or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs
New Posts  All Forums:Forum Nav:

Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How... - Page 163

post #4861 of 9556
Quote:
Originally Posted by AustinJerry View Post

Here are your room modes:




If I were looking for a starting point for sub placement, I would locate it along the 17' wall, at either the 1/4 or 3/4 mark, i.e. 4' 3" from the front wall, or 4' 3" from the back wall.  This places the sub in the 66Hz null associated with the long wall.  A second potential location is in the middle of the 12.5' wall, which would be in the 45Hz null.

I would then measure with the sub only, Audyssey off.  It is going to be somewhat difficult achieving a flat bass response with only one sub.  On the other hand, your room dimensions are more friendly than others I have seen, because your length and width have no common modes below 100Hz.  Looking forward to seeing new measurements!

When you mean measure with the sub only, you mean only to 80 hz? Here's the measurements at the 3/4 wall placement ( side 17 ft wall ) at 4.5 feet:

Gear mentioned in this thread:

post #4862 of 9556
Here's a graph of the mid point of the back wall:



When you say measure sub only, I'm not sure what you mean exactly. In the preferences I've got to select High Def Audio 1 through 8, the numbers corresponding to the speakers ( 1 being left and 2 being right fronts ) ... I don't see an option for Sub only ...

And lastly here's my original position for the sub: ( Sorry says l and r front but is L + sub and R + sub


Edited by Sean Spamilton - 10/2/13 at 8:12pm
post #4863 of 9556

Sean, what is with the left and right limits on your graphs?  I thought we had discussed that.  Click the limits icon:

 

 

Set the left limit to the low end of your measurement, and the right limit to the upper end of your measurement.  Here are two examples:

 

 

 

Both of these graphs show the same frequency measurement.  Which looks better?  Which looks like the graphs you posted?  Understand limits now?

 

OK, having said that, something is still seriously wrong with the results of your measurements.  The 40dB dips in the measurements indicate a major issue, and I can't tell if it is your measurement technique, or the actual response in your room.

 

How to take a measurement of the sub only?  My understanding is that one of the eight channels available to you using HDMI is the LFE channel, which is the sub channel.  It should be either channel 3 or channel 4.  Select that channel, run a sweep from 15-300Hz, and see which of the two options outputs to the sub only.  A second alternative is to physically disconnect the left and right speakers from the amps, but you need to be careful if you do this.

 

What you should be shooting for is a much flatter response in the bass region, closer to what is displayed in the examples I posted above.

post #4864 of 9556
Quote:
Originally Posted by Sean Spamilton View Post

In the preferences I've got to select High Def Audio 1 through 8, the numbers corresponding to the speakers ( 1 being left and 2 being right fronts ) ... I don't see an option for Sub only ...

Sub is #4.
post #4865 of 9556
Quote:
Originally Posted by AustinJerry View Post

Sean, what is with the left and right limits on your graphs?  I thought we had discussed that.  Click the limits icon:

We did, but When I read the post to measure sub only, I inferred it to mean I should just measure what I have the sub crossed over at and below, so 20 to 80 - I see from the next post how to actually measure the sub alone, and will try it again tomorrow as soon as I can .. ( It's after 11 pm here and I don't like putting the neighbors through bass sweeps at this time of night smile.gif )


Quote:
Originally Posted by AustinJerry View Post

Both of these graphs show the same frequency measurement.  Which looks better?  Which looks like the graphs you posted?  Understand limits now?

Yes sir I get it. Sorry. Lol. I Feel like a high school kid getting a lesson from the prof tongue.gif

Quote:
Originally Posted by AustinJerry View Post

OK, having said that, something is still seriously wrong with the results of your measurements.  The 40dB dips in the measurements indicate a major issue, and I can't tell if it is your measurement technique, or the actual response in your room.

How to take a measurement of the sub only?  My understanding is that one of the eight channels available to you using HDMI is the LFE channel, which is the sub channel.  It should be either channel 3 or channel 4.  Select that channel, run a sweep from 15-300Hz, and see which of the two options outputs to the sub only.  A second alternative is to physically disconnect the left and right speakers from the amps, but you need to be careful if you do this.

What you should be shooting for is a much flatter response in the bass region, closer to what is displayed in the examples I posted above.

Did you receive the PM I sent with my Cal file? did it work out on your end? ... Maybe the sub only measurements will be more revealing tomorrow - or else yeah - I have serious room issues at play... I've already started contemplating / figuring out how to afford a second PC12.
post #4866 of 9556
Quote:
Originally Posted by Sean Spamilton View Post


Did you receive the PM I sent with my Cal file? did it work out on your end? ... .

