RossoDiamante Theater REW Measurements - Page 5 - Home Theater Forum and Systems - HomeTheaterShack.com

Old 10-20-13, 10:04 PM Thread Starter
Senior Shackster

Join Date: Aug 2013
Location: United States
Posts: 101
Re: RossoDiamante Theater REW Measurements

Part 2
Attached Files impulse2-2.mdat (9.98 MB)
RossoDiamante is offline

Old 10-21-13, 01:38 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: RossoDiamante Theater REW Measurements

Quote:
RossoDiamante wrote: View Post
I'm certain I understand at least 2% of this!

Starting from the end, as it were, there are two adjustments that are suggested (SWs +3 ft and CC +2 ft.).
Sorry, I got confused the CC distance should have been -2 ft not +2 ft. I edited the above post to correct this.

Quote:
Inputting a greater distance in the AV processor presumably serves to alter the timing of the beginning of a given waveform relative to some sort of reference. Am I assuming that putting in a distance of "0" for every channel would result in no "artificial" delay introduced into the signal?
Yes, all at 0 would add no extra delays. The same is true for any distances that are all equal such as all at 10 ft. The delays are added to speakers that are closer in distance than the furthest speaker. This makes the signal arrive at the LP as the same time.

Quote:
In that case every channel that I do put some value in that field will result in an amount of "artificial" delay proportional to the value input.
No, see above. The furthest speaker has no delay the closer speakers have just enough delay to compensate for difference in distance.

Quote:
If I recall, the distances that were input were SW = 14', FL = 14', FC = 13.5', FR = 14'.
So there would be no delay to any of these channels except the CC (FC). The CC would be delayed about 0.5 ms to compensate for the -0.5 ft difference.

Quote:
If I'm understanding your analysis correctly, the delay that is currently introduced appears to be 28 ms. Your analysis is also showing that a delay of 31 ms is more ideal, necessitating an increase in the value input into the AV processor of 3 ft. For the suggested SW adjustment of + 3 ft, we are in effect asking the AV processor to delay sending the SW signal by 3 more ms to get its signal to more closely align in timing to the signal coming from the FL and FR speakers.
Since we carefully entered the actual distance measurements into the AVR we can assume all the signals are arriving close to the same time except for the SW which has a little additional delay. We would expected it be delayed an additional 1-2 ms. Remember that we don't know the actual delays because we don't have loopback timing capability and REW moved the original positions of the IRs to align the largest peak near 0 ms. The peaks are not the initial arrival time. The initial rise of the IR is the initial arrival time. That is why I manually realigned them to how they would have aligned if loopback was in place. To do this I moved the SWs IR -30 MS in REW (which is actually 30 ms of additional apparent delay). It is not real delay it is just correcting for the SW IR shift that REW applied. This all gets confusing fast as:
> A manual IR shift of -30 ms in REW is increasing the delay of that channel relative to the other channels.
> If an increase in delay is needed in a channel and we intend to do that with a distance change in the AVR we need to reduce the distance setting. A +1 ms delay is obtained by reducing the distance in the AVR by about -13.5 in. or about -1.1 ft; roughly -1 ft for each +1 ms of delay needed.

Quote:
Two questions relating to this first part: First, the need for an increase delay in the SW signal seems to run counter to your assumption that the DSP/processing in the iNuke introduces its own delay. Second, would the same net result be obtained by decreasing the distance in the FR and FL fields by 3 feet each?
We found the IR alignment that we believe is the actual alignment very close to where we first estimated it to be. We guessed it would be near the -30 ms REW adjustment and a then found it was actually at an REW shift of -31 ms. We found this by matching the calculated SPL trace with the -31 ms shift to the actual SPL trace. Had we used loopback this is where we would have found the relative positions of the 2 IRs to start with on these measurements.

From that starting point we found that the REW SWs IR position needed to be shifted +3 ms to maximize the SPL. A +3 ms manual shift in REW is an -3 ms relative delay for the SW or approximately a +3 ft AVR adjustment.

I still get confused on the directions when doing this sometimes as I did on the CC above.

