Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How... - Page 38

Your first and third graphs are at 1/12 smoothing, and the second is at 1/24 smoothing.  Consistency, please!  And didn't we agree to use 1/6 smoothing for full-range frequency measurements?

 

Where is the bass?  If the left and right speakers are set to "small", we should be seeing the sub fill in the lower frequencies.  Or is it turned off?

 

What would be nice to see is for you to temporarily move the sub to the center position, place the left and right where Sanjay recommended, re-run Audyssey, and then show a "left+right+sub" full frequency sweep, 1/6 smoothing.

 

What AVR do you have?  If it is a Denon, you can save the current calibration and re-load it later.  If not a Denon, then unfortunately SOL. 

Can you also confirm whether the mic level issue has cleared up with the 100% setting?

 

Jerry, sorry for the inconsistency on the graphs.  Still getting the hang of REW and I like to look at my graphs at 1/24th to get an idea of what's going on as I find 1/6th really hides a lot!  Also, I believe the latest version of the guide says that 1/6th OR 1/12th is acceptable if noted.  In any case, I updated my graphs to 1/6th.

 

These plots were just of the Left, Center and Right with no bass.  Here's what each looks like before and after the change to L and R with bass from sub included (but sub is still in current location which is front wall right corner loaded):

 

Right + Sub:

 

 

Left + Sub:

 

 

Center + Sub:

 

I agree with what you're saying on the next set of measurements but since I know I want to relocate the sub to the mid front wall location, I'd prefer to wait on the Audyssey calibration until I can make the change permanent (rather than measuring with the sub in front of my false wall as I did yesterday just to see if there was any improvement.  I'm going to need some help getting my screen down so maybe I can do these measurements next weekend.

 

As I mentioned previously, as long as I met the headroom threshold (for me it was around 75), then there was no difference in measurements taken with the UMM-6 mic level at 75, 80 and 100.  However, dropping below 75 (I tried 60 and 40) resulted in sporadic results but mainly the SPL was noticeably lower almost as if a negative db offset was in place.

 

I guess I'm wondering if there's any harm in just setting it to 100?  The headroom in my measurements was nowhere close to clipping range at a UMM-6 mic recording level of 100.  The average was about 25 dB with a bottom of around 20 dB when I did L+R+sub measurements.  Hope this helps!

The current measurements don't look bad above around 150Hz, and we know the bass response will improve once you re-position the sub.  And of course a fresh Audyssey calibration will improve things as well.

 

You seem to be reluctant to measure the combined left+right speakers.  Keep in mind that sounds below approximately 100Hz are omni-directional, and will be produced by the left, right and sub in combination.  This is why measuring left and right together will give you the best representation of how the low frequencies are looking.

 

Nice job so far, and I would leave the mic level at 100%.  More fun next weekend!

 

Thanks for the feedback!  So far, I haven't been reluctant to measure anything!  I just didn't post the L+R (and L+R+Sub) graph.  Check out the legends in my graphs and you'll see I've got all the data!

 

I'll post the comparison of measurements 8 and 16 (L+R+Sub before and after) tomorrow since I just shutdown the laptop for the evening.

I use hunecke loudspeaker calculator to get a rough idea of room placement and it was showing that my two subs moved to the centre of the front wall would actually give a flatter response in the room. So I tried it... two subs co-located behind the TV screen in the middle of the front wall. And it did indeed measure a lot flatter (and still nice and low) than what I was getting with them in the two front corners.

Just goes to show the importance of measuring your own room to find what actually works rather than blindly following the 'common wisdom' on interweb forums.

 

 

Can you upload the measurement file as well?

 

This doesn't surprise me, as subs in the front corners is not the best placement.  If you are an experimenter, try the two subs at 1/4 and 3/4 of the width.

That configuration is not one of the optimal ones either. If you have more than one sub, you want to put them on separate walls (center positions). That way, they remove modes in both dimensions. Here are the recommended ones when you take into account the least variations seat to seat, smoothest response and maximum power:



The corner configurations always produce more power. As such, if you can have four subs, put them in the four corners. Putting the same subs in the center walls substantially reduces the total power but provides a slight improvement in seat to seat variation.

Yes, measure your room, find the areas in the bass where the response shows a peak and there is a related ringing in the time domain (as visible on a waterfall, spectral decay, spectrogram). Those are your modal frequencies. Now to find their distribution the best way is to play a sine wave at the frequency of the room mode and walk around the room. You will hear the SPL change from loud to soft. That's how you find the nulls for sub placement etc.

Well I am an a experimenter so I did try that today. Still didn't work/measure particularly well unfortunately. It also put the two subs at different distances from the LP, so I tried adjusting the phase on one of them. I usually like to go for sub placement that puts both subs at the same distance from the LP as my AVR only treats both sub outputs as one.

 

Sorry if I missed it but what are your room dimensions?

 

Ok, here are the graphs you requested:

 

L+R 1/6th smoothing before and after the change to 42" from sidewall for L and R:

 

 

Here's the same plus sub (in right corner location):

 

 

That null at 111 Hz is interesting...will need to see how Audyssey does once I get the sub relocated.  Also, I should reiterate that everything is crossing at 90Hz currently.  I'd be interested to learn more about selecting a different crossover based on the interaction of the mains with the sub?  I'm still curious why Audyssey chose 40Hz on my surrounds (even though the manufacturer specs rate them 40Hz-20kHz, I'd be surprised if they performed that well down there).