I received it and will test it out tomorrow, thanks.
post #4867 of 9556
I couldn't resist trying just one at this time of night - Hope the limits are set more to the required specifications:



Still looks bad. Seems like everything from 40 - 70 is messed up.

Just for the record, here's a sketch of my room - It's actually closer to 15' long then 17, and 12' 8 wide. There's an unusual corner in the front left, and the right wall juts out 5 inches into the room at around 9 foot mark on the right. I doubt this makes a difference, but for the record here it is, if it helps at all.


Edited by Sean Spamilton - 10/2/13 at 10:49pm
post #4868 of 9556

First of all, the fresh measurement and its graph are much better.  With the limits set properly, it is much easier to read.  And with the sub isolated, we can now clearly see that the frequency response irregularities are a result of the sub and its interaction with the room.

 

And the sketch shows that the room shape makes it somewhat difficult to apply the Room Mode calculator to predict a likely spot for the sub.  Next step, IMO, would be to try the "sub crawl".

 

- Temporarily place the sub at your primary listening point, at ear height if possible (perhaps placed on a chair or a couple of boxes)

- Take the REW mic, adjust it so it is at the same height as the cone of the subwoofer (if the sub were on the floor, of course)

- Walk around the room and run a frequency sweep at each potential spot where the sub could be placed.

 

The hope is that this process will result in the "best" placement, given the room dimensions, and the fact that there is only one sub.  Once you have placed the sub in that spot, take another measurement with the sub only, and then three additional measurements, sub+left, sub+right, and sub+left+right (all 15-300Hz).  Post these measurements (separate graphs, which will be easier to read), which will represent "as good as it gets", given that you only have one sub.  Depending on the measurement results, and your listening tests, you can decide if you want to consider a second sub.  BTW, what make and model sub do you have?  A second sub should be a matched sub, if possible.

 

Finally, I have tested out the UMM-6 calibration file that you sent.  The file has a sensitivity parameter in the first line:

 

 

When I use this file, and open the REW SPL tool to calibrate the mic, I get this message:

 

 

REW will indeed not allow the mic to be calibrated if a sensitivity parameter is loaded.  I'm not sure I would have coded the program this way, but there is a simple workaround, should you want to calibrate the mic manually.  Simply open the calibration file, delete the line with the sensitivity parameter, and save the file with a different name.  Then use this edited version the next time you run REW, and you will be able to manually calibrate the mic.  So relax, Sean, you are not crazy (what a relief!).  ;)

 

Edit:  I think adding this information to the Guide might help others in the future.

post #4869 of 9556

@Sean Spamilton: I agree a sub crawl is probably the way to go from here but have you also tried the midpoint of the right or left wall to the side of the couch?  Midpoints tend to be a good position for single sub placements but as Jerry noted, irregular room configurations can throw off the standing waves.  I was wondering if you might get a bit of nearfield bass effect though especially with the mid right wall placement?  Tough to tell from your sketch how close the sub would be to the MLP but if this is feasible then perhaps mid room right behind the couch would be a good option to eliminate much of the room effects at the MLP?  Here's a modified standing wave chart based on your new dimensions:


 



 


The other thing I didn't see mentioned was your ceiling height?  I used 8ft because that is what Jerry had in his chart but I didn't see where you confirmed this dimension?  Probably won't have too much effect though doesn't hurt to try raising the sub as well if this is feasible.

post #4870 of 9556
^
Nearfield would be within 1-2 ft of the ears. If feasible this is the simplest method to get superior bass.

"Sub crawl" isn't necessary if measurement gear is at hand. Measuring is far superior to putting the speaker at the listening position and using ears as measuring instruments. Play pink noise and observe the frequency response live with an RTA while moving the sub.
post #4871 of 9556
You may be right, but sometimes subs are difficult to move around.
post #4872 of 9556
^
If subs can't be moved then a "sub crawl" is out of the question too.
post #4873 of 9556
Quote:
Originally Posted by AustinJerry View Post

You may be right, but sometimes subs are difficult to move around.

 

I can support that view unequivocally :)

 

At least with your method, the sub only has to be moved once, onto the chair. 

post #4874 of 9556
You can use the 'magical' furniture sliders to easily move the sub around w/o lifting, similar to these.
post #4875 of 9556
Quote:
Originally Posted by shpitz View Post

You can use the 'magical' furniture sliders to easily move the sub around w/o lifting, similar to these.

