48 minutes ago, crtexcnndrm99 said:

too loud

Should sound, over the speakers being tested Fz operating range, same magnitude to the ear. Should be about 20dB above the noise floor of the room, 75/80dB is a good level, reasonable and will be fairly loud. If the sweep tone seems to disappear or rise dramatically in magnitude, that's showing there could be some response issue at those frequencies.

3 hours ago, crtexcnndrm99 said:

does turning up the sweep noise to the recommended decibels sound far too loud?

It will sound loud.

Wear hearing protection or leave the room.... of not too concerned about noise floor and distortion, then measuring quiter can be ok for general response.... although it is always interesting to look at frequency response vs drive level (as this can be a big source of distortion)... but probably ok with big speaker/horns  

 

3 hours ago, crtexcnndrm99 said:

Anyhow, took multiple measurements with speakers moved forward/back and LP forward/back, to observe how it may influence the frequency response. 

If you are not already, test L and R separately.    Sweeping them together will show you things to do with how accurately the speakers are aligned, and you probably don't want to look at those errors all the time.    Comparing a L and R and L+R together is good information (how well do L and R sum) ... but for what you are doing, look at L and R seperately.

 

3 hours ago, crtexcnndrm99 said:

Can't seem to improve the 60hz null with the current space, however the 90-100hz dip was improved. The <30hz followed the improvements in sub-bass I perceived by listening after the ports were changed to a more optimal alignment. 

Consider tilting up the bass a little bit, or perhaps a lot depending on preference.

 

3 hours ago, crtexcnndrm99 said:

Edit: NB. crossover frequency  ~600Hz 

There is a dip at this frequency.    I would be wanting to understand is this an acoustic effect (diffraction, etc.) ... or if this your crossover not summing correctly.   If the former, then perhaps not worry... if the later, then it is problem.    Or at least, whichever one it is the solutions will be different to each other.

Interestingly, after the first feedback about the 600hz dip, I wondered if physical time alignment of the horn could be to blame. I just now set about moving the mid horn within the adjustable range in the current setup, to see whether a cancellation was the issue. It does appear to have much effect at all. 
 

Subsequently, measured again with the LP further forward again, due to it working well in earlier measurements. Interesting how that dip around the 350hz region changes. Overall, seems a bit flatter nearfield which makes sense I think.  
 

Room width is 4m however open on the rearward (not directly behind LP - diagonal setup) wall to other spaces, so harder to quantify. 
 

19 hours ago, crtexcnndrm99 said:

Interestingly, after the first feedback about the 600hz dip, I wondered if physical time alignment of the horn could be to blame. I just now set about moving the mid horn within the adjustable range in the current setup, to see whether a cancellation was the issue. It does appear to have much effect at all. 
 

Subsequently, measured again with the LP further forward again, due to it working well in earlier measurements. Interesting how that dip around the 350hz region changes. Overall, seems a bit flatter nearfield which makes sense I think.  

It isn't entirely clear to me what traces are what....   but you likely have at a very rough guess (super hard to tell, but at an educated guess)

 

floor at 350Hz

cabinet and/or horn mouth diffraction at 600

modes at 35 and 60

an incorrectly summing crossover 800 - 2.4khz

incorrect filtering on the horn +3khz

wrong bass filter and/or level < 150Hz

Last time I used REW was 14 years ago and I've recently got back into taking a few measurements (lot easier now with USB mic & laptop!).

Thought I would post a couple of measurements from my subs after moving them around a bit to ascertain the best positioning.

 

Some info:

Room is 5.2m W x 10.6m L with apex ceiling 2-2.9m H.

Rear wall consists of a staircase (down) and two doors either side leading to other rooms- so basically open.

I have an older pre/pro with only a single sub output. There is currently no DSP or eq involved.

Subs SVS PB13 Ultra.

MLP is 4.4m from front wall.

There is substantial bass trapping in front wall corners

"Nominal position" for the subs is 1.3m (1/4 & 3/4) along front wall facing into each corner.

Room sim for something else to look at

 

Red- nominal position - however Lsub facing front  wall, Rsub facing into room

Black- nominal position - subs facing into corners with 80hz xover

Green- Lsub at 1.5m facing front wall, Rsub 5.3m along RHS wall.

Blue-nominal position - subs facing into corners with 120hz xover

 

From the images here I am thinking the blue measurement looks the best response, good decay + most convenient position - thoughts?

