Print

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

[David Sturzenbecher]

Interesting that the HF boost stops abruptly at about 10kHz. Did they not want the top octave..?


The curves look much better now.

Chris

This is incredibly common.  Typically the boost needed to maintain flat response to 20k is crazy... so it’s best to simply not even try.  You would be surprised the number of boxes that claim to hit 20k, however the DSP curves show otherwise.

[Rob Spence]

This is incredibly common.  Typically the boost needed to maintain flat response to 20k is crazy... so it’s best to simply not even try.  You would be surprised the number of boxes that claim to hit 20k, however the DSP curves show otherwise.

Looking closely, it is the coherence that dropped at 10k, the magnitude plot continues on toward 16k but got masked by the coherence.

[Rob Spence]

Looking closely, it is the coherence that dropped at 10k, the magnitude plot continues on toward 16k but got masked by the coherence.

Spent some more time and I think I am done with this phase. Next I need to set up one sub with one top and use a measurement mic. I was thinking about 10' out from the speakers and the height about midway up the top cabinet?

Here are the screen shots for the HF and LF. The HF looks pretty good to me though I don't understand why the coherence goes crazy at about 10kHz. I could not get the LF to look better at this time but will work on it more when running the actual speakers and amps.

 

[Tim McCulloch]

Your delay could be off by a sample or 2, try changing the delay time +/- the smallest increment and see what happens with the phase trace.

That said, it probably doesn't matter.

[David Sturzenbecher]

Spent some more time and I think I am done with this phase. Next I need to set up one sub with one top and use a measurement mic. I was thinking about 10' out from the speakers and the height about midway up the top cabinet?

Here are the screen shots for the HF and LF. The HF looks pretty good to me though I don't understand why the coherence goes crazy at about 10kHz. I could not get the LF to look better at this time but will work on it more when running the actual speakers and amps.

Rob what are we looking at here?   These again look like acoustical measurements, but the naming in Smaart says Nav vs Venue, implying that that you are using the output of the Sabin as the reference trace, and the output of the Drive Rack as the measurement.  This would be a completely valid test, yet i would prefer to just overlay traces.

The amount of phase wraps i see on the LF looks excessive no matter if it is an acoustical measurement or electrical

When I copy DSP settings from one device to another....I honestly don't even worry about taking acoustical measurements anymore.   If the transfer functions are the same...It. Is. The. Same. (**ruling out dynamics - limiting - or any other changes in the signal path)

[Dave Garoutte]

Be careful of the ground reflection with hard surfaces.
If you can measure on a lawn, better.

[Rob Spence]

Rob what are we looking at here?   These again look like acoustical measurements, but the naming in Smaart says Nav vs Venue, implying that that you are using the output of the Sabin as the reference trace, and the output of the Drive Rack as the measurement.  This would be a completely valid test, yet i would prefer to just overlay traces.

The amount of phase wraps i see on the LF looks excessive no matter if it is an acoustical measurement or electrical

When I copy DSP settings from one device to another....I honestly don't even worry about taking acoustical measurements anymore.   If the transfer functions are the same...It. Is. The. Same. (**ruling out dynamics - limiting - or any other changes in the signal path)

I posted earlier with overlay traces. I captured the transfer function of the Sabine, one for HF and one for LF , then tweeked the crossover and EQ for each on the driverack till they matched as good as I could.

I was thinking a listening test was in order before my next festival so no surprises and while set up, might as well measure.

[Rob Spence]

So, yesterday I set up the amp rack and a QRX212 on top of a TH118 in the driveway. I don't have a lawn as my home is on a rocky hillside in the woods. Half the project was cables... first, hooking 50a tails in a panel to my distro, and then a 30a cable out to the amp rack. Part of the test was to verify the power changes I made as well as listen to the rig with new DSP.

In hind sight, I should have also listened with the old DSP as it is still in the rack and only 3 XLRs need moving... Oh well, I may try again next week.

The bottom end seemed weak to me which is why I need to calibrate against the old unit.

A couple of anomalies showed up in measurement. There was a consistent huge dip at 170hZ and a smaller one at around 500hZ. Neither is near a crossover point. I am thinking that the 170hZ dip contributed to my impression of a weak low end. I did test inverting the polarity of the sub but other than the expected reduction of low end, it didn't change anything.

I worried that perhaps there was an issue with the speaker, so I repeated the measurements with another cabinet. The results were close enough to rule out a driver problem. I have each speaker in my inventory labeled with an id so in the event of something sounding wonky at a gig, I can note the unit for a checkup back at the shop.

So, I have 2 screen shots to post. One with just the QRX sitting on the sub and a second with the sub enabled. In each case the confidence line for the HF section looks pretty good. The LF has those anomalies mentioned above.

Comments and/or suggestions?

Thank you all!!

[Chris Grimshaw]

I'd put £5 on floor bounce.
Change the mic or speaker height, and the frequency will move.

Chris

[Frank Koenig]

Another approach to getting a "reference delay" for sub-top alignment is to measure each in the extreme near field. The phase (not magnitude) obtained this way agrees with free field measurements quite well and does not require an anechoic environment. This is what I do in "living room" situations where there are modes galore and no chance of getting a pseudo-anechoic measurement at any frequency, least of all low ones.

See "Low-Frequency Loudspeaker Assessment by Nearfield Sound-Pressure Measurement", D. B. KEELE, JR., JAES APRIL 1974, VOLUME 22, NUMBER 3

On EQing the top octave I agree with David. You choose your battles. I usually look for the last usable peak as things drop off and make that the top frequency for EQ. The correction curve is flat above that. I can't hear anything up there anymore anyway but I do my (measured) best for the youngsters 

--Frank

[ProSoundWeb Community]