Not happy with flat response (1 Viewer)

[ClintonM]

About two weeks ago I finished my sealed Q=.7 Shiva subwoofer. My goal all along was to create an enclosure that would give me flat response in a particular corner of my family room. After living with it for a while I'm not totally happy with the output. Don't get me wrong, it's a huge improvement over my old subwoofer which couldn't get below 35Hz, but it just seems like something is missing. It's as if there is only low bass or no bass, if that makes sense. My Harman Kardon receiver crosses over at 100Hz and it seems like the subwoofer doesn't have oomph below that until it gets to about 30Hz. Could it be that my Shiva isn't broken in yet? Or is it something I'll just get used to?
Thanks,
Clinton Myers

 

[Fred Seger]

what are you using to power the sub? A plate amp? if so do you have the crossover on that set right.
Is the lack of bass just from subjective listening or are you using a SPL meter?

 

[Neil Joseph]

What is it tuned to? I am thinking of the possibility that it is tuned too low for the driver/enclosure used and gain was sacrificed for low output.
------------------
Link Removed
------------------
My Favourite DVD's

 

[Dave Poehlman]

I had the same sort of experience when I finished my Shiva. The bass seemed too overpowering and disconnected from the rest of the audio. After buying Stryke's test CD and a Radio Shack sound level meter, I was able to get things reasonably flat.
I was surprised how far off my "earballing" was.

 

[ClintonM]

I'm using a 120 watt plate amp from my old KLH subwoofer (don't laugh to hard). I have the crossover on the amp set at the highest point, I think around 160Hz, thinking that it would be better to have the receiver set the crossover point. The gain on the amp is turned up one fourth of the way. Up until now I've been using my ears which haven't let me down. Regardless, I do have a Radio Shack SPL meter and test tone CD so I'll measure and see if anything jumps out at me.
Thanks again.

 

[Fred Seger]

Alot of plate amps that are in manufactered subs have bass boost built into them for the specific drivers they use. My guess is your KLH has a lot of eq down low and that might be the problem.

 

[Dave Poehlman]

Another thing I just thought of, the amp you are using may be designed with a level boost at some point. You may want to check with the manufacturer. Do you have a model number on that amp?
(oops! I see Fred beat me to that idea... sheesh, you get up to use the can in the middle of a post and someone steals your thunder

)
[Edited last by Dave Poehlman on November 05, 2001 at 03:29 PM]

 

[Greg_R]

Even worse, the KLH amp may have a hardwired LF cutoff built into the amp (i.e. remove all signals below 30Hz). I'd test you sub with a different amp if possible and see if that improves your situation...
Greg

 

[Wayne A. Pflughaupt]

My goal all along was to create an enclosure that would give me flat response. After living with it for a while I'm not totally happy with the output. It's as if there is only low bass or no bass, if that makes sense.

Clinton, this does not sound like flat response. Believe it or not, flat response usually sounds “thin” or lacking in deep bass.
Just curious, how did you determine the response is flat? Have you taken in-room measurements or are you assuming, based on modeling data?
If you post your data, it will be easier to have some idea of what is wrong.
Regards,
Wayne A. Pflughaupt
------------------
My Equipment List

 

[ClintonM]

This morning before work I measured the in-room response from 200Hz to 15Hz. There is a big dip at 140Hz and 90Hz. The 90Hz dip is probably what's bugging me. Here it is...
(Hz)-(SPL)
015-66
020-77
025-90
030-87
035-88
040-92
045-93
050-91
055-91
060-84
065-83
070-94
075-96
080-95
085-90
090-70
095-86
100-87
110-78
120-82
130-80
140-66
150-84
160-84
170-84
180-85
190-90
200-93
I remember hearing about "corrected values" for the Radio Shack meter. Were can I get those?

 

[Wayne A. Pflughaupt]

Forgot to ask, what is your crossover frequency?
Regards,
Wayne A. Pflughaupt
[Edited last by Wayne A. Pflughaupt on November 06, 2001 at 09:40 AM]

 

[ClintonM]

Wayne, the receiver's x-over frequency is 100Hz.

 

[Patrick Sun]

If you measure your sub at a different place, do the dips migrate to other frequencies? (I'm suggesting you have a room null situation).
------------------
PatCave; HT Pix; Gear; DIY Mains; DIY CC; Sunosub I + II + III; DVDs; Link Removed

 

[Wayne A. Pflughaupt]

I found that—I should have checked your original post first. Sorry.quote: My Harman Kardon receiver crosses over at 100Hz and it seems like the subwoofer doesn't have oomph below that until it gets to about 30Hz.[/quote]Clinton, the measured data certainly does not reflect your seat-of-the-pants observations. Did you take the measurements from the primary listening position?
The 90Hz null is very severe yet very narrow, so I’m guessing it is cancellation between two speakers. I wouldn’t worry about it. Based on the measurements, your biggest problems are the two response spikes, +12dB centered at about 77.5dB, and +9dB centered at about 45Hz. These are both pretty substantial. The first place to start is eliminating them with a parametric equalizer.
In any event, response this strong to 25Hz is excellent. This is a very impressive sub, especially if the family room is the typical fare that opens to other rooms. This usually gives a total listening area of more than 5000 cubic ft., and it takes some serious “woofage” to deliver 25Hz in a space this large. Equalizing out the peaks will buy the driver some headroom, so most likely you could safely boost 20Hz several dB to further enhance extension.
Regards,
Wayne A. Pflughaupt
------------------
My Equipment List
[Edited last by Wayne A. Pflughaupt on November 06, 2001 at 09:42 AM]

