Home Theater Forum and Systems banner

1 - 20 of 22 Posts

·
Registered
Joined
·
35 Posts
Discussion Starter #1 (Edited)
Hi,

Please skip "Background" if you're in a hurry :)

Your assignment
A subjective comment on the "flatness" of my 75 dB target frequency curve! It's a FL+FR+Sub average based on three measurements.


There is a horrible dip at 100-200 Hz and a smaller one at 30 Hz.

Background
I've been interested in HT for a while now, but my current apartment will only permit a 2.1 setup (no room for a 5.1). Until now I've relied Audyssey to perform room correction, but a good colleague of mine showed me the ropes of REW so I've started a long and hard journey to learn what actually is happening - I have been surprised! For starters, the crossover setting on my Onkyo does not perform as expected when meauring with REW, more on that later, first, the equipment!

Equipment


Connections


REW Setup
I'm using a UMIK-1 mic with my MiniDSP 2x4, which has been used to equalize the subwoofer. Presently I can only manipulate the FL+FR+Sub blending by adjusting the crossover on my amplifier. I've applied a 24 dB slope to my subwoofer starting from 120 Hz.



Measurements
There are four different screenshots I'd like to show you. My target is a 75 dB "flat" frequency response. The range on all the charts are 10 - 300 Hz and 45 - 105 dB. All of the measurements shown are averages based on three listening positions in my sofa (center, left, right).

Here is a description of each chart:

NB: Disregard the 'FR' written in the charts and REW data it's actually 'FL+FR'

1) FL+ FR + Sub (crossover @ 80 Hz, 100 Hz, 150 Hz, and 200 Hz)


2) 80 Hz crossover on fronts (FL+FR) adjusted on the amplifier (subwoofer solo, FR solo, FR+Sub)


3) 120 Hz crossover on fronts (FL+FR) adjusted on the amplifier (subwoofer solo, FR solo, FR+Sub)


4) 200 Hz crossover on fronts (FL+FR) adjusted on the amplifer(subwoofer solo, FR solo, FR+Sub)


Conclusion
I've opted for the 200 Hz crossover setting on my amp, which I wouldn't in my wildest dreams would have imagined would provide the "flattest" respsonse. Usually the recommendation I followed prior to using REW was to calibrate using Audyssey, set the fronts to small, then set the crossover to 10-20 Hz above the rated minimum FR +/- 3 dB of the fronts... This would have given a pretty bad result.

I thought the crossover setting on my Onkyo TX-606 would result in a rolloff at differnt Hz, and not just a steeper curve?

The entire .mdat file is available here.

*Update*
Thanks to the input from jtalden, I've done some tweaks that has alleviated the dip between 100-200 Hz. Here is what I did:

1) Added minor toe-in to my speakers and reran Audyssey (Audyssey put my Sub 10-20 dB HOT compared to the mains so had to turn that down)
2) Adjusted the distance setting for my sub in my receiver to the fronts until I got the best curve
3) Measured FL+Sub and adjusted FL distance until I got the best curve
4) Measured FR+Sub and adjusted FR distance until I got the best curve
5) Ran FL+FR+Sub with 3 point measurement and voila!

Result of tweaking distance settings. XO remains at 200 Hz, but I might be able to turn it down once I get my foam plugs:

FL+FR+Sub 200 Hz XO (10 - 300 Hz) - individually adjusted distance setting


FL+FR+Sub 200 Hz XO (10 Hz - 22 kHz) - individually adjusted distance setting


Comparison - before and After


Final question. Could manipulating the distance setting in the receiver cause the sound to be out of sync? I mean I got a nicer curve now, but is there some impact on synchronization?