Quote:
For the Center, the suggested modification is to add 2 feet to the AV processor's distance. On a relative basis, I currently have the center marked as 0.5 ft. less than the FL and FR channels. This decreased distance would presumably serve to increase the delay the AV processor uses in that channel relative to the FL and FR signals. By putting in a greater distance for the center channel, it moves the relative measurement of the center channel to "behind" the FL and FR speakers (i.e. greater distance to the listening position) which would change the timing of the signal to the center from "after" the FL/FR to "before" the FL/FR. This doesn't make intuitive sense when there is presumably no additional electronic delay (i.e. iNuke) for the center relative to the FL/FR and the speaker is physically closer to the listening position. The only thing that I can think of is whether we are "measuring" and accounting for something like the length of speaker wire.
Ignoring the direction as I confused the direction in the previous post, you are correct in the basic idea that the CC distance would logically not need adjustment relative to the FL and FR if the original measured distances were all correct. In practice however most all the SPL support distances vary a little from measured distances due to differences in room influence. Note that, the CC is not at the same distance to all boundaries as the FL and FR are. My room setup is very asymmetrical and there is a significant difference in the SPL support between the FL and FR. I chose not to compensate for this and simply try to compensate with EQ setting. I want the direct signal to arrive at same time even if the XO SPL on one side is significantly weaker and requires more EQ. In a home theater environment the CC is primarily dialog and to have the dialog arriving 1-2 ms different only shifts the persons apparent distance by that much. For music I mentioned in the previous post that I would probably not make that shift and allow the direct signal to arrive at the same time. This can be handled either way; you choose.

Quote:
I'll make these changes in the AV processor and remeasure. Thank you very much for your thoughtful analysis.
I try to look at these tomorrow. We need to compare the SPL of the new setting to the old setting an see if the expected improvement is found. The comparison should be done with the mic at the same LP.

Quote:
One thing that troubles me to a certain extent (and I'm surprised it doesn't trouble anyone else) in looking at these graphs is what appears to be a significant roll off from about 5 kHz through 20 kHz. It is consistent between the FL/C/FR graphs so I'm assuming at has to do with the actual design of the speaker or its enclosure. All 3 speakers are identical models and are mounted in nearly identical boxes. Is this something that should be EQ'd to flat? As it is, the system sounds very "bright" to me already. But that may have something to do with the lack of any sort of acoustic treatments in the room (yet). Or is my perception of "bright" a function of a relative peak from 500 Hz - 5 kHz? Perhaps it something to do with "ringing" which seems prevalent throughout the upper frequencies in the waterfall plots I've made from these latest supplied measurements.
Are you going to manually EQ the main channels? PEQ? GEQ? or maybe FIR filters? I wouldn't worry about it until the room is completed as it will all change. The house curve is very much a matter of taste and will be impacted by the speaker directivity and the room acoustics. You may want to start with the X-curve or a downward sloping curve and then decide what changes you want to try. This is a very big subject in itself. Wayne Pflughaupt is good source for threads/discussion on this subject.

Quote:
Once this timing issue is optimized, what is the next suggested step in tuning?
A glass of wine.
jtalden is online now
Old 10-21-13, 05:26 AM Thread Starter
Senior Shackster

Join Date: Aug 2013
Location: United States
Posts: 101
Re: RossoDiamante Theater REW Measurements

Sounds like I'll have to make another measurement with the CC corrected to the -2 ft. value. Right now it should be even farther off as I made a +2 ft. adjustment in the AV processor.

But gotta go to work. Work always seems to get in the way of my fun! Thanks again for your detailed analysis and explanations.
RossoDiamante is offline
Old 10-21-13, 08:57 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: RossoDiamante Theater REW Measurements

Quote:
RossoDiamante wrote: View Post
Sounds like I'll have to make another measurement with the CC corrected to the -2 ft. value. Right now it should be even farther off as I made a +2 ft. adjustment in the AV processor.

But gotta go to work. Work always seems to get in the way of my fun! Thanks again for your detailed analysis and explanations.
Oops, I got so involved in the explanation of the process I was using that I forgot to advise you that you do not really need to understand that math process. I use that process because I cannot take a bunch of measurements on your system and I want to see the polarity of the SWs vs the Mains and find the distance settings that I believe is appropriate so that I don't misdirect you any more than necessary.

All you really need to do is to set the measured distances in the AVR and then:
> Measure the SPL
> Adjust the AVR SW distance 2 ft and take another measurement.
> Repeat until you have measurements covering ±8 ft for the FL+SWs
> Use the overlay SPL view to select the distance setting that provides the most SPL support in the XO area.
> Repeat for the FR+SWs and select the SW distance that best satisfies both FL and FR. They should be very similar particularly if the room setup is relatively symmetrical. Some compromise may be needed.
> Confirm that the CC+SWs is okay at the same SWs setting or adjust the CC distance a little if desired. If it is very different (like 6 ft) it may be better to invert the CC polarity and see if it is then closer. Not all speakers used for the CC are the same design as the FL and FR. If they are not specifically designed to match then the distance may be different. I don't know how common a problem this is, but mixing speaker designs can be problematic.