 

I do use those but even with them, moving 2 Submersives is not something to be undertaken lightly ;)

post #4876 of 9556
Quote:
Originally Posted by markus767 View Post

Nearfield would be within 1-2 ft of the ears. If feasible this is the simplest method to get superior bass.
That's what I would do. Since his sub is a 3-foot tall cylinder, it would mean laying it sideways behind the couch. Don't know if he is willing to do that.
post #4877 of 9556
Quote:
Originally Posted by AustinJerry View Post

First of all, the fresh measurement and its graph are much better.  With the limits set properly, it is much easier to read.  And with the sub isolated, we can now clearly see that the frequency response irregularities are a result of the sub and its interaction with the room.

And the sketch shows that the room shape makes it somewhat difficult to apply the Room Mode calculator to predict a likely spot for the sub.  Next step, IMO, would be to try the "sub crawl".

- Temporarily place the sub at your primary listening point, at ear height if possible (perhaps placed on a chair or a couple of boxes)
- Take the REW mic, adjust it so it is at the same height as the cone of the subwoofer (if the sub were on the floor, of course)
- Walk around the room and run a frequency sweep at each potential spot where the sub could be placed.

The hope is that this process will result in the "best" placement, given the room dimensions, and the fact that there is only one sub.  Once you have placed the sub in that spot, take another measurement with the sub only, and then three additional measurements, sub+left, sub+right, and sub+left+right (all 15-300Hz).  Post these measurements (separate graphs, which will be easier to read), which will represent "as good as it gets", given that you only have one sub.  Depending on the measurement results, and your listening tests, you can decide if you want to consider a second sub.  BTW, what make and model sub do you have?  A second sub should be a matched sub, if possible.

Finally, I have tested out the UMM-6 calibration file that you sent.  The file has a sensitivity parameter in the first line:




When I use this file, and open the REW SPL tool to calibrate the mic, I get this message:




REW will indeed not allow the mic to be calibrated if a sensitivity parameter is loaded.  I'm not sure I would have coded the program this way, but there is a simple workaround, should you want to calibrate the mic manually.  Simply open the calibration file, delete the line with the sensitivity parameter, and save the file with a different name.  Then use this edited version the next time you run REW, and you will be able to manually calibrate the mic.  So relax, Sean, you are not crazy (what a relief!).  wink.gif

Edit:  I think adding this information to the Guide might help others in the future.

Great! thanks for trying that for me. I've started running some tests, recalibrating the spl at each different position. The graphs all look similar to the ones from earlier, if not identical, but at least the equalized spl at all the positions keeps it in the same spot on the graph for overlays smile.gif
Quote:
Originally Posted by jkasanic View Post

@Sean Spamilton: I agree a sub crawl is probably the way to go from here but have you also tried the midpoint of the right or left wall to the side of the couch?  Midpoints tend to be a good position for single sub placements but as Jerry noted, irregular room configurations can throw off the standing waves.  I was wondering if you might get a bit of nearfield bass effect though especially with the mid right wall placement?  Tough to tell from your sketch how close the sub would be to the MLP but if this is feasible then perhaps mid room right behind the couch would be a good option to eliminate much of the room effects at the MLP?  Here's a modified standing wave chart based on your new dimensions:


 



 


The other thing I didn't see mentioned was your ceiling height?  I used 8ft because that is what Jerry had in his chart but I didn't see where you confirmed this dimension?  Probably won't have too much effect though doesn't hurt to try raising the sub as well if this is feasible.


My ceilings are exactly 8 feet. Good guess! I've tried the near filed placement directly behind the couch, as you will see evidenced in the graphs below. I haven't started the sub crawl yet, as I wanted to take some " Sub Only " sweeps while moving the sub around the room first. The near field behind the couch didn't look to good, huge nulls at 30 and 60. So far, the best positions ( if you could call it best ) would be the 2/5 front wall, as there is no dip at 30 hz, but isn't great from 45 - 80, and the half back wall, which near eliminates the dips from 40 - 70, but introduces a huge one at 30...

Quote:
Originally Posted by sdurani View Post

That's what I would do. Since his sub is a 3-foot tall cylinder, it would mean laying it sideways behind the couch. Don't know if he is willing to do that.