 

Cheers

 

 

You are pretty lucky to have a room like that! And you still have bass at 10Hz - wow! All 4 sweeps look excellent, only within a couple of dB relative to each other. I don't think you would hear too much of a difference between them. Yes, the blue sweep looks the best. 

 

I assume these are all sweeps of the sub only, without the mains? From the main listening position? 

29 minutes ago, Keith_W said:

I assume these are all sweeps of the sub only, without the mains? From the main listening position?

 

Yes,  I have been measuring subs together & individually from MLP. This is what I'm concentrating on ATM.

The above sweeps represent combined subs.

Have also done sweeps of L, R & C speakers for integration / xover purposes. 

 

Edited October 29, 20232 yr by Craigandkim
EDIT

3 hours ago, Keith_W said:

You are pretty lucky to have a room like that! And you still have bass at 10Hz - wow! All 4 sweeps look excellent, only within a couple of dB relative to each other. I don't think you would hear too much of a difference between them. Yes, the blue sweep looks the best. 

 

Agree with all that Keith said. No obvious nulls below 80Hz. And at 1/48 smoothing! Being able to have the subs a fair way in from the side walls and flush up against the back wall is working wonders.

 

You can Shift-Click and drag across a frequency range (on an individual trace graph) if you want some slopes and numbers to back up your best trace choice.

 

My small room (3.92x3.65m) with one sub on left looks like this at 1/48 smoothing, same 50-100dBSPL range,  20-80Hz... 🙄 It's basically impossible to get the speakers and sub in any position to avoid cancellation nulls.

 

I'm listening almost in near field, with plenty of acoustic treatment to get the system sounding good.

Better than you'd ever expect in reality. I love it.

 

PS. I roll bass off below 28Hz in my small room (via PEQ available in the Auralic Altair G1) because I don't listen to much music at all with 10-20Hz content.

 

Edited October 29, 20232 yr by PeterB7858
Clarification

Been busy re-arranging the room these past few weeks and my Sov's are basically back in position #1 from years ago where I found overall was the best spot for depth and imaging at the expense of a boosted bass response however 20mins with REW got me the following result (red = before, green = after)

 

]

17 minutes ago, tubularbells said:

Been busy re-arranging the room these past few weeks and my Sov's are basically back in position #1 from years ago where I found overall was the best spot for depth and imaging at the expense of a boosted bass response however 20mins with REW got me the following result (red = before, green = after)

 

]

I'd be completely happy with the green response and sit back and listen to the music!

 

Mike

1 hour ago, almikel said:

I'd be completely happy with the green response and sit back and listen to the music!

 

Mike

 

Thought id throw on a few LP's to see how it sounded and here I am hours later still enjoying. Difference is HUGE sonically.

On 29/10/2023 at 3:39 PM, PeterB7858 said:

Hi @PeterB7858,

 

do you hear than deep null just above 50Hz? serious question

It looks quite steep/sharp on the graph - and the ear is not sensitive to sharp/high Q  changes in the frequency response...

...in a previous post in a different thread I made an observation that a dip would not likely be audible as it looked sharp/high Q on the graph, but others on SNA pointed out it was broader/lower Q than my first interpretation.

 

I'm interested in your subjective opinion on whether you can identify that null in your room? maybe never on real music but possibly on a sweep or test tones?

 

@davewantsmoore

if we look at this snip of Peter's graph

 

Looking where the green FR curve crosses the aqua average line, let's say the dip is approx 45Hz - 55Hz wide in that region - which is 0.29 octaves wide.

If we instead look at the peak red line crossings let's say it's approx  43Hz - 60Hz - which is 0.48 octaves wide.

 

In your opinion would this dip be audible?

Is there a function in REW to calculate the Q of a dip?

Are there guidelines on what sharpness/Q of dips would be audible vs not audible?

 

With my DEQX, and likely similar with mini DSP etc, I can overlay an EQ curve and shape it to the dip in FR, and DEQX will tell me what the centre frequency, Q and gain of the band of EQ is - likely the easiest way for me to measure the Q of a dip if REW doesn't do it.

 

Not that I would attempt to correct a dip like that with EQ!

 

I'm just trying to get an understanding of what constitutes sharp/high Q dips that aren't audible vs broader/lower Q dips that would be audible?

