Frequency Sweep Range for Aligning Speakers

[crossrh]

I'm using the Alignment Tool in the All SPL control panel to try to fix the phase issues that Dirac left me with. (My L,C, and R, all Zaph Audio designs, have 2nd order crossovers, and thus reverse-polarity drivers, and this is throwing Dirac for a loop. They're aware of it, but haven't offered any solutions yet.)

Is there a preferred frequency range I should use during the sweeps to get the most accurate info? Is the range different for full-range speakers vs. subs? Is there anything else I should know about this process? I'm using the farthest-distance speaker (SBL) as the Acoustic Timing Reference.

On a side note, my Land R speakers are sealed, but my C speaker is ported, and my Subs are IB. Would this make it harder to match phases, or reduce the frequency range that I could hope to get a good phase match? I can rebuild the Center speaker into a sealed design, but I'd like to know if it's worth it, from a phase-matching perspective.

Rick

 

[jtalden]

I suggest you use the alignment tool to adjust the SW delay to achieve good SPL support with the FL and FR. The CC can then be changed in distance to either match the timing in the midrange or in the SW XO range; whichever you prefer. I personally think it may be best to align the phase in the midrange, but have not seen discussions on this topic.

Sweeps should be full range for all measurements; SWs and mains. Be sure the XO is active and you are measuring the redirected bass. That is, use the FL or FR channel for the SWs measurement, but disconnect that main.

In an ideal anechoic world where we want ideal timing across all speaker we would need identical speakers. In the practical world the room is a major contributor to the timing we select so the speaker design is not a major factor for the SW XO.

A woofer with an inverted polarity is no issue for the alignment tool process. We normally select the best phase tracking through the XO range without prejudice as to what expectations are. The room and acoustic rolloff rates impact what works best in practice.

I would expect Dirac would have no trouble finding a favorable timing when the woofer is wired with negative polarity. That is a very common design choice. I expect it is more likely that some room situations are the cause of some automated setup systems having trouble choosing a favorable timing. At least that is the reason I often have to spend a lot of time in analysis doing it manually.

 

[DanDan]

Dirac, Sonarworks, and I presume most DRC software will create in interchannel delay if the central mic is not precisely central. I have never achieved a 0 time offset in Dirac.
Oddly, and hard to understand why, but this small delay absolutely wrecks stereo in a critical listening environment.
Standalone Dirac Live has a bypass, but the Minidsp version does not have bypass nor does it allow viewing the delay, if there is one.
Surround systems seem to use such delays a lot, but I am certain that it causes very significant problems with normal stereo program.

 

[crossrh]

Be sure the XO is active and you are measuring the redirected bass. That is, use the FL or FR channel for the SWs measurement, but disconnect that main.

I tried doing that, and got some really weird distance/delay readings. Like, 148 feet. I ran the sweep a few more times, but the distance/delay never got below eight feet, which is way off.

I'm also having problems getting consistent results from REW. The FR plots are consistent, but the time/distance can be all over the place. I've attached some pics of my room. It's fairly symmetrical. The smaller, brown speakers are John Krutke's SR-71's (zaphaudio.com) that I use for two-channel listening. The IB subs are powered off the Monolith 7X, along with the L, C, and R. So no additional processing delay that one would see with a powered sub. I did end up rebuilding the CC to make it a sealed alignment, but the phase plot still comes up way different than the mains, which are very consistent between the two of them.

I guess my biggest problem is getting some consistency with REW. I've been using it for years, now, and have never had this problem. The rest of my equipment is listed under "My AV System."

 

[jtalden]

It appears you rely on Dirac for all settings. It usually does a good job. I don't see a way with your equipment list to manually readjust delay timing even if it is unfavorable. We still can use REW to see how well Dirac did with the timing and EQ however.

Are you hearing the timing chirp from the SBL on all measurements? Can you post an mdat with an example of the acoustic timing repeatability issue you are having? Maybe just repeat measurements using one speaker several times. All the repeated measurement impulses should overlay each other when acoustic timing is working properly,

 

[crossrh]

Thanks for the reply, John. On the HTP-1 I can make adjustments to the Dirac delays and gains for each speaker. It sets the farthest speaker (SBL) at 0ms, and then positive values for all the others. I am not, however, able to go to negative delays. To do that I would have to add, say, 5ms to all the speakers and then subtract from the speaker that REW says needs it.

I am hearing the timing chirp from the SBL, the farthest speaker. I think I found at least part of the problem with some of my measurements. My theater is over my garage, which has a small, window-mounted AC unit. Turning that unit off, as well as the AC for the theater, seams to have cleared thing up, at least a little.

I'm still working on getting a good phase alignment between the subs and the mains. All of them are sealed alignment (IB counts as sealed, right?) I'll take some more measurements this morning, before it gets too hot and the AC wants to run all the time.

I'm doing all of this without Dirac enabled. I'm entering the delay data into an unused Dirac slot on the HTP-1. My goal is mainly to get pretty close with the FR for the bed speakers, using the very robust PEQ on the HTP-1. (I can run a Dirac cal with or without PEQ enabled.) I wanted to get close to a good FR so that after I run Dirac, the changes I make with PEQ won't affect the phase relationship that Dirac has set up, especially around the crossover freq.

Rick

 

[jtalden]

Yes, IB is the same as sealed.
If delays/distances are adjustable after running Dirac then it is easy to check the timing with REW and adjust if needed to a more favorable setting.
I wouldn't expect PEQ or timing before Dirac to be particularly helpful, but it is not going to hurt anything except maybe the sound quality. It's fun to experiment to see if it helps.