I tried the sub behind the couch as mentioned earlier, but did so with it standing up. I'm not willing to lay it down, as there is still a fair amount of space back there that I use with a table and to store my bike. In other words I'd be kicking it all the time, so laying it down is out smile.gif



The individual lines correspond to the descriptions of the sub placement indicated on the bottom of the graph.
Since it's a little small, the colours are as follows:
Red is 3/4 wall
Purple is 1/2 wall directly inbetween couch and wall on right side
Green is the Nearfield behind the LP right behind the couch
Cyan is half back wall
Orange is Front Wall 2/5

Oh and AustinJerry asked earlier, It's a SVS PC12
Edited by Sean Spamilton - 10/3/13 at 11:22am
post #4878 of 9556

Looks like 2/5 front wall to me?!  Did you already say if you're using RC?  Looks like a perfect job for XT32...

post #4879 of 9556
Quote:
Originally Posted by jkasanic View Post

Looks like 2/5 front wall to me?!  Did you already say if you're using RC?  Looks like a perfect job for XT32...

It looked that way too for me, but doing bass sweeps reversed, I get a huge suck out at around 50 there.
post #4880 of 9556

So the multiple measurements in post #4877 was actually with the sub at MLP and microphone in the sub locations (i.e. sub crawl)?

post #4881 of 9556
Ok, Well I've settled on the front 4/5 wall for now, as I wasn't sure about the strange suck out at 50hz when the mic position was reveresed with the sub on a box at the LP, when at the 2/5 position. I'm tight on time, and have to shower then run to work, but these are the final measurements for the 4/5 wall position. Ironically this is where I had the sub in the first place rolleyes.gif .. Although the 2/5 might indeed have slightly better response, according to the first set of tests ( when the sub was there and the mic at the LP ) ... it involves completely removing my AVR and wiring, and moving it it under the center channel in my cabinet. Although this CAN be done, it will take me QUITE some time, and am not quite sure if it's worth it or not. Maybe you guys can chime in. In the mean time, I'm touching base with my dealer about the possibility of getting a second PC12. My logic is that according to the graphs, if I leave one at the 4/5 or 2/5 front wall, I'll still have great response until 40, but can move a sub to the 1/2 back wall, which as displayed above has a huge suck out at 30 - 40, but takes care of the 40 to 70 region. I think that's probably the best case scenario. I just have to wait until pay day and resign myself to NOT getting a new winter jacket until November. And hurting my credit card a bit more. Oh well.



post #4882 of 9556
Quote:
Originally Posted by jkasanic View Post

So the multiple measurements in post #4877 was actually with the sub at MLP and microphone in the sub locations (i.e. sub crawl)?

No those were the actual measurements with the sub in place. Post 4879 were while doing the sub crawl method. I actually had all 5 positions mapped out but while moving my PC unplugged it and lost everything. So I just re-mapped what the best 2 were. The Back wall was best for 40 - 70 .. but unfortunately my mic didn't reach back far enough to map it anyways. I just thought I'd go with the best of my options from what I saw. While the 2 / 5 did show signs of improvement over the 4 / 5 .. they weren't drastic, and since I've resigned myself to add a second sub anyways, can do more measurements then. That's my thinking anyways! If both the 2/5 and 4/5 are similar up until 40 hz, and the 2/5 just slightly better after 40hz, since the plan is now to add a second sub to even out the 40-70, it's not as crucial at the moment.
Edited by Sean Spamilton - 10/3/13 at 12:22pm
post #4883 of 9556
Quote:
Originally Posted by Sean Spamilton View Post


The individual lines correspond to the descriptions of the sub placement indicated on the bottom of the graph.
Since it's a little small, the colours are as follows:
Red is 3/4 wall
Purple is 1/2 wall directly inbetween couch and wall on right side
Green is the Nearfield behind the LP right behind the couch
Cyan is half back wall
Orange is Front Wall 2/5

Oh and AustinJerry asked earlier, It's a SVS PC12

 

Orange looks best to me.  Too bad it involves a lot of work.  A second sub will yield significantly different results, so I wouldn't move anything around yet.

post #4884 of 9556
Quote:
Originally Posted by Sean Spamilton View Post

Green is the Nearfield behind the LP right behind the couch

How far from the mic was the driver for the near field setup?
Could you plug the port and measure again?
post #4885 of 9556
Quote:
Originally Posted by AustinJerry View Post

Orange looks best to me.  Too bad it involves a lot of work.  A second sub will yield significantly different results, so I wouldn't move anything around yet.

