Home Theater Forum and Systems banner

1 - 20 of 23 Posts

·
Registered
Joined
·
20 Posts
Discussion Starter #1 (Edited)
I have made some measurements, and I think I have rather long decay times in my room, and some peaks and valleys in de frequency response.

I have measured with the subwoofer in a corner, and along a wall, there does not seem to be a big difference between the two placements.

I have also attached a waterfall, and a spectrogram that shows the long decay times in my opinion,

Then I added a house curve (hard knee) to REW, and I asked REW to create filters to match the target curve. As you can see there were 4 filters generated, and I have added the two filters with the largest adjustment in dB to MiniDSP.

You can see the two filters in the MiniDSP parametric equalizer. I have not tried these out yet, I only played with them on my computer.

I would like some feedback to my conclusions, and would like some advice how to best improve the sound in my room.

I suspect that the MiniDSP will not really help me correct the room response, and that I would need some bass dampening, is that correct?

Thank you for your time reading this.
 

Attachments

·
Premium Member
Joined
·
1,400 Posts
- You'll want to do both ( some Equalization & some Room-Treatment ) to achieve decent results .

- You're going to need to educate yourself on what treatment does ( & does not ) offer ( at a reasonable price point ) .

- For instance , one can expect to only modify the decay times ( below 100hz ) with the standard ( though not necessarily inexpensive ) products sold by GIK & others .
- It takes a very large budget to actually modify the amplitudes in the lower octaves ( to effectively smooth out the frequency response ) .

- That's why you will still want to apply some EQ ( after some careful room treatment choices ) .

- This website has an acoustics forum run by a GIK rep.
- ( I believe it best ) to read all the stickies ( at the top of the forum ) before posting there.
- I'd suggest asking him what products they ( GIK ) sell that will get your low frequency ( 20hz to 120hz ) decay times down to ( say ) a more reasonable 200 to 300 ms range .

:sn:
 

·
Registered
Joined
·
54 Posts
View attachment sub-out-of-corner-Suggestion.mdat

I have assume this is sub+mains and not sub alone?

You should add just the sub measurement and define the cut over frequency you are after.

I had a quick look at the file and has added some filters and changed the target level to give you an idea of what to target as a starting point.
 

·
Registered
Joined
·
20 Posts
Discussion Starter #4
View attachment 45934

I have assume this is sub+mains and not sub alone?

You should add just the sub measurement and define the cut over frequency you are after.

I had a quick look at the file and has added some filters and changed the target level to give you an idea of what to target as a starting point.
This was just the sub, I had Audessey on the AVR turned off with the crossover at 250Hz so the AVR would not do anything to the signal. And I had the amplifier for the front speakers turned off. The signal from REW was channel 1 through HDMI.

Thank you, your filters look a lot nicer than mine.
 

·
Registered
Joined
·
54 Posts
What is the frequency range of your main speakers? 250 seems quiet a high cross over?
 

·
Registered
Joined
·
20 Posts
Discussion Starter #6
What is the frequency range of your main speakers? 250 seems quiet a high cross over?
I only did that for the measurement, normally the crossover is 80 for the front speakers, and 100 for the surround speakers. The idea was that my AVR would leave the signal alone.
 

·
Premium Member
Joined
·
1,562 Posts
egraaf,
When comparing your results to others it is important to do so on an equal basis. I note that the average measurement level is about 87 dB and the peak SPL levels are >95 dB. Most of the charts and recommendations here are based on results using measurements levels nearer 75 dB average level.

Your higher SPL level makes the charted decay time appear to be somewhat worse than it actually is. The peaks have the same effect. Lower the measurement level and apply the EQ to get a much better idea of how your results compare to others.

This will help in the decisions regarding acoustic treatments.
 

·
Registered
Joined
·
20 Posts
Discussion Starter #8
egraaf,
When comparing your results to others it is important to do so on an equal basis. I note that the average measurement level is about 87 dB and the peak SPL levels are >95 dB. Most of the charts and recommendations here are based on results using measurements levels nearer 75 dB average level.

Your higher SPL level makes the charted decay time appear to be somewhat worse than it actually is. The peaks have the same effect. Lower the measurement level and apply the EQ to get a much better idea of how your results compare to others.

This will help in the decisions regarding acoustic treatments.
Thank you. I will try to do another measurement this evening (CET), and post results.
 

