X.O. will offer a greater degree of phase shift. The amount of phase shift that a capacitor offeres is not linear as well. It varies as frequency varies.

For all those that I've confused the heck out off:

Use your ears to tune. Playing with filter slopes and frequency centers affects both phase and frequency response. For you rich buggers, get Clio from Orca design. It measures and plots in the time domain for under a $1000 US.


And...Cheers Dennis,

Adam

 

Drivers are a "minimum" phase device.

Here is an intresting write up, I think it somewhat agrees to what I was trying to point out.

http://www.geocities.com/kreskovs/Phase-B.html

*download* and read in Word

[ April 24, 2004, 06:03 AM: Message edited by: Seahag ]