 

cheers,

Mike

 

PS: if anyone is interested in calculating the octave gap between 2 frequencies the calc is:

Octave Gap=Log "base 2" (F2/F1)

or 

Octave Gap=Log "any base" (F2/F1)/ Log "same base" (2) 

Most scientific calculators have Log base 10 and natural log base "e" as standard functions - either would work using the 2nd function

 

Edited November 3, 20232 yr by almikel
typo

2 hours ago, almikel said:

do you hear than deep null just above 50Hz? serious question

It looks quite steep/sharp on the graph - and the ear is not sensitive to sharp/high Q  changes in the frequency response...

 

Hi, Mike.

 

The graph looks worse than it possibly is because I used 1/48th smoothing to match the smoothing Craig had on his graphs above - which are uber impressive! I would never normally even look at 1/48th.

 

Exactly the same graph using psychoacoustic smoothing looks like this... +/- ~2dB.

 

 

There would possibly be a softening on a bass note in the recessed region but I haven't really ever noticed it and the psychoacoustic graph seems to bear that out.

 

I've actually minimised that null a bit since that measurement, too.

 

Regards,

Peter.

 

Edited November 3, 20232 yr by PeterB7858
Clarification

16 hours ago, almikel said:

It looks quite steep/sharp on the graph - and the ear is not sensitive to sharp/high Q  changes in the frequency response...

It is probably broader than it appears (1 octave below is 25 and above is 100Hz)

.... but perhaps lost on most music.

 

I'd be more concerned with the rolloff below 40Hz (at least want to understand why it is there, it might be the natural response of the speaker, or it might be a "problem" .... I don't know anything about the setup).

 

 

OTOH, this is exactly the sort of things that you are looking to fix by moving (or adding) bass sources, as it cannot typically be solved any other way.

Edited November 4, 20232 yr by davewantsmoore

Looking for opinions on my room measurement in respect to RT60.  Thinking it may have too much absorption?

 

 

 

 

37 minutes ago, lucmor444 said:

Looking for opinions on my room measurement in respect to RT60.  Thinking it may have too much absorption?

 

 

 

 

The decay times are reasonable across the graph. Nothing pops out as an issue and fairly normal for perceivably slightly treated area  thin absorb some furniture, in a light construction( I might be wrong.).  Would be good to see the <100Hz part of the graph.

 

The fundamental at 20Khz is fairly (+ 10dB) loud. It's a very good tweeter 🙂 and was on axis to the mic. mic position at the listening position ? Looks like it.

Can you describe the Acoustic Treatment regime  in place and the symptoms you find problematic,  photos are good.

Thank you Playdough.

 

I have attached 30-100Hz graphs.

 

Spot on in respect to construction (its a family room with plasterboard lining and colorbond exterior cladding).  A few windows and doors in the room. Wooden floor with large rug and a 2.5 seater couch.

 

Measurements are without room correction.  Tweeter is a JBL 2453h-sl CD and is centre speaker measured with mic vertical, on axis and at MLP.

 

Room is 5.4mx4.2m with raked ceiling (3.6m max).  I have curtains along the sides and a mixture of absorption and diffusion panels.

 

Room treatment panels include 9 x Artnovian Verona panels, 9 x Artnovian Loa for absorption and mixture of diffusers -  Artnovian Sienna, Artnovian Douro,  Quadratic wooden diffusers and  Sonitus Acoustics 3D diffusers.

 

I'll post some photos later.

 

Perhaps thinking my RT60 should be a bit higher for HT - more between 200-300 rather than low 200 for greater spaciousness?

🙂 Looks in order, I guess a strategy would be remove a few absorb panels. Tune to your personal  preference. Especially if you are cranking the treble to get more "spaciousness" bring the room back into play a little, walls.

I'm running the JBL 2426 CD (from 1200HZ), which will barely make 18khz 🙂 dam good JBL CD you have there.

