Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[scott foster]

 

Anyone have experience using the DCX 2496 Auto Align function?  I’m curious about how well it has worked for others and what it’s been tried on.  The manual is very sparse on this feature.  What was your procedure?  Does anyone happen to know what goes on in the box as it attempts to execute this task – what it is measuring?    

[Ivan Beaver]

Never used it, but it probably works as badly as other "auto" systems.

[Paul O'Brien]

scott foster wrote on Tue, 23 October 2007 13:50

Anyone have experience using the DCX 2496 Auto Align function?  I’m curious about how well it has worked for others and what it’s been tried on.  The manual is very sparse on this feature.  What was your procedure?  Does anyone happen to know what goes on in the box as it attempts to execute this task – what it is measuring?

I thought the manual gives a pretty good explanation.. all things considered. "The DCX calculates the run-time differences, phase shift and polarity. The OUTs are automatically adjusted to the calculated values."

This is not an EQ adjustment but time alignmet between outputs. It requires at least 2 outputs be active(not muted), otherwise there is nothing to compare. And like all AUTO functions, it is only valid for the exact test mic location at that time. It's best used to setup a baseline for the speaker system in an anechoic environment, or where the processor is used to drive delay stacks.

[Chris Semler]

I have tried it.

I have only done it once so I do not have a whole lot of experience with it.

One thing I was having problems with was that it would flip the polarity of a driver then delay it by a huge amount. So I believe it was making an error in the polarity flip and trying to compensate with the delay.

I would recommend determining polarity yourself and not allowing the dcx to guess at that. Look at the delays it gives you and ask yourself if it is reasonable. If not try a different microphone position. Or a different order of doing things. You may want to try to only align 2 drivers at a time in a 3 or 4 way system.


Like any measurement tool the results are only as good as the person reading them.

Never blindly accept results from anything automatic. Question them.

This may help to explain some things.
http://www.prosoundweb.com/install/tech_corner/ty2/time.shtm l

[Josh Billings]

If you are aligning subs the best way to do it would be to flip the polarity on your mids and play a sine wave pule at the frequency you cross over at (80 hz or whatever) then adjust delays for maximum cancelation and you're all set. Just flip polarity back inside the dcx2496.

Love the box btw

The autoalign didn't do a very good job for me

-Josh Billings

[Kevin Ballard]

Yes I have.

I tried outside with a pair of biamp wedges pointing skyward (at 60 deq angle). Mic on a boom about head hight on centre of the box (seemed logical at the time).

Boxes are homebrew and loaded with celestion drivers on an NL4 speakon.

The two boxes each run from a chanel of the DCX to Left and Right of a pair of appropriate power amps. It popped away for a few bursts of sound before selecting a short delay and applying it. Listening afterwards, there was a slight improvement in the sound but in blind testing with my assistant selecting delay/No delay to pink noise and tracks it was hard to discern, voice was easier? However once a few eq tweeks were introduced to flatten the monitor response(above 120Hz) it was overall a much better sound and I have very few problems with on stage feedback in small stages these days (Beta57a mics).

So it can work, I think I tried a number of times to replicate the results and basicly got the same each time except one time it reveresd polarity and dialed in lots of delay instead! So check the results and prove the logic Allways. Use Joshs' method to prove what you have or fine tune.

[scott foster]

Thanks for replies.  I noticed the same tendency of unit to flip polarity.  Another question.  Is there an advantage to selecting steep slopes (48 db/oct) in terms of minimizing the region of potential problems between drivers?

[Bennett Prescott]

scott foster wrote on Thu, 25 October 2007 12:08

Is there an advantage to selecting steep slopes (48 db/oct) in terms of minimizing the region of potential problems between drivers?

No, there is instead a huge disadvantage (at least with traditional filters... lay out the $6K for something like the Dolby Lake processor that will do linear phase 180dB/octave filters and it's a different story). The high slope filters will ring, which negatively impacts sound quality.

24dB/octave LR is a good generic choice.

[John Roberts {JR}]

I would be careful about all sweeping generalizations about speaker crossovers. Non speaker types tend to look at crossovers as idealized dividing networks, in practice there are huge driver transfer function, physical spacing, and power handling considerations that get factored into practical crossover designs. One of the best sounding crossovers I ever heard was also the ugliest looking curve on paper.

I am known to be a strong supporter of power boxes with crossovers designed by the same engineer that specified the drivers, designed the box, and tweaked it in the chamber.

Short of that, if you can get usable preset advice from the factory that has to be a good starting point. I also rank specific advice from the speaker factory over general advice from the DSP box maker.

Free general advice from some of the strong speaker types who hang around here is good and worth digesting but perhaps lower on that list, and I am not even a speaker guy so I'm down near the bottom.

YMMV

JR

PS: to answer your specific question, steepness or order of the cut off filter is a trade off between protective attenuation against out of bandpass signals and phase shift/amplitude response for combining in the transition region. More poles of roll off help one at the expense of the other. Correct answer is it depends.