 

[Allan F]

Wayne,
Ever consider leveling a home theater based on A or B weighting as opposed to C weighting? The C weighting scale evidently reflects the sensitivity of the human ear to various frequencies at high levels (over 85dBSPL). The next curve down (the B scale) reflects the sensitivity of the human ear to more moderate levels (btw 55 and 85 dBSPL). This curve is also noticeably less sensitive to lower frequency regions. I'm assuming, based on the above info, that calibrating a home theater with the SLM set for B weighting would result in more energy in the lower frequency region. Anyone ever done this? I wonder if we would get a more desirable "room curve." My SLM only offer A and C weighting options.
Allan

 

[BruceD]

Allan,
I think you need to take into account that the A weighting on the RS SPL meter's scale only measures freuqencies in the 500Hz to 10kHz range.
This means the bass below 500Hz will not register on the SPL meter's scale in the A mode, no matter how much bass is present.
This would not be an appropriate way to calibrate an HT system.
The C weighting scale is designed for music/HT measurement in the 20-20kHz range, even though the meter is not quite flat.
The reason for the A weighting scale is to measure the strength (loudness) of frequencies that can damage the human ear and then calculate the time exposed to that level of SPL.
This would be a smart thing to test for when listening to your favorite movie at your preferred volume level.
BruceD
[Edited last by BruceD on November 07, 2001 at 11:42 AM]

 

[ClintonM]

Thanks guys! I'll be looking for a parametric equalizer now. That should do the trick.

 

[Allan F]

Bruce,
I think a couple of your points need some clarification.
"I think that you need to take into account that the A weighting on the RS SPL meter's scale only measures frequencies in the 500Hz to 10kHz range."
The A weighting scale is not at all limited to a frequency range of .5-10kHz. The A weighting scale is simply more sensitive to this frequency region ie.. a steep rolloff above and especially below that range. I have not taken the time to confirm this but even the manual to the RS SPL meter does not indicate this.
"This means the bass below 500Hz will not register on the SPL meter's scale in the A mode, no matter how much bass is present."
I seriously doubt this to be the case.
"This would not be an appropriate way to calibrate a home theater system."
I wasn't intending to imply that. I simply inquired into the possibility of calibrating based on a B scale. This weighting scale rolls off closer to 150Hz.
"The C weighting scale is designed for music/HT measurement in the 20-20kHz range, even though the meter is not quite flat."
I'm quite confident the C weighting scale was around a long time before home theater or even home music reproduction. The C weighting scale represents the human ear's sensitivity to various frequencies at high input levels. This scale represents a 100 phon contour. The response curve of the human ear is level dependent.
"The reason for the A weighting scale is to measure the strength (loudness) of frequencies that can damage the human ear and then calculate the time exposed to that level of SPL."
This is typically what the A scale is used for but not why it was developed. The human cochlea (our primary organ of hearing) has kindof a built in defense mechanism to low frequency sounds and that is why we are notably less sensitive to them. They simply are not important for communication.
Allan

 

[BruceD]

Allan,
You may have missed the point entirely.
We were discussing how to take advantage of the RS SPL meter capabilities not just various weighting scales. The B weighting scale you discuss is not available on the RS SPL meter.
your quote:
----------------------------------------------------------
The A weighting scale is not at all limited to a frequency range of .5-10kHz. The A weighting scale is simply more sensitive to this frequency region ie.. a steep rolloff above and especially below that range. I have not taken the time to confirm this but even the manual to the RS SPL meter does not indicate this.
----------------------------------------------------------
Well I have both read the manual and tested the response.
Manual says: "When set to A, the meter primarily measures frequencies in the 500 to 10,000 Hz range". I have also tested and found little response (-20dB level is a pretty substantial reduction) from the meter on bass frequencies at high SPLs.
An example: when testing bass warble tones 200Hz-20Hz the SPL meter readings were as follows:
freq----C weight----A weight
200Hz-----88------69---------(almost 20dB less)
160Hz-----94------73---------(almost 20dB less)
125Hz-----90------66---------(more than 20dB less)
100Hz-----90------60---------(more than 20dB less)
your quote:
----------------------------------------------------------
"This means the bass below 500Hz will not register on the SPL meter's scale in the A mode, no matter how much bass is present."
I seriously doubt this to be the case.
----------------------------------------------------------
Do you actually have some experience measuring with the RS SPL meter in A mode? I do. See my results above.
your quote:
-----------------------------------------------------------
I wasn't intending to imply that. I simply inquired into the possibility of calibrating based on a B scale. This weighting scale rolls off closer to 150Hz.
-----------------------------------------------------------
The B weighting scale is not available on the RS SPL meter. Do you have a source for an inexpensive SPL meter with a B weighting scale? That would be a helpful addition to this discussion.
As I stated, the A weighting scale is not approriate for any calibration duties in a HT. It is useful for protecting your ears.
I don't want any HT enthusiasts here to get the wrong idea about how to use the RS SPL meter in making their HT a more enjoyable experience.
BruceD