My first stab at REW

[norliss]

Hi there guys. I've just recently got into this whole REW thing and have just spent a few hours getting to grips with it. I would be grateful for some feedback on my results thus far.

[brucek]

I would be grateful for some feedback on my results thus far.

First you have to tell us what we're looking at?

Next, for subwoofers, always use the standard Vertical graph axis of (45dB - 105dB) and the Horizontal graph axis of (15Hz - 200Hz) using the Graph Limits button in the top right corner of REW.

For full range, use the standard Vertical graph axis of (45dB - 105dB) and the Horizontal graph axis of (15Hz - upper limits you desire, i.e. 1KHz -20KHz - certainly no higher than your soundcard can extend). For full range only, enable smoothing to eliminate the comb filtering. Use a 1/3 octave smoothing.

brucek

[norliss]

brucek wrote:

First you have to tell us what we're looking at?

Apologies. This was the "subwoofer" setting. I did 2 512k sweeps from 0-200KHz. The different graphs relate to various gain levels and crossovers.

My main concerns were not wanting to be able to localize the bass as the sub is right next to the seating position.

Where would setting individual sound levels i.e. with test patterns on something like the Avia disc fit in with this whole process?

Thanks again.

[brucek]

This was the "subwoofer" setting

So, this is the subwoofer without the mains, or the mains without the subwoofer, or subwoofer + mains?

My main concerns were not wanting to be able to localize the bass as the sub is right next to the seating position.

Yeah, certainly 80Hz would be your upper limit for the crossover. A lower crossover does help with that, but unfortunately precludes modal resonances being eq'd that may be caused by the lower crossed mains.

Where would setting individual sound levels i.e. with test patterns on something like the Avia disc fit in with this whole process?

REW takes care of all that. Generally it helps to run your receiver test tones to set up the speaker levels first.

brucek

[norliss]

brucek wrote:

So, this is the subwoofer without the mains, or the mains without the subwoofer, or subwoofer + mains?

It was sub + mains.

I've just been running some more sweeps so will post some more graphs if that's ok?

Thanks for your help.

[brucek]

The different graphs relate to various gain levels and crossovers.

It was sub + mains.

Huh. Since the extension of the sub is changing quite a bit with a change in crossover (normally wouldn't occur), it can only mean that your mains are quite capable and the sub is quite challenged. Tell us what type of mains and sub.

I've just been running some more sweeps so will post

Be sure to include what we're looking.

brucek

[Wayne A. Pflughaupt]

norliss wrote:

Apologies. This was the "subwoofer" setting. I did 2 512k sweeps from 0-200KHz. The different graphs relate to various gain levels and crossovers.

Strange, it looks like just the opposite, i.e. the mains though a high pass. Note how the bass below 100 Hz rolls out more severely with each graph...

Regards,
Wayne

[brucek]

Note how the bass below 100 Hz rolls out more severely with each graph...

Exactly, and that's the reason behing my comment above of : Since the extension of the sub is changing quite a bit with a change in crossover (normally wouldn't occur), it can only mean that your mains are quite capable and the sub is quite challenged.

As he raises the crossover, the mains don't support the sub as much and the level drops. The sub simply isn't adding that much to the mix. A more capable sub is required..

brucek

[norliss]

Hi again guys.

The sub in question is a BK XLS300/PR (I can't post a url yet). It is well regarded this side of the atlantic and certainly not something that lacks ooph.

The sub is next to the seating position and I didn't want to be able to localise the bass, hence my using a fairly low gain level setting.

I'll post some more graphs very soon with supporting info which will hopefully be easier to make sense of!

Many thanks again for your time.

[norliss]

Ok here are some more graphs. For reference, the dials on the sub have 10 "marks" and these images basically span from a gain level of 3, 3.5 & 4, with crossover levels (where 0 is 40hz and 10 would be 120hz) of 3-5 and phase ranging from 3-4 (mark 0 would be 0degrees and 10 being 180degrees).

The gain level increases from top-to-bottom. Sorry I couldn't individually mark-up the graphs (although the file names are indicative). I suppose advice might be to increase the gain levels and I'll happily do this if I can avoid localizing the sub?

