REW room measurement analysis and fix help

[haygeebaby]

Hi guys

I've finally reached a point where it was time to measure the room and see what really was going on with the acoustics.

I've spent a few hours with REW sofware and a minidsp umik1, watched some youtube vids, and read some FAQs.

Here is a picture of the basic curve I get. I ran measurements in about 12 positions around the single listening positions and all are very similar.

 

Measurement db was calibrated to 75db for the room.

Room is 5.9m L x 4.4m W x 2.5m H.

Looks like we have a big dip around 40hz. Any tips on how to fix?

Would I be right in saying that everything else on the graph looks ok?

 

[andyr]

 @haygeebaby, I'd be very interested to see what people recommend you do to fix the 40Hz dip.  I'd suggest it's a room-mode issue - can you change your listening position by, say 500mm?

 

But, to me, you have the top end too 'hot'.  It would sound better if the response above 2kHz was at - or slightly below - the 75dB level ... not up to 80dB.

 

Andy

 

[frednork]

would be good to know which of the speakers you are running

[haygeebaby]

[haygeebaby]

Agree. Room mode.

[frednork]

Ok, might be worth redoing with the mic not on a big (resonating) box, (preferably a nice solid stand) and see if it improves. (might not though) , can you perceive  the dip?

[frednork]

rough simulation shows a dip at that frequency is possible depending on where everything is

 

[frednork]

this one looks a bit closer

 

 

[haygeebaby]

Thanks for the responses guys.

Yeah - I've done the test without that box - no chair - on the floor - raised higher than normal - over the to the left and right - etc.

That dip at 40 is always there.

I also looked at the room simulator software as well - a think 40hz dip is a characteristic of this room size.

Can I hear it ? Yeah - I have a few tracks that I know have some deep percussion bass that I've heard on other speakers in other rooms and setups.

 

I've moved the mic back and forward in steps of 50cm and moving forward slightly will reduce the dip by about 5db but the top end and other parts suffer.

 

How to tame the sound above 2khz? More absorption at the front?

 

I am thinking a sub at the rear of the room might be needed to sort out the dip. Was hoping someone had a similar case and could comment.

[frednork]

that would appear to be a reasonable solution

[Nigel]

First thoughts were the same as Andy's, hot treble (presumably you can lpad the tweeters down to 75dB or below) and a room mode which can never be fixed if you sit in it.  My system and others are aligned diagonally across the room, somewhat, with the listening position determined via REW, then the speakers placed isosceles from there.  You get more bass against the walls and in corners, but REW then tends to look like a rollercoaster.

 

[aechmea]

I have one sub at the back of the room.  Works for my room, so worth a try.

 

My problem was out-of-phase reflections (SBIR) rather than a standing wave (mode), but in both cases cancellation (and reinforcement, depending on distances and geometry) occurs.

[Al.M]

Try orienting speakers across the longest wall length (90 degrees to current) and seated against near the rear wall. Treat the rear wall behind seat with 200mm+ insulation to help cancel rear reflections to improve clarity if audible.

 

Cant see what’s in the rest of of the room, but fill it more randomly with useful furniture and objects rather than being sparse to naturally breakup parallel reflections vs having a tidy, sparse and but very echo-y room and then having to randomly place treatments. Add treatments after furniture is placed, assess it and treat further if needed. See what that does.

[davewantsmoore]

On 28/12/2019 at 7:17 PM, haygeebaby said:

Here is a picture of the basic curve I get. I ran measurements in about 12 positions around the single listening positions and all are very similar.

Did you capture the output of the L and R speaker playing at the same time?    Don't do that.

 

Put your mic in the listening position.... and take 2 sweeps.  L, and R, with 1/12 octave smoothing.

 

Averages are hard to interpret without knowing what data created them..... Posting the 12 individual plots all stacked on top of each other is probably more helpful.    (but again.... not of L and R playing together..... .Just of L.... or R

 

 

Also.... remove the chair, and put the mic on a microphone stand (so nothing is close to the mic).   It will be impacting your data quite significantly around 1, 2, 3khz.

 

Quote

Looks like we have a big dip around 40hz. Any tips on how to fix?

Move the listener and/or the speaker position.

Add more bass sources

... or build an incredibly big bass trap (not a practical suggestion)

 

Quote

Would I be right in saying that everything else on the graph looks ok?

It's really hard to say, below 500Hz sure.   I don't trust 1-3khz (ish) due to your chair.... and everything above 1khz could be 'anything' due to the averaging of multiple measurement.... depending on where you measured (how far away from the listening position did you go, etc.)

 

Quote

Put your mic in the listening position .... and take 2 sweeps.  L, and R, with 1/12 octave smoothing.

 

without the chair.... make something bodgy from broom handle and tape, if you have to..... but anything close to the mic is a nono if you are interested in > 1khz

Edited December 29, 2019 by davewantsmoore

[davewantsmoore]

22 hours ago, frednork said:

Ok, might be worth redoing with the mic not on a big (resonating) box, (preferably a nice solid stand) and see if it improves. 

No, that's not causing the 40Hz issue.

 

9 hours ago, Al.M said:

but fill it more randomly with useful furniture and objects rather than being sparse to naturally breakup parallel reflections vs having a tidy, sparse and but very echo-y room

You might be talking more generally ..... but just FWIW, this will do nothing about bass modes.   The sound waves go around furniture like it's invisible (the waves are too big / furniture is too small).