Edited November 7, 20232 yr by playdough

Just building on lucmor444's post above, can anybody please tell me which REW graphs are going to be most beneficial to work out if my acoustic treatment approach is working/needs more etc? I've been reading the REW Help Index pages but a lot of it is going over my head. I really wish the pages would include a paragraph explaining X graph shows Y and a desirable result is Z (look for x ms etc). I've strategically added absorption in my room and will be targeting diffusion soon, but want to correlate what I've done with a measured improvement. Any links or guidance appreciated. 🙂 

1 hour ago, OzJustin said:

Just building on lucmor444's post above, can anybody please tell me which REW graphs are going to be most beneficial to work out if my acoustic treatment approach is working/needs more etc? I've been reading the REW Help Index pages but a lot of it is going over my head. I really wish the pages would include a paragraph explaining X graph shows Y and a desirable result is Z (look for x ms etc). I've strategically added absorption in my room and will be targeting diffusion soon, but want to correlate what I've done with a measured improvement. Any links or guidance appreciated. 🙂 

 

Waterfall or Spectrogram. 

8 minutes ago, Keith_W said:

 

Waterfall or Spectrogram. 

Thanks. Could you elaborate on exactly what I should be looking for in these graphs? 🙂 

 

Here are some graphs based on my current room treatment. I'm about to add a couple of extra absorber panels to the ceiling unless the results indicate otherwise. 

 

Front left speaker:

Front right speaker:

Centre speaker:

 

With the waterfall graphs, there is an X, Y, and Z axis. X = frequency response, Y = SPL in dB, and Z = time in milliseconds. You can think of it as a normal frequency response graph with fingers that extend outwards, the length of the finger showing which frequencies are resonating and how long they take to decay. 

 

The spectrogram shows the same information, but uses colour to represent SPL instead of a Y axis. It is easier to look at frequency response on a waterfall, and easier to spot resonances on a spectrogram. 

 

Both these graphs depend on a microphone that has been calibrated to an SPL meter so that you know what the noise floor of your room is. For example, if your speakers are very soft, and you have to turn up microphone gain to compensate, it will make the noise floor seem louder than it actually is, and make your waterfall look better than it actually is because the fingers appear to disappear into the noise floor sooner. Most rooms have a noise floor of about 40-45dB. 

 

In the graphs that you posted above, it appears that you have very little room resonance, but this depends on whether your mic was calibrated or not. The target is between 250ms - 500ms. Some people prefer more "wet" (i.e. more resonance), some people more "dry" (less resonance). I am not a fan of rooms that are excessively dry, but it is up to your taste. You will find that adding more room treatment will dampen the upper frequencies more than lower frequencies, which has the effect of making upper freqs too dry, while the bass remains boomy. That is, unless you specifically target the lower frequencies with membrane traps tuned to target specific frequencies. 

 

There are two bass "fingers" visible at 45Hz and 70Hz, corresponding to a first order mode (half wavelength) of 3.81m and 2.45m respectively. So you look around your room and look for dimensions which are multiples of these first order modes, e.g. 3.81m, 7.6m, 11.4m, etc. 

@OzJustin

 

I notice that you have a MiniDSP and subs, so I would suggest that you can lop off the 2 peaks at 35Hz and 70Hz with EQ.

 

IMO, no other EQ or treatment is required since the mids and highs are flat enough and apart from a couple of small "rings" at 600 and 1500Hz which you won't hear.

 

The cancellation at 110Hz is more of a problem in that EQ won't fix it because you have some direct sound being cancelled by other waves that are 180 degrees out of phase. 

 

It could be that the subs aren't in phase with the mains (is 110 the XO point?).  In which case adjusting the delay between the subs and the mains might help bringing them into better alignment phase wise.

 

The other possibility is that there is a strong reflection somewhere in the room, ie. the reflections are out of phase with the direct.  I had something similar which I solved (somewhat) by moving the subs to different locations in the room.  Time consuming and a nuisance. 

 

Or you could not worry about the anomaly at 110 but miss the occasional whack on a drum.

20 hours ago, Keith_W said:

With the waterfall graphs, there is an X, Y, and Z axis. X = frequency response, Y = SPL in dB, and Z = time in milliseconds. You can think of it as a normal frequency response graph with fingers that extend outwards, the length of the finger showing which frequencies are resonating and how long they take to decay. 

 

The spectrogram shows the same information, but uses colour to represent SPL instead of a Y axis. It is easier to look at frequency response on a waterfall, and easier to spot resonances on a spectrogram. 

 

Both these graphs depend on a microphone that has been calibrated to an SPL meter so that you know what the noise floor of your room is. For example, if your speakers are very soft, and you have to turn up microphone gain to compensate, it will make the noise floor seem louder than it actually is, and make your waterfall look better than it actually is because the fingers appear to disappear into the noise floor sooner. Most rooms have a noise floor of about 40-45dB. 