Notes:
> We expect to find the best setting with the SW distance at the starting distance or adjusted maybe +2 ft from it's starting distance. It is of course not necessary to really do all the measurements; just enough to satisfy ourselves that we found the best SWs distance setting.
> If we want to fine tune it closer, we can then move ±1 ft from that position to see if it improves a little more. 2 ft increments are adequate and there is definitely no need to go closer than 1 ft as that is a very small increment for an 80 Hz XO.
> If the distance shift is found more like 6 ft to 8 ft from the starting distance that suggests that the SWs are the reverse polarity of the midwoofers in the mains. In that case we can either inverse the polarity of the SWs and start over or just leave the setting where it was found. There is nothing really wrong with either choice. There are tradeoffs either way. Having the same relative polarity is considered the more conventional setting however. Your SWs distance was found where we expected it to be and suggests that the relative polarities of the SW to the mains is the same and also that we found the conventional setup.
> I do usually suggest that we should favor a good setting by increasing the SWs distance rather than reducing the SWs distance very far from the starting point. [The reason is beyond this scope.]
> It may be easier to drive both FL and FR together with the SWs to find the best SW distance and then confirm the CC measures okay at that setting. I don't do it that way, but I would expect the same result. I like to see what is going on channel by channel to confirm that there is no special issues with any channel; like one front being wired with the wrong polarity or a major dip at the XO of one verses the other due to room effects.

You can go through this entire process or just do the minimum to confirm to yourself the settings I suggested are actually improved compared to the starting settings. Remember it was only an educated guess on my end as to the starting point for the math because loopback alignment was not used. You also remember I sometimes get confused doing this process!

When satisfied with the fronts, you can check the surround distance setting in a similar way. There we would only adjust the distances of the mains not the SWs as the SWs must remain properly aligned with the fronts.

I'm sure this really cleared things up for you.
jtalden is online now
Old 10-21-13, 09:23 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: RossoDiamante Theater REW Measurements

Quote:
RossoDiamante wrote: View Post
Part 2
I just took a quick look at this new data - no surprises.

The FL+SWs and FR+SWs look exactly as we expected. A little improved from the initial settings.

The CC+SWs looks much worse and that makes sense since I had you move them the wrong way.
jtalden is online now
Old 10-21-13, 09:48 AM Thread Starter
Senior Shackster

Join Date: Aug 2013
Location: United States
Posts: 101
Re: RossoDiamante Theater REW Measurements

This probably means your initially guesses were on target. I will correct the CC distance and remeasure.

My understanding at this point is that the surrounds would be correlated with the SWs as above and that any corrections would be implemented only on the surround distances. No changes should be made to the subs or the FL/CC/FR at this point since they should be locked together as a unit now.

It is also my understanding that these corrections have nothing to do with the frequency response characteristics of the speakers or the "ringing" that needs to be dealt with on the waterfall graphs?
RossoDiamante is offline
Old 10-21-13, 11:50 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: RossoDiamante Theater REW Measurements

That is correct.

This only optimizes the timing of the handoff of the mains to the SWs in regard to SPL and phase alignment. It has little to do with other concerns. It is a small portion of the overall effort. Most people do not bother with it, but the risk is the possibility of low SPL through the XO range.
jtalden is online now
Old 10-21-13, 07:25 PM Thread Starter
Senior Shackster

Join Date: Aug 2013
Location: United States
Posts: 101
Re: RossoDiamante Theater REW Measurements

The final SWs and SWs+CC measurements with the corrected, corrected speaker distances.

I'll be able to do an audition of the speaker system tomorrow evening.

At this point, I think I will try to take a look at the surrounds next. I assume the measurement and analysis techniques are the same? Measure SWs alone and then SWs with each of the surrounds?

Then I think it's time to take a break and actually finish this room. These seem like good "baseline" measurements of my system in a raw, untreated room. To my ear, there is quite a bit of work to be done in controlling the top end which seems overly bright and reverberant to me.

Are there any measurements I should add to my "baseline" set before I start adding carpeting, and furniture, and acoustic treatments on the walls? What about an "all channels" test? Or just a surrounds (2 sides + 2 rears) test? What about just the 3 front mains? Would any of these be tested in the future in the final tuning phases such that a baseline measurement would be either beneficial or at least instructional?