[brucek]

sub in question is a BK XLS300/PR (I can't post a url yet). It is well regarded this side of the atlantic and certainly not something that lacks ooph

Perhaps not, but in your situation it isn't doing its job, and maybe a result of having the volume to low. The spec on the BK XLS300/PR is 20Hz -3dB. You aren't getting that. Your best extension is about 25Hz. It should be better. Locate the sub in a corner and see if you can't excite your room a bit more. And turn it up.....

As long as your crossover is 80Hz or less, you won't localize the sub.

BTW, the Measured tab in REW is for overlaying and comparing graphs - rather than posting multiple plots, one plot would make it easier on you.....

brucek

[norliss]

brucek wrote:

Perhaps not, but in your situation it isn't doing its job, and maybe a result of having the volume to low. The spec on the BK XLS300/PR is 20Hz -3dB. You aren't getting that. Your best extension is about 25Hz. It should be better. Locate the sub in a corner and see if you can't excite your room a bit more. And turn it up.....

It kind of is in a corner: well, by a recess but turn it up I will!

brucek wrote:

As long as your crossover is 80Hz or less, you won't localize the sub.

Well that will be good

brucek wrote:

BTW, the Measured tab in REW is for overlaying and comparing graphs - rather than posting multiple plots, one plot would make it easier on you.....

I couldn't find the embarrassed smilie but thanks for that I will bear that in mind!

Many thanks for your input. Another little round of REW'ing beckons

[norliss]

Ok I've just done another couple of sweeps (this time measuring at 70db).

This time I've left the phase and crossover the same but increased the gain to 5 (the midway point on the dial) and then 6. What sort of levels should I be looking at for the lower frequencies?

[norliss]

And some more with the gain turned up to 6.5. Each of these has the crossover changed slightly.

[Wayne A. Pflughaupt]

These graphs all look like they have smoothing applied. Have they?

Regards,
Wayne

[brucek]

Well, your room is certainly cooperating, as you have no real problem modal peaks with a smooth response and crossover region, but your low end extension isn't what I would expect from that sub. It's stopping quite consistently at 25Hz rather than 20Hz ........ other than that the response is quite good.

brucek

[norliss]

Wayne A. Pflughaupt wrote:

These graphs all look like they have smoothing applied. Have they?

Yes, 1/3 octave smoothing.

[norliss]

brucek wrote:

Well, your room is certainly cooperating, as you have no real problem modal peaks with a smooth response and crossover region, but your low end extension isn't what I would expect from that sub. It's stopping quite consistently at 25Hz rather than 20Hz ........ other than that the response is quite good.

Thanks for your feedback.

I just watched a Blu-ray disc "The Host" and it literally shook the room so much so I had to take the sub gain level down.

This is my first sub and I've never sat in anyone else's home cinema to have any kind of benchmark but that just seemed too much for me. Even at the lower gain level I could feel a serious "thud" everytime the creature's feet hit the floor!

As for the low-end thing I wonder whether it's a room thing? I must say I did notice when calibrating the soundcard that the curve tailed off considerably at the low end so perhaps I'll need to try again with a different laptop?

Also, I feel very conscious of the location of the sub i.e. right next to me: something I thought I could calibrate out. A bit disappointing really given that it really can't go anywhere else.

Thanks again for all help.

[brucek]

Even at the lower gain level I could feel a serious "thud" everytime the creature's feet hit the floor!

Yes, that's the desired result.

As for the low-end thing I wonder whether it's a room thing? I must say I did notice when calibrating the soundcard that the curve tailed off considerably at the low end

Yes, that's normal. Before you removed the loopback cable used in the soundcard calibration, you were advised to take a measure of the loopback cable with the newly created soundcard cal file in place. The result, of course, if the soundcard cal file was valid, would be a perfectly flat line. If you did this and got the desired result, then that channel is calibrated and should be trusted as flat.

conscious of the location of the sub i.e. right next to me

Yeah, when a sub is super close, its proximity can make you sensitive to the soundwave. It would be nice if you could get it nearer the source.

brucek

[Wayne A. Pflughaupt]

norliss wrote:

Yes, 1/3 octave smoothing.

We typically suggest no soothing for subwoofer graphs. Just prepare you, that'll make it look worse than it does now.

I feel very conscious of the location of the sub i.e. right next to me: something I thought I could calibrate out. A bit disappointing really given that it really can't go anywhere else.

At least make sure you have delayed appropriately.

Regards,
Wayne