Strange graph

[mra68]

Hi All,

I had good luck using REW to set up my previous sub over a year ago. I'm now using it (with the latest version) to set up a (new to me) used SVS PB12-plus/2. Please take a look at the graph and let me know why you think the response is dropping so low at 40 hz. The crossover on my receiver is set to 80 (I've also tried it at 100 with no real difference), the PEQ on my sub is disabled (and set to 120hz just in case) and I've tested it enabled and set to 40hz so it does work. I've tried several different locations in the room, and the curve looks very similar above 40hz in all of them.

Red is 25hz tune, green is 20hz tune, and blue is 16hz tune.

It almost looks like there's something wrong with the sub's amp or EQ. Any thoughts?

Thanks,
Alex

[Wayne A. Pflughaupt]

Well, it looks like the lower tuning is tilting the lower end up while depressing the upper end. That may well be the way it’s supposed to work, but the guys at SVS would be better at answering that. Would you like for me to move your thread to that Forum?

Regards,
Wayne

[mra68]

Sorry, I didn't realize there was an SVS specific forum, I've only surfed the REW forum here. I'll head over that way to do some reading.

Thanks,
Alex

Edit: Created a new thread there, in case anyone is interested:
http://www.hometheatershack.com/foru...-plus-2-a.html

[mra68]

Well that was useless, the mod of the SVS forum just moved me back here...

[JohnM]

You could eliminate any possible receiver issues by connecting your soundcard output directly to the sub's input. When you say you have tried different locations do you mean for the mic or the sub or both? If you've only been moving the mic may be worth trying a measurement with the sub moved.

[brucek]

The peak at 38Hz area is likely causing a lot of energy in your setup levels for REW and may result in the overall level of the plot being somewhat low in relation to the target.

So, if I overlay a target on your graph that has been dialed down a bit from 75dB, the graph to me looks like a large room peak at 38Hz and a dip at 50Hz. Fairly normal.

I would work on the peak first and then do some re-measuring. I think you'll find the response will be better.

You could also drag the sub to the middle of the room and do a near-field measure and see how it looks...


test1.jpg


brucek

[mra68]

JohnM wrote:

You could eliminate any possible receiver issues by connecting your soundcard output directly to the sub's input. When you say you have tried different locations do you mean for the mic or the sub or both? If you've only been moving the mic may be worth trying a measurement with the sub moved.

Ok, I'll try connecting it directly to the sub input later today. I'm also going to make a drastic placement change, from the front to the back of the room to see how that affects things. When I said I tried different locations, I meant that I had moved the sub to 3 different places in the front of the room. The mic has stayed fixed at my listening position.

[mra68]

Hi Brucek,

That seems plausible, however in my other thread on this topic that I posted in the SVS forum I tried using the built-in EQ to take the peak at 38hz. It seems (to my relatively untrained eye) to not have much of an effect. I can try reducing it even more and will take some more measurements, in conjunction with the ideas John gave above.

Edit: I should explain, this graph is the 16hz tune with the green line being no EQ, and the yellow line with some reduction of the peak at 38 hz.

Thanks,
Alex

[JohnM]

mra68 wrote:

Edit: I should explain, this graph is the 16hz tune with the green line being no EQ, and the yellow line with some reduction of the peak at 38 hz.

In that case the PEQ filter being used on the 38Hz peak is too wide (Q is too low).

[JohnM]

mra68 wrote:

When I said I tried different locations, I meant that I had moved the sub to 3 different places in the front of the room. The mic has stayed fixed at my listening position.

You should try other mic positions, even if it is impractical for you to change your listening position, as then you can see how much the listening position is contributing to the dips in the response.

[mra68]

Thanks for the tip on the Q, I wasn't clear on that one. The first image is the before and after with equalization on the PEQ of the sub. The second image is the resulting measurement with the curve. It still seems remarkably low below 45hz.

I moved the mic over 2 feet (from the sweet spot to a second listening position) and it looks remarkably different! See image 3 (blue line is new mic position, purple is previous eq'd measurement 2 feet to the right).

Thanks,
Alex

[Wayne A. Pflughaupt]

JohnM wrote:

In that case the PEQ filter being used on the 38Hz peak is too wide (Q is too low).

Unless SVS has changed their parametric EQs since the last time I looked into them (a year ago), wide bandwidths is all they’re capable of.

Alex, not sure I’m getting what the problem is at this point. When you asked in your opening post why response was “dropping so low at 40 Hz,” and presented the graph comparing tunings at 25, 20 and 16 Hz, I assumed you were wondering why 40 Hz went lower as the tuning dropped. Did I miss something?

Regards,
Wayne

[mra68]

Hi Wayne,

Indeed, I believe I have caused confusion. My intent with showing all three tunings was that no matter which tune I used, the levels are all quite high and loud in the low frequencies, crash from 40-50, and then continue to drop. This effectively causes my house to shake at the low frequencies (cool! ) but a loss of bass in the 45+hz ranges.

I did not have this experience with my previous sub in the same location, and am trying to ascertain if there's an issue with my sub, or if anyone has ideas to try to improve the response in my room. Thanks to the input so far I've made some progress, but still am wondering if there's something weird with my sub.

[Wayne A. Pflughaupt]

mra68 wrote:

I did not have this experience with my previous sub in the same location...

I’ll bet your old sub was um, kinda wimpy compared to this one?

...and am trying to ascertain if there's an issue with my sub, or if anyone has ideas to try to improve the response in my room.

You have an excellent and highly capable sub. Your problem is the same as every one else’s: the room.

As far as improving response, around here we typically recommend parametric equalization. Trust me, we’ve seen much worse response. At least yours is such that it could be equalized. Just make sure you don’t have the sub’s crossover enabled in addition to the one on the receiver. You might also try playing with the phase control.

Regards,
Wayne

[DrWho]

What kind of mic are you using for your measuring and is it calibrated and all that?

Do things sound like what you think you're measuring?

[mra68]

Hi Dr. Who,

I'm using the new analog Radioshack meter with the calibration file from the download page on this site. I do believe that I hear what the graph is reporting. For contrast, and in reply to Wayne's question about my previous sub, here's the graph from 7/2006 using an older version of REW on my Onix UFW-10. As you can see I had a similar peak at 38hz but above that did not experience as significant of a drop as the SVS is showing.

It's looking to me like it is indeed a room problem, so I'll continue to try different placement options.

Thanks for the continued feedback,
Alex