 

Finally, here's another Yousendit Link for the baseline measurements from yesterday as well as the measurements after the change to 42" for Left and Right.

 

Are you familiar with the sub distance tweak to improve the result around the XO frequency? (Once you have run Audyssey).

^^^^ I thought so, but I was thinking that was more about tweaking Audyssey for 2 subs?  Are you saying it will improve my Xover region with only one sub?  I guess I need to re-read the FAQ...again!

My mic is in the UK, at the local mail depot. I can't do anything until I have paid the Customs tax on it and I can't do that until they send me a letter telling me how much I owe them. If the letter arrives tomorrow, then I should have my mic sometime this week for sure.

 

The FAQ won't help you - I just checked and the link to the sub diatance tweak procedure is missing in d)1, penultimate bullet point. 

 

Anyone got the link so I can fix it?

Got my UMM-6 on Saturday and took a couple measurements. Haven't got too deep into it, but just couldn't wait to share my first measurements!

This measurement is SUBS ONLY (I have 3), Audyssey OFF, crossover at 80hz. This is after trying multiple placements to get the FR as flat as possible.

The FR doesn't look too bad to me, but the waterfall looks pretty ugly to me under 35hz or so - lotsa ringing. Will room treatment help at all with ringing that low?

Let me know how these look to you guys!

^^^^ What are your room dimensions and current sub locations?  Are all 3 the same?

So I sat down & done a preliminary test & I'm not too secure with the results. I measured from the mlp. I used the narrow band 90 degree cal. file. In order to get REW to recognize the levels I had outpu the signal at -10db & the avr(Onkyo 3009) was at ref level. I sit 12' away. This seems off. So when I ran a full range response(with sub in all graphs) I get the following results. I tried lookin back at what we agreed upon for posting but couldn't quickly locate it. Maybe that should be in your signature too. If it is, my mistake. So here are my results:




Outlaw Sub set at max ouput
Fronts are EMP's, the large set.
Concerned about the large slope in the higher frequencies.

Here's a drawing of my room.

Sub 1: PSA XS15
Sub 2: Velodyne F1500 circa 1995 - Just using it because it still works well and seems to smooth the response a bit.
Sub 3: PSA XS15

^^^^ What is your ceiling height?  Is it flat or vaulted ceiling?

 

EDIT:  Also, what is the distance from the side wall to the first sub as well as the distance between the two (to the middle of each is probably the best) on the front wall?  You have a complex room setup due to the wide open space so we may need the big guns (aka AJ, Kbarnes, sdurani etc.) to chime in at some point!  I'm just trying to gather the intel!

Edited by jkasanic - 2/11/13 at 11:43am

5.58 metres wide x 3.63 long x 2.4 metres to the ceiling. I have a long-wall setup.

 

By long wall setup do you mean that you are projecting your image (hence LCR speakers) on the longest wall and that's why you used 5.58m as the width?  If so, here's your room modes:

 

 

Could you also clarify which sub placements you have tried thus far?  I believe you mentioned trying the 1/4 and 3/4 width placements to no avail.  Do you also have any FR curves of this that you could show (15-300 Hz, 45-105 dB)?  As I mentioned to Alan P., I probably know just enough to be dangerous so I'm trying to help get the pertinent info ready so other more experienced members can comment on your particular setup and issues.

Ooops. Ceiling is 8'.

Just guessing because I'm at work, but each sub on the front wall is probably 3' from each corner making them 9' apart. I'll measure when I go home for lunch.

 

Is this the link you were using?  Or were you using a Craig John post?

 

http://www.avsforum.com/t/1141416/official-hsu-uls-15-thread/1980#post_21557746

Guys, I'm experiencing so many back-n-forth communications here on room details. Would it be possible for members to submit a couple of photos of their room instead? If someone thinks it's too much personal than it's OK. Although I don't think any one of us would like to see how tidy the room is, that's not the concern. We're talkin' acoustics here, aren't we?

Whaddaya think?

FYI, you can see my room details (with some dust included) in my sig. LOL

Yes, that's right.



Here is the FR with the green being when they were both co-located in the centre and the purple is with them at 1/4 and 3/4 after trying to optimise it as best I could with phase and distance settings.



(hopefully this will be bigger and easier to read when you click on it... otherwise I will upload it to another site and post the link)

edit... ok, if you click on it, then click on "original" it should be easier to read.
Edited by kiwi2 - 2/11/13 at 12:34pm

 

 

 

Alan, that looks like quite a challenging room.  With its irregular shape, and its opening to the kitchen, it is difficult to determine room modes.  And it looks like you have limited options on where to place the subs. What flavor of Audyssey do you have, XT or XT32?  Where is your MLP, in one of the two seats, or exactly in the center of the room (left to right)?

 

I would like to see a few additional measurements when you get the chance:

 

- Full frequency sweep, 20-20,000Hz, including front left and right speakers, Audyssey off

- Same measurement, with Audyssey on

- Re-run the Waterfall, increasing the Time Range from 300ms to 600ms (to see if we can see where the ringing stops)