What about adding a 3rd sub? I am thinking of adding a 3rd sub in the back of the room which is completely different in manufacturer, size, and configuration. It is an M&K sub from 20 years ago at least. I think it has a pair of 12" drivers in a push-pull configuration if I'm not mistaken. I will dig it out along with an extra 2 channel amplifier I have lying around. What would the measurement/testing process be to integrate a third sub into the system?
RossoDiamante is offline
Old 10-21-13, 07:29 PM Thread Starter
Senior Shackster

Join Date: Aug 2013
Location: United States
Posts: 101
Re: RossoDiamante Theater REW Measurements

Quote:
jtalden wrote: View Post
That is correct.

This only optimizes the timing of the handoff of the mains to the SWs in regard to SPL and phase alignment. It has little to do with other concerns. It is a small portion of the overall effort. Most people do not bother with it, but the risk is the possibility of low SPL through the XO range.
Is the drawback of low SPL in the XO range also correctable with EQ instead of timing? On the face of it, it seems to me that it should be possible, but that compared to correcting the timing like you've done, it is more like using a sledgehammer to get the square peg through the round hole instead of just picking up a round peg instead.
RossoDiamante is offline
Old 10-22-13, 10:13 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: RossoDiamante Theater REW Measurements

Quote:
RossoDiamante wrote: View Post
The final SWs and SWs+CC measurements with the corrected, corrected speaker distances.
I took a look at the new SWs+CC measurement and compared it to the calc SPL at this distance setting. It was pretty close but not quite as good as the calculated (possibly the mic was in a little different position). The SPL is only marginally better than your original measurement at the measured distance setting. There are 2 CC nulls in this XO area at about 80, 107 Hz, if I recall correctly. For the reason you mentioned (direct signal arrival time) and also because the room is not completed yet, it is probably best to use the measured distance setting. So that would be the corrected, corrected, corrected setting for the CC, i.e., back to where we started.

Quote:
At this point, I think I will try to take a look at the surrounds next. I assume the measurement and analysis techniques are the same? Measure SWs alone and then SWs with each of the surrounds?
All that is needed is to set the measured distances and confirm each is okay with the SWs.
So measure the SL+SWs and see if the SPL support in the XO range looks okay. If there is a very large dip try inverting the SL polarity as see if it is better. Then move to the next speaker.
[You could also change the SL distance setting and see if it is a significant enough improvement to make it worthwhile. Most Surrounds are not the identical designs to the FL and FR anyway so there will be compromise anyway. On the positive side it is also not as critical as the fronts so, unless there is a very large problem at the XO, I would leave them as measured.]

Quote:
Are there any measurements I should add to my "baseline" set before I start adding carpeting, and furniture, and acoustic treatments on the walls? What about an "all channels" test? Or just a surrounds (2 sides + 2 rears) test? What about just the 3 front mains? Would any of these be tested in the future in the final tuning phases such that a baseline measurement would be either beneficial or at least instructional?
Possibly the Acoustics Forum or others here have better advice on this matter. I have no real experience with before and after measurements of acoustical changes, but I would not think there is any real value beyond curiosity/learning unless you are using them to make design changes.

The SWs to mains timing setup you are doing now will not be changed unless the speakers or the LP are moved so that work was worthwhile.

Quote:
What about adding a 3rd sub? I am thinking of adding a 3rd sub in the back of the room which is completely different in manufacturer, size, and configuration. It is an M&K sub from 20 years ago at least. I think it has a pair of 12" drivers in a push-pull configuration if I'm not mistaken. I will dig it out along with an extra 2 channel amplifier I have lying around. What would the measurement/testing process be to integrate a third sub into the system?
Your current SWs SPL response is very smooth so adding another SW is not needed in that regard.

SPL capacity is the other consideration. This is dependent upon; capacity and freq range of the SWs, the size of the room, lost headroom due to EQ boosts, intended volume levels. I suppose it also depends upon the level of distortion you are sensitive to.

You can see SW capacity is a very personal determination.

If you do decide to add it, just go through the SW setup procedure again until they are working together well as single unit again. This may be easier or harder depending on the position of the new SW in the room and the particular design of the additional SW. In general, the more SWs in the room the easier it is to achieve a smooth SPL response without undo EQ and the more SPL headroom that be available. I would not expect it to help the lower freq limit significantly. Unless the new SW is equal or greater in capacity to the other two I would be wary about adding it.
jtalden is online now

 Bookmarks

 Tags measurements , rew , rossodiamante , theater

Message:
Options

## Register Now

Random Question
Random Question #2

User Name:
OR

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.