Update November 4, 2015
I finally received my foam inserts for my B&W front speakers. They are two part and I've done measurements with half and full inserts. I've managed to lower the X-over from 200 Hz to 120 Hz and my FR is also a little better :)

Here is the comparison of the FL+FR (no sub) with and without the foam inserts:


Here is the final result before (200 Hz x-over) and after (120 Hz x-over and foam inserts) 10-22 KHz


Same as above but with 10-250 Hz range - it's more obvious where the improvements are here:

What do you think? :)
 

·
Premium Member
Joined
·
1,562 Posts
200Hz XO setting is pretty high. It may be the best compromise in a particular situation, but you should try to find a lower setting that will work. The timing/delays/distance setting between the SW and the mains impacts the SPL support in the XO range and of course the room modes can also be problematic there.

It is important that the proper delay timing was chosen for evaluation of each XO setting. If you did that and 200Hz was the only workable setting for reasonable EQ, I would suggest that you also consider moving the SW and/or the mains to a different location to see if that changes the situation. If that is not an option and lower settings aren't workable then 200Hz may be the best option.

just my thoughts in case it is helpful.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #3
jtalden,

Thanks for the quick response. I'm not really happy with the high cross-over either... I've left the distance measurement for the front/sub as per the Audyssey calibration. However, I did try to adjus the phase and measure it for both 0/180. Running it at 180 will cancel the overlapping waves so I left it at 0.

Could the altering the distance setting for the FL and FR enable me to run at a lower XO without runing my somewhat flat FR?
 

·
Premium Member
Joined
·
1,562 Posts
I just looked at the data file. I didn't initially notice it.
The SW response is great I would not change the location. The mains response is the issue. There is just too much LF SPL there.

Can you move the mains closer, wider/narrower, or possibly higher/lower to see if the situation is improved?

You may be able to augment the receiver 80Hz XO to add a high-pass filter using the MiniDSP to try to get the acoustic slopes to better follow the LR-24 roll-off that is the standard target. Used that way it is only on the mains an not on the SW. There are lots of optional setting and experimentation is probably the best approach.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #5
Yeah, It's odd the 685s output so much low frequency SPL... Maybe it's because they're ported? I've ordered a set of foam plugs (lost the original ones) and will try plugging them in and re-measure.

The mains are located next to my TV and my sub is actually next to my couch... I'll try to see if I can move the speaker up!

Right now I'm using the sub-preout to connect to my DSP so I can't filter my mains :'( I thought I'd rely on Audyssey for that, but you're seeing the filtered responses for the mains already...

Thanks for all the help!
 

·
Premium Member
Joined
·
1,562 Posts
jtalden,

Thanks for the quick response. I'm not really happy with the high cross-over either... I've left the distance measurement for the front/sub as per the Audyssey calibration. However, I did try to adjus the phase and measure it for both 0/180. Running it at 180 will cancel the overlapping waves so I left it at 0.

Could the altering the distance setting for the FL and FR enable me to run at a lower XO without runing my somewhat flat FR?
I would not look at phase or timing/distance settings at this point. Just leave phase at zero and distances where they are.

If you don't want to move the mains. I would try to set a 80Hz XO in the receiver. Then add an 80Hz B-24 HPF (high-pass filter) on the mains in the MiniDSP and see what the mains roll-off looks like. The frequency and slope of the HPF filter can then be changed to other values to until you get close to the acoustic shape of an LR-24 XO target. The final MiniDSP settings for slope and frequency are not important. Just find settings that work. If that doesn't work out then change the receiver XO freq up or down and try again with new setting the MiniDSP to see that works better. You can also augment the Mains roll-off issue using EQ filters in the MiniDSP if necessary.

Once you have more reasonable acoustic XO slope slopes, then the timing/distance can be set such that the SPL through the XO is supported when both SW and mains are measured together. So for now I would just concentrate on getting the SW and Mains roll-offs to approach the target shapes. Don't worry about the combined measurement until that is done.
 

·
Premium Member
Joined
·
1,562 Posts
Yeah, It's odd the 685s output so much low frequency SPL... Maybe it's because they're ported? I've ordered a set of foam plugs (lost the original ones) and will try plugging them in and re-measure.

The mains are located next to my TV and my sub is actually next to my couch... I'll try to see if I can move the speaker up!