Yes.... I'm leaning towards a second sub, but if something comes up and I can't quite stretch myself that thin I'll utilize that as my sole position I think... If and when I get the second sub ( maybe next week depending on my take for the week ) - that position could be used as 1 with the other near the back 1/2 wall - or as symmetry so I could have one at 2/5 and 4/5... I'll have to see how the frequency turns out though. No point in doubling up the 20 to 40 hz region and leaving the 40-70 out to dry. I guess we'll see! ( I'm Pretttttty sure I'm going to get it smile.gif I don't REALLY need a winter coat until November! It's all about priorities ..shelter > audio > food > clothing

Quote:
Originally Posted by markus767 View Post

How far from the mic was the driver for the near field setup?
Could you plug the port and measure again?

It was directly behind the MLP on the couch, - the drivers on the bottom of the sub and the port comes out the top of it ( PC12 is a cylinder ) so the port was about 2 feet from my head, while the driver was 2 feet back and 3 feet down. I can't plug the port on the PC12 it's got a protective mesh over the top. Thanks for the suggestion though.
post #4886 of 9556
^
That explains why the near field response doesn't look better. It's like two sources (driver and port) that have a different response and one is too far away. You would need to plug the port and get the driver as close to your head as possible. If that's not doable then you're out of luck.

Near field sub is better than multiple subs in the room (except double bass arrays):
http://mehlau.net/audio/dual_nearfield_sub/
post #4887 of 9556
Quote:
Originally Posted by Sean Spamilton View Post

It was directly behind the MLP on the couch, - the drivers on the bottom of the sub and the port comes out the top of it ( PC12 is a cylinder ) so the port was about 2 feet from my head, while the driver was 2 feet back and 3 feet down. I can't plug the port on the PC12 it's got a protective mesh over the top. Thanks for the suggestion though.

 

Sean, I used to own those subs. You can in fact remove the top mesh coating on the PC12. HST, the PC12 is not designed to have its port plugged (unlike some of the other SVS cylinder subs which are provided with plug ports specifically for the job). Plugging the port on a PC12 might actually damage it and also invalidate the warranty, so I would definitely steer clear of it.

post #4888 of 9556
Quote:
Originally Posted by kbarnes701 View Post

Sean, I used to own those subs. You can in fact remove the top mesh coating on the PC12. HST, the PC12 is not designed to have its port plugged (unlike some of the other SVS cylinder subs which are provided with plug ports specifically for the job). Plugging the port on a PC12 might actually damage it and also invalidate the warranty, so I would definitely steer clear of it.

I didn't know you could take that mesh off, but yeah I didn't think you could plug it anyways. Don't worry, I don't plan on destroying it in the hopes of a better frequency response. Better to just get a second sub.
post #4889 of 9556
^
The sub doesn't get destroyed by plugging the port. That's nonsense. It's quite the opposite. By plugging the port you prevent the sub from mechanical damage. Audyssey doesn't care about the type of sub and might boost frequencies below the driver's resonance frequency. In any vented sub design such a boost can lead to potential damage of the driver (voice coil hits the pole plate).

Again, you don't get better sound quality just by using two subs. Modal decay is as important as frequency response. Near field subs give you better modal behavior whereas multiple subs don't guarantee faster modal decay.
post #4890 of 9556
Quote:
Originally Posted by Sean Spamilton View Post
 
Quote:
Originally Posted by kbarnes701 View Post

Sean, I used to own those subs. You can in fact remove the top mesh coating on the PC12. HST, the PC12 is not designed to have its port plugged (unlike some of the other SVS cylinder subs which are provided with plug ports specifically for the job). Plugging the port on a PC12 might actually damage it and also invalidate the warranty, so I would definitely steer clear of it.

I didn't know you could take that mesh off, but yeah I didn't think you could plug it anyways. Don't worry, I don't plan on destroying it in the hopes of a better frequency response. Better to just get a second sub.

 

I dont think you'd destroy the sub, but as it was specifically NOT designed to have the port plugged (unlike the other cylinder subs from SVS) I'd personally be very careful about blocking the port. If SVS had intended this sub to be run with the port blocked I am sure they would have provided blocking plugs as they do with the other subs. ICBW, but I think it's best to err on the side of caution. Or at least drop SVS an email and ask their view - IME they reply to emails almost before they receive them :)

 

I think a second sub can help in various ways, but it is usually used to smooth the bass response and create better bass over a larger area of the room. I found that adding a second sub was the only way I could deal with my room modes satisfactorily and achieve much nicer bass quality (rather than quantity). I ran dual PC12s up to the time I bought the dual Submersives.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat

Gear mentioned in this thread:

AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs