YPAO RSC Re- measurement

[Dunkc]

About AU$1000

Maybe SB 2000 might be better and more compact? A few hundred AU$ more costly.

 

[lovinthehd]

I expected worse ! (that's roughly $690 US)

 

[Dunkc]

OK , here are REW test results for 6 separately measured channels. The LFE high cut dial was set to its maximum hz position. I applied the EMM-6 calibration file. I don’t think I needed to calibrate the sound card. Also attached a rough picture that shows the layout of the room. Moving the SW around in a metre radius did not make any difference.
In REW can I select FR +LFE for combined signal output and measurement?
In REW can I simulate a FR + LFE response using the curves from each speaker test and summing them?
Can somebody please interpret these results and provide some guidance - thanks

 

[William Lemmerhirt]

Hey, cool room!
So, here’s my quick n dirty on your sweeps.
First you want to change the vertical limit to 45-105db. That will give a 60db window(want this). Then remove smoothing. 1/6 is ok, but it can hide stuff when trying to evaluate. So...

The left red mark shows where your subwoofer starts to roll off(basically quits responding) which seems like about 38hz. This is very high, for a subwoofer and a more capable one should reach down to 20hz. The right red mark is where the subs higher response is is rolling off. It’s about 55hz which I can’t figure out. That makes sense if the XO is set at 50hz but that would be low for your bookshelf speakers. In any case I suspect a very boomy type of sound with all the bass energy so high between the 30hz and 60hz marks.
The blue marks are showing a 10db difference in output from where it starts dropping at55hz down to the 70hz range where it tapers back up to about the 150hz range. There is a lot of musical bass in that range that you are missing, and a lot of movie bass that you’re missing below about 40hz. IMO, a new sub and some time to place and integrate it will benefit a lot. As far as another Yamaha, it might balance out better, but the low end will Nadine the same. Some actually just hit me now as I’m writing this. The frequency range of these sweeps might start too high. What range did you use? Do your sweeps from 10-200hz lfe(or sub) only.

Here’s another doodle showing some differences between graphs. This one is a sweep of my room(eq’d). You can see that output is pretty good starting around 15hz and continues up to just under 100. This is what you hope to see! The yellow marks are roughly where your sweeps are and used just to illustrate missing content. Hope you don’t think I’m busting you up. Just trying to be informative. Now I need some coffee! Wakey wakey!

 

[Dunkc]

Thanks for that I will read it more thoroughly during the daytime hours!
Pretty sure the sweeps were 20-200, 20-1000, not sure why they look truncated at the lower end.
Just occurred to me that the AVR could have been applying YPAO RSC corrections causing the SW and mains to roll off early in the 60-90hz range.
I should redo the sweeps using a DIRECT/PURE or YPAO THROUGH setting on the AVR, or just reset.
The SW specs are 20-160hz (-10dB). Maybe I also need to try and test its output in isolation to see if it is still working over this range.
If this is what “boomy” looks like, maybe that is why I have to turn down the SW volume a notch or I get ear ache.

 

[Dunkc]

Sorry but I haven’t got the hang of getting the presentation correct. :-(
First plot
Here is a sweep of the SW isolated in the middle of a large room with mic near driver (green) I think it shows that the frequency range meets manufacturer’s spec roughly and that it can produce output in the 60-90hz range. I agree that its output volume is not very consistent and does not go much lower than 32dB
The top blue graph shows the sweep of the speaker back in its original place with mic near driver - similar to first measurement (I did not manage to get volume levels the same)
The bottom blue sweep show the output with the mic back at the MLP. Lots of response lost, so I guess this is room effects showing up.
Moved the SW to rear left corner behind seat about 200mm from each wall, port facing out, and 1m from the MLP.

Second plot
Showing response at new location (purple) vs old position mid room against wall (red) vs FR (YPAO Through, green) vs FR (YPAO Flat, blue). The corner position works much better and integrates better with the YPAO corrected front speaker (others were the same).

The last plot
Shows the SW+FL+FR (80hz crossover, YPAO Flat) averaged in REW

Looks like I am getting somewhere with understanding this SW !
Does it look like the gap between 60-90hz is now covered with small front speaker crossover set to 80hz?
Am I done or should I do more analysis? I will redo YPAO at three positions and maybe recheck REW sweeps..
Is a better SW going to work similarly in this location but give deeper bass?
If I can afford it one day I would probably consider SVS SW. What is recommended: PB or SB, 1000 or 2000 ?

 

[PENG]

Sorry but I haven’t got the hang of getting the presentation correct. :-(
First plot
Here is a sweep of the SW isolated in the middle of a large room with mic near driver (green) I think it shows that the frequency range meets manufacturer’s spec roughly and that it can produce output in the 60-90hz range. I agree that its output volume is not very consistent and does not go much lower than 32dB
The top blue graph shows the sweep of the speaker back in its original place with mic near driver - similar to first measurement (I did not manage to get volume levels the same)
The bottom blue sweep show the output with the mic back at the MLP. Lots of response lost, so I guess this is room effects showing up.
Moved the SW to rear left corner behind seat about 200mm from each wall, port facing out, and 1m from the MLP.
View attachment 29570
Second plot
Showing response at new location (purple) vs old position mid room against wall (red) vs FR (YPAO Through, green) vs FR (YPAO Flat, blue). The corner position works much better and integrates better with the YPAO corrected front speaker (others were the same).
View attachment 29572
The last plot
Shows the SW+FL+FR (80hz crossover, YPAO Flat) averaged in REW
View attachment 29573
Looks like I am getting somewhere with understanding this SW !
Does it look like the gap between 60-90hz is now covered with small front speaker crossover set to 80hz?
Am I done or should I do more analysis? I will redo YPAO at three positions and maybe recheck REW sweeps..
Is a better SW going to work similarly in this location but give deeper bass?
If I can afford it one day I would probably consider SVS SW. What is recommended: PB or SB, 1000 or 2000 ?

As suggested earlier, you should change the vertical axis scale to 45-105 dB, or at least 35 to 100, and smoothing to 1/24, or 1/48, even no smoothing is preferred, to make it easier to evaluate low frequency resonances.

I also have the following recommendations for your next REW plots:

1. Turn the volume up by 3 dB. I know your system cannot play very now, but if your seating distance is 10 feet or less, it should be able to get you about 70 dB average from the main mic position.

2. Set the XO for the Tannoys to 90 Hz, 100 Hz may even be better. It is best to plot one for 80, 90, and 100 Hz so you can see which one is best.

3. Do the sweeps for a) L+R, b) L+R+Subwoofer, in addition to what you have already done.

4. Repeat the step 2., but with YPAO turned off, just to see if YPAO is doing a good job on its own, or you may need some manual adjustments, such as adjusting the subwoofer's phase know bump up the levels etc.

And, don't forget to set the vertical axis (dB) to 35 to 95, horizontal axis (F) to 20-300 Hz; and 1/48 or no smoothing. As William explained before, 1/6 is fine for final comparison of results, but to see the room related issues, it is best not to hide anything by applying too much smoothing, not to the range below 300 Hz anyway.

 

[Dunkc]

Thanks for the further advice - much appreciated.

I found a deal on an SVS SB-2000. So will start again with your advice when it arrives.

I read that 1/3 averaging better represented the human’s sound perception?

Setting the axes on the Mac version does not fix them. It does not set the limits correctly and I have to use the scroll bars and zoom to get into the region suggested, and this is not very satisfactory. Also, there is no LFE + FR/FL output option available. Also, there is no proper user guide to using the Mac version of REW that I could find.

I find setting the system up for constant output on each channel tricky. Each speaker has a different efficiency so for a set volume on the dial the dB is not constant (IIRC). Not sure what frequency in the sweep the volume level needs to be set on. In REW, does level checking select the best frequency considering the sweep range automatically?