·
Registered
Joined
·
54 Posts
Is that with or without any EQ?
 

·
Registered
Joined
·
54 Posts
In an ideal world you would try and move the subwoofer to a different position to address the room mode at 55hz which is causing the big dip you are seeing.

I know that is not always possible, therefore understanding the crossover, try the attached file as a starting point.

Once you have the curve looking much better, than you can start to look at waterfalls and RT graphs.
 

Attachments

·
Registered
Joined
·
20 Posts
Discussion Starter #14
In an ideal world you would try and move the subwoofer to a different position to address the room mode at 55hz which is causing the big dip you are seeing.

I know that is not always possible, therefore understanding the crossover, try the attached file as a starting point.

Once you have the curve looking much better, than you can start to look at waterfalls and RT graphs.
I will see if I can move my subwoofer, but I think my wife might not agree.

You have some large negative gain adjustments in your filters (-15dB), would it not be better to move the target level up a little, and have smaller adjustments and maybe some positive gain filters?
 

·
Registered
Joined
·
54 Posts
I have the same issue about subwoofer placement, so totally understand.

The problem with boosting is you will be fighting the natural room modes and therefore can end up stressing the amp and not actually gaining any boost.

If you do a search you should be able to find a number of threads whereby boosting is covered.

If it helps here are the settings I use on my sub and the results are amazing, see here:

http://www.hometheatershack.com/forums/rew-forum/70657-hard-knee-house-curve-results.html

Freq DB Q
43 -7.0 14.0
64 -9.0 8.0
77 -14.7 4.0
90 -6.0 4.0
 

·
Registered
Joined
·
20 Posts
Discussion Starter #16
I have the same issue about subwoofer placement, so totally understand.

The problem with boosting is you will be fighting the natural room modes and therefore can end up stressing the amp and not actually gaining any boost.

If you do a search you should be able to find a number of threads whereby boosting is covered.

If it helps here are the settings I use on my sub and the results are amazing, see here:

http://www.hometheatershack.com/forums/rew-forum/70657-hard-knee-house-curve-results.html

Freq DB Q
43 -7.0 14.0
64 -9.0 8.0
77 -14.7 4.0
90 -6.0 4.0
In reading Wayne's article about the hard knee house curve he says:

For my latest re-equalizing attempt, I tried a new tact – raising the Target Level up between the peaks and valleys before equalizing.


So that's what I tried to do, I agree that it does not make sense to try to do something about room modes, but other than that I don't think it's bad to do a little positive gain filtering.

Does that make sense?
 

·
Registered
Joined
·
54 Posts
Wayne article is a good one, so yes agree with that.

My insight was given the big dip at 55hz, I moved the target lower to try and address that.

I think the best thing is to do what you feel looks right based on what you have read and then measure to see what happens when the EQ is applied. You may well find my suggestion still leaves a big dip at 55hz, therefore moving the target up and having less reduction may work better. :bigsmile:
 

·
Registered
Joined
·
20 Posts
Discussion Starter #18
I have been able to move the sub to a new location, and the graph looks better to me there. Green is the new location, and red is the old one. I moved the sub farther along the wall, away from the corner.
 

Attachments

·
Registered
Joined
·
54 Posts
That does seem to have made an improvement around the 55hz dip.

Looks like time for the miniDSP to be engaged?
 

·
Super Moderator
Joined
·
9,203 Posts

Ok, I have attached some files with lower SPL levels. The waterfall looks less scary now I think
Sure, reducing the signal level is a great way to make a waterfall graph look better, even if no improvements in low-freq decay times (aka “ringing”) have been instituted or realized. This thread may shed some light.

That said, parametric EQ could help bring decay times for the mode at ~42 Hz back in line with what you have at other frequencies.

The issue at ~32 Hz – anytime you see where a certain frequency seems to just “jut out” of the graph – i.e., shows virtually no decay – it usually means you have something in the room generating steady-state noise at that frequency. We more typically see this at around 60 Hz, as electrical noise from things like refrigerators, but 32 Hz is an approximate relative of 60 Hz. So make sure there is no furnace, washing machine, cars idling in the driveway, etc. going on when you take your measurements – at least as far as generating a pretty waterfall is concerned

Regards,
Wayne
 
1 - 20 of 23 Posts
Top