Right now I'm using the sub-preout to connect to my DSP so I can't filter my mains :'( I thought I'd rely on Audyssey for that, but you're seeing the filtered responses for the mains already...

Thanks for all the help!
Are you sure Audyssey dynamic EQ is off? It can add a lot of unwanted boost to the low end. If it is on that could change the whole situation. It may be responsible for the low freq boost in the mains.

I suggest all measurements be done using left and right channels in stereo mode with the receiver XO active. This way the confusing influence of the LFE channel is avoided. To measure the SW just unplug the mains. We need to see the shape of the SW roll-off using the redirected mains signal not using LFE channel. Otherwise we can't see the shape of both 'acoustic' XO filters. You have already shown the LFE channel is very flat.

With Audyssey in the picture I am not as comfortable with all this work. You will probably be inclined to rerun Audyssey after the XO problem is resolved. That may or may not work out too well. If we can get it resolved with the current Audyssey settings then we could leave Audyssey it as is. The other option is turn Audyssey off and get good XO settings and hope Audyssey doesn't make too big a mess again when it is rerun.

My inclination is to leave the current Audyssey setting active (with dynamic EQ set to off) and not run Audyssey setup again after the work we do.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #8
Hi, all the measurements were done with Audyssey EQ on, but with Dynamic EQ off (incl. M. Optimization).

I'd like to have a flat FR and then use Dynamic EQ to my own taste if I need a bit more low-end boost :)

The measurements are performed for front speakers only (no subwoofer), subwoofer alone and FL+FR+Sub. There are measurements for the mains with full-band and XO settings dialed in on the receiver, but the sub if OFF for all these measurements.

I think based on you're input I'll have to find a way to EQ my mains manually (incl. cross-over) it's too imprecise to use the Onkyo XO settings...

Big OOPS - I might have written 'FR' in REW, but ALL measurements are done in stereo FL+FR...

Sorry for this confusion!
 

·
Premium Member
Joined
·
1,562 Posts
Also...
For this XO work there is no need to average measurements across the LP. Just take 2 sweep measurements; 1) L+R mains, 2) SW. The mic should be centered within maybe ±2cm.

Averaging can be helpful when evaluating EQ.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #10
Hi jtalden,

Do I need to change the LP filter on my sub or are you refering to the possible HP filter on my mains?

I've added a screenshot of the low-pass filter I've added to my sub using the MiniDSP 2x4 (see above).

Oh, and I've thought about adding some toe-in to my mains. The FR is located near a sidewall and I've heard it may help reduce reflections...
 

·
Premium Member
Joined
·
1,562 Posts
Hi, all the measurements were done with Audyssey EQ on, but with Dynamic EQ off (incl. M. Optimization).
That's good.

I'd like to have a flat FR and then use Dynamic EQ to my own taste if I need a bit more low-end boost :)
Nothing wrong with that. :)

The measurements are performed for front speakers only (no subwoofer), subwoofer alone and FL+FR+Sub. There are measurements for the mains with full-band and XO settings dialed in on the receiver, but the sub if OFF for all these measurements.
Yes, I understood. Those aren't the best set of measurement to resolve the problem efficiency. For this problem just the 2 sweeps mentioned above are needed for the XO setting you want to try probably ~80Hz.

I think based on you're input I'll have to find a way to EQ my mains manually (incl. cross-over) it's too imprecise to use the Onkyo XO settings...
The option I mentioned is to put the MiniDSP in the left/right output and not the SW. The current Audyssey EQ on the SW is as good as you will get. It is also probably okay for the midrange and high frequency. Only the mains near the XO have a problem.

So you could leave Audyssey on and just correct a 80Hz XO with the help of the Mini DSP on the mains. We use the MiniDSP to get control of the mains roll-off so it matched the classic target. It should mirror the SW roll-off around the XO frequency. Once the SPL roll-off is correct we set the delay/distance properly and we are all set. With MiniDSP help, the XO should then look pretty good.

Audyssey can then be left is. You could use some the Mini DSP EQ filters if you want at higher frequencies, but that may not be necessary with Audyssey having EQ'ed there already. A little touch-up of the HF roll-off can be helpful in some cases though if it too bright, or too dull.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #12
I really appreciate all the feedback!

Unfortunately I have the "cheap" Audyssey 2EQ, so all the filtering I've put on the sub has been done using REW and the miniDSP... The Audyssey 2EQ only performs level correction for the .1 channel :(

If I need to follow your suggestion I'll have to invest in another miniDSP in order to EQ the mains.

Let me think of a way to solve this problem. I'm using Sonos (SPDIF), TV (Optical), PS3 (HDMI) all using my PS3 so investing in another 2x4 or MiniDSP Digital may only provide EQ'ing for some of my inputs.

Did you have a look at my front full-band measurement, it looks terrible in the low end. Would you suggest a 80 Hz crossover and HP filter for these based on a 12 dB or 24 dB slope?
 

·
Premium Member
Joined
·
1,562 Posts
Oops - sorry for the confusion. I often have problems imagining all the system constraints. With a AV receiver there is no way to put the MiniDSP into the mains is there. I am used to thinking of a pre-pro and separate P-Amps. That kills my idea.

I will see if I can think of another path. Maybe others here will have good solution.
 

·
Premium Member
Joined
·
1,562 Posts
To your initial question the 200Hz XO setup you have above looks reasonably good for SPL as is. If there is no issue with being distracted by localization issues with the high XO then all is pretty good. I was just carried away with a mistaken possibility to lower the XO. :duh:
 

·
Registered
Joined
·
35 Posts
Discussion Starter #16
Hi again. Finally managed to fix the pictures on the first post. Had some issues with dropbox and linking directly.

With that sorted, hopefully it'll make much more sense!

Based on my LW 24 dB roll-off at 120 Hz, do you think I'd be able to suffer from localization issues?

Any frequencies below 200 Hz or from the discrete .1 LFE channel *should* follow the SUB chart?
 

·
Registered
Joined
·
35 Posts
Discussion Starter #17
jtalden, I tweaked the distance settings as per your instructions and managed to remove the dip between 100-200 Hz! I'm not sure if the procedure I used was the right one, but I updated it in the first post. Could you have a look and let me know :)? Thanks a ton for all the help.

I'm pretty satisfied with the latest measurement based on the fact I don't have any bass traps of the sorts. Obviously my waterfall chart might be horrible, but at least my FR is somewhat "ok" :D
 

·
Premium Member
Joined
·
1,562 Posts
Yes, you found a setting that results in a smooth SPL trace through the bass range. I would expect that to sound very good. The LFE will follow the Sub chart so that is great.

I don't know if the SW location is likely to be noticed. With test signals, I would expect you can locate it. With program material and without specifically trying to locate it may not be a distraction at all. Only you can say if it distracts you.

If you intend to do more testing when the foam plugs arrive:
> You can also just fill the ports with something a little denser for a trial to see the affect in case the foam is not enough to tame the low end.
> I would always test using 3 measurements; mains, SW, Mains+SW. This allows you to easily see the XO range interaction.

You have good result as is so more experimentation is more a matter of whether you enjoy the activity of fine tuning the system, or prefer to sit and enjoy program material. :)
 

·
Registered
Joined
·
35 Posts
Discussion Starter #19
You're on to something about the sub, it's not an ideal location so close to the listening postioning, but that's what I have to work with. So far I I'm not suffering from any localization issues, but, on certain songs I do feel it's more obvious than others. It might be because I'm expecting it :D

I'm very happy with the results so far, and I've certainly come further than I thought possible with the limited variables I have to work with.

I'll certainly do some more fine-tuning once I get the plugs so I can (hopefully) lower the XO to somewhere closer to 60-80 Hz.

For now I'll enjoy it!

Thanks for all the help :)
 

·
Registered
Joined
·
35 Posts
Discussion Starter #20
Hi, I got my foam inserts for my fronts and successfully lowered the x-over setting on my receiver.

Please see the updated charts in the first post :)
 
1 - 20 of 22 Posts
Top