 

Although something big and soft like a large couch, does have quite a bit of absorption.

 

 

Sitting against a wall is about the worst thing you can do.....   It's good he has such a long listening room, to have space behind him.

 

My suggestion would be:

 

Sit further forward.

Move the speakers back (towards the front wall some if necessary)

Toe the speakers in more....    and if you end up moving closer to the speakers slightly, then that requires even more toe it.  (Make it so the speakers "direct sound" crosses at least 30cm in front of you.... or more.

 

 

This may make the SBIR "worse" (really, just "different") as the speakers are closer to the front wall(s) ..... but it will get the listening position out of the strong mode(s) where it is now.

 

 

As for the upper frequencies being too hot.

 

1) spend more time figuring out if your data is accurate, before you trust it too much

2) more toe-in will modify this

 

Yes, it will look strange. 

[Al.M]

There’s useful info in this new post I created with attached link 

 

 

[haygeebaby]

Thanks Dave - agree with you points.

I've taken a load of measurements today.

I'll take some more again tomorrow.

I've managed to avoid the 40hz null. Can't sit in the center width wise. So I moved the chair 50cm to the right.

The Harman Axial Standing Wave Calculator came in handy and was easily understood.

So far I've moved my listening spot forward slightly, to the right, and increased my ear height. All 3 help me avoid the standing waves.

But this means all my symmetrically placed acoustic panels are off compared to the new optimized seating position.

And I need a longer power cable for one of the woofer subs.

We've opened up a can of worms!

 

 

 

[Hytram]

Sorry if this takes the thread somewhere where it shouldn't.. 

 

Would waterfall graphs be helpful in this measurement scenario? 

 

Probably not for the 40hz null, but the overall tuning of the room? 

 

Still trying to understand all the REW measurement stuff. 

[almikel]

15 hours ago, haygeebaby said:

I've managed to avoid the 40hz null.

nice work on the null, but a shame you have to sit off centre - lovely room and setup BTW.

 

Have you tried moving your subs around instead? including off the floor?

 

8 hours ago, Hytram said:

Sorry if this takes the thread somewhere where it shouldn't.. 

 

Would waterfall graphs be helpful in this measurement scenario? 

 

Probably not for the 40hz null, but the overall tuning of the room? 

 

Still trying to understand all the REW measurement stuff. 

I find waterfalls useful, and the RT measurements

 

cheers

Mike

[almikel]

21 minutes ago, almikel said:

Have you tried moving your subs around instead?

in the room photo are the subs underneath the horns or not in the photo?

It's hard to tell from the photo what the "boxes" under the horns are...

 

subs don't need to be co-located with main speakers - subs get placed to create the smoothest bass at the listening position - I think it's likely with 2 subs and mucking with the placement of each you could fix that 40Hz null and go back to a central listening position...

 

cheers

Mike

[haygeebaby]

Cheers on the comments.

in this room size - width wise I’m going to have a 40hz null and then a 118hz null in the centre. I can’t defeat the 118hz. Too high for subs. 
You have to move to get out of those nulls. But maybe a pro who’s solved the same could comment.

the woofers on the horns are at the bottom. They are not subwoofers but they do go down to subwoofer frequencies. 
I’ve had a few subwoofers in the room. I was never able to get the results that I liked enough for me to justify spending the money.

moving the seat to the right didn’t cost money. I’ve been listening in the new position for a while now. Some low end is more audible now. Everything else seems similar to the original position. 
question will be if the changes are worth the gains. More listening needed to decide.

Edited December 30, 2019 by haygeebaby

[deafenears]

17 hours ago, haygeebaby said:

in this room size - width wise I’m going to have a 40hz null and then a 118hz null in the centre. I can’t defeat the 118hz. Too high for subs. 

Mid-bass module? HSU used to sell this, but looks like no more.

 

http://www.hsuresearch.com/products/mbm-12mk2.html

 

I'm sure there's other alternatives. Also maybe:

 

http://rythmikaudio.com/F8_specs.html

[almikel]

On 30/12/2019 at 5:55 PM, haygeebaby said:

moving the seat to the right didn’t cost money

fair enough

On 30/12/2019 at 5:55 PM, haygeebaby said:

in this room size - width wise I’m going to have a 40hz null ..... in the centre.

with only your current bass sources, agreed...and I'm assuming those bass sources run all the way up to the large horn, so they have to remain co-located with the horns.

 

But a sub (or subs) could be placed in the best spot to ameliorate the dip.

 

On 30/12/2019 at 5:55 PM, haygeebaby said:

But maybe a pro who’s solved the same could comment.

Try pinging @Paul Spencer - he's based in Melbourne, but he also provides remote room analysis services.

 

cheers

Mike

 

[haygeebaby]

Agree. Professional help might be next. Have been listening in the new position for two days now. It’s much better sounding in the sub bass region. 
when time permits I might move some more things around.

everyones comments have been helpful. 
many thanks guys and happy 2020

[Al.M]

According to the experts link I attached earlier it’s not possible to resolve the big nulls and modes, they are too severe and yours is a typical one. Changing seat position can resolve it mostly, massive amounts of treatment and EQ can only alleviate but not eliminate it if you still choose to sit where the problem is. And this is after he has had a go at it, not amateurs and semis guessing at it.

Edited January 2, 2020 by Al.M