 

In the graphs that you posted above, it appears that you have very little room resonance, but this depends on whether your mic was calibrated or not. The target is between 250ms - 500ms. Some people prefer more "wet" (i.e. more resonance), some people more "dry" (less resonance). I am not a fan of rooms that are excessively dry, but it is up to your taste. You will find that adding more room treatment will dampen the upper frequencies more than lower frequencies, which has the effect of making upper freqs too dry, while the bass remains boomy. That is, unless you specifically target the lower frequencies with membrane traps tuned to target specific frequencies. 

 

There are two bass "fingers" visible at 45Hz and 70Hz, corresponding to a first order mode (half wavelength) of 3.81m and 2.45m respectively. So you look around your room and look for dimensions which are multiples of these first order modes, e.g. 3.81m, 7.6m, 11.4m, etc. 

Thanks Keith. I really appreciate you taking the time to explain it to me in plain English. That is exactly what I was looking for in the guides.

 

I'm using a Umik-1 with the USB calibration file and run my measurements at 75dBFS so to my knowledge everything should be accurate. Noise floor is about 45dB at my MLP.  My pre-treatment measurements were in the 300-350ms range generally while post-treatment seems to be 200-250ms range. I can hear a significant difference so hopefully I'm at a reasonable position now and not overdamped. 

 

One question I had is why does the ringing at 500Hz and 1100Hz show on the waterfall graph but not the spectrogram? Aren't they both showing a sound that is taking longer for the energy to decay than other frequencies?

11 hours ago, aechmea said:

@OzJustin

 

I notice that you have a MiniDSP and subs, so I would suggest that you can lop off the 2 peaks at 35Hz and 70Hz with EQ.

 

IMO, no other EQ or treatment is required since the mids and highs are flat enough and apart from a couple of small "rings" at 600 and 1500Hz which you won't hear.

 

The cancellation at 110Hz is more of a problem in that EQ won't fix it because you have some direct sound being cancelled by other waves that are 180 degrees out of phase. 

 

It could be that the subs aren't in phase with the mains (is 110 the XO point?).  In which case adjusting the delay between the subs and the mains might help bringing them into better alignment phase wise.

 

The other possibility is that there is a strong reflection somewhere in the room, ie. the reflections are out of phase with the direct.  I had something similar which I solved (somewhat) by moving the subs to different locations in the room.  Time consuming and a nuisance. 

 

Or you could not worry about the anomaly at 110 but miss the occasional whack on a drum.

I appreciate you taking a look at these graphs for me.

 

Won't the 45Hz and 70Hz apparent room mode issue (as Keith pointed out) from my speakers disappear once I set my crossovers above that threshold e.g. 120Hz for subs to take over? (For context, these measurements were taken at my MLP with all crossovers set to 40Hz and Anthem ARC turned off). If the crossovers won't fix it, can strategic room treatment fix the apparent room mode issue (i.e. sound bouncing off two surfaces and lingering longer)? At this low frequency I'd think it would be difficult to tame and is there a way to work out which surfaces are the reflecting problem?

 

I haven't been able to work out the 80-150Hz cancellation across my three main speakers and it has been bugging me for a while (see my raw response graph below for these speakers and all crossovers set at 40hz with ARC off). My surround left and right also have a dip 80-200Hz range but my rear and Atmos speakers are ok. I've moved speakers, lounge and subs with no success. Is there anything I can do to get the speakers performing in this range? My fall back position is to assign crossovers for these speakers at 130Hz and set the High Frequency Extension (HFE) at 250Hz on my receiver. That should mask the lack of speaker bass.

 

I was intending to add approx. 3.6 - 5.6m2 (6-10 panels) worth of Sonitus 3D Diffusers (BigFusor 2s) strategically to my side and rear walls and on the ceiling above my MLP. That will add a small amount of additional absorption in concert with diffusion. Does that sound reasonable based on my measurements?

 

4 hours ago, OzJustin said:

One question I had is why does the ringing at 500Hz and 1100Hz show on the waterfall graph but not the spectrogram? Aren't they both showing a sound that is taking longer for the energy to decay than other frequencies?

 

Because of different windowing. If you go to your spectrogram, click on the top right "controls" button (it looks like a gear). Change "Window" to match your waterfall. You should see the 500Hz/1100Hz ringing reappear.