|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
208.104.235.145
I've been working on integrating a Velodyne Optimum-8 sub with my NHT Classic Three's in a very difficult room. Please take a look and let me know your thoughts. I'm not done playing but this is as good as I could get tonight.The blue line is in room at listening position with no sub
The red line is 0-phase w/ 40hz crossover
The green line is 270-phase w/ 40hz crossover
All measurements have been corrected for the errors in the RatShack SPL meter.
The "green-line" measurement is yielding pretty good sounding results right now.
Edits: 01/12/09Follow Ups:
We have a saying in Pro Audio. Flat is where it's at. With the exception of some high-mid frequencies( which the human 'hears' better than others) that I prefer attenuated slightly, this rule of ear holds true.
I think you answered your own question with regards to your green line comment.
I played around some more this evening. This is as good as I think I'm going to get it - Phase 270 and volume @ 15 seems to get me as good as I'm going to get. The thick blue line is no sub, the dashed orange line is what I settled on (40hz cross -as low as it will go- and 270 phase and volume @15).
Don't get me wrong - I prefer how it sounds with the sub than without, it fills in the kick-drum and string bass on my jazz and rock very nicely without seeming to impact the sound of the Three's in a negative way, which was the aim of integrating it. My room situation is temporary (a 14x14 square room with lots of furniture in it), thank goodness, and I'll have an 11x16x9 dedicated room in the near future.
I guess what I'm looking at is input on how bad or good the dashed 270 line is on the graph. I understand, Richard, that the method I'm using is not the best but it's what I've got available to me (a RatShack meter and a Stereophile CD).
Just guessing from the charts that the overall sound quality would be better without the subwoofer.
Your room has a strong resonance around 40Hz.
Using a 40Hz. crosover, with the subwoofer phase control set 180 degrees out of phase, should help.
The 270 degree chart does just what I would expect with the phase control set at 180 degrees (subwoofer crossover frequency and phase controls are notoriuosly inaccurate versus what's printed on the dials!)
You really need parametric EQ to eliminate the 40Hz. boom and much less subwoofer output over 50Hz. -- a 24dB/octave low pass filter rather than the current 12dB/octave filter (my guess)
.
.
.
Richard BassNut Greene.....................................................................
The "Cliff Claven" of Audio
and the "The Floyd R. Turbo of Bingham Farms Michigan"
It's about 11x16x9 - we're doing an addition to the house and I'm planning on putting my gear in that room as a dedicated listening room.
I do have a decision to make as to if I want to do a flat ceiling or a... well - not sure what to call it?!? ... a ceiling with vaulted sides up to a flat area - like this /---\ instead of this |---|. My inclination is the semi-vaulted (???) to avoid corners but I need to talk to a professional about it I suppose.
There may be nothing wrong with 11'w x 16'l x 9'h but if you were able to use the magic Fibonacci ratio (1 : 1.618 : 2.618), you couldn't get much better than this as a basic room.
Re. the ceiling ... IMO the vaulted ceiling you are thinking of will deliver better sound than a flat 9' ceiling and if you can go to a peaked ceiling /\, that will sound even better (that's what I designed my "listening room" with, 22 years ago).
If you send me a PM with your email address, I'll send you an Excel s/sheet which originally came out of THX, which gives you graphical feedback in terms of the room nodes intrinsic to the selected dimensions. Then you can experiment.
Regards,
Andy
It's an additon - I might be able to expand the width (11' wall) by a bit - but not more than a foot or so. I'm pretty much locked in on the dimensions. I also can't do a full vaulted ceiling but do have room to do a "semi-vaulted" which is what I'm leaning towards.
"I prefer how it sounds with the sub than without, it fills in the kick-drum and string bass on my jazz and rock very nicely without seeming to impact the sound of the Three's in a negative way, which was the aim of integrating it."
How it sounds overall is what really matters. You seem to be getting peaks that would coincide with what you're hearing. Perhaps a little too good for my taste. Whatever you decide, I believe you'll find inching the voluume down a bit on the sub will benefit you greatly.
Well, The 270 phase 15 curve is slightly worse overall than your earlier 270 phase 17 curve. The new curve is =/- a little more than 4 dB each way but the previous curve was roughly the same in the peaks but the main dip was a little shallower. There's not much between them but I'd call that very slightly in favour of the earlier curve.
I'm still in favour of the no sub trace, however. Yes, it is down a bit on the others when you go below 100 Hz but it has a 1 dB peak and is only 4 dB down at 31.5 Hz. The sub is filling in the dips, certainly, but it's also raising the peaks and the sub traces are not in as tight an envelope overall as the no sub trace down to 31.5 Hz. If you take in the further dip to -7 dB at 25 Hz, the no sub trace falls within an 8 dB envelope while the 270 phase 15 trace falls within a roughly 8.5 dB envelope and the 270 phase 17 trace from your earlier post falls within close to an 8 dB envelope.
I can understand the attraction of the traces with the sub because, as I said, they do fill in the dips but the peaks get pushed much higher also. The total variation doesn't significantly change, both peaks and dips are simply being raised somewhat.
I don't think you realise how good that plot without a sub actually looks. Dips no lower than -4 dB down to 31.5 Hz is extremely good and that peak of +1 dB is even better. Keeping things within a 5 dB envelope down to 31.5 Hz is extremely good response. Provided you don't have traces that are significantly elevated above 1 kHz to make that natural bass response sound a little too light, that's the sort of result many people would love to see in their rooms even with a sub. Yes, it will sound lighter relative to the sound with the sub but it shouldn't be sounding excessively light. In my view it should probably sound very close to just right.
And my experience of adding a sub to my audio system and then shifting it to my separate HT system taught me one thing. That was that it's hard to beat the overall transient response you get without the sub. Subs are usually placed further away than the mains and their active circuitry plus the additional distance means the sound from the sub arrives a few ms after the direct sound. That delay in my system certainly wasn't enough to make me notice the sub as a separate source but it certainly took the crispness of transients in the bass range. That's not a problem in the HT system which compensates for the delay from the sub by delaying the sound from the other speakers. In fact it was the gain in crispness in bass transients in the audio system when I removed the sub that helped persuade me to leave it in the HT system and to run without the sub in the audio system.
If you're getting that good a result without the sub, then I think that you can probably improve on that by paying attention to speaker and listening position placement. I'd be prepared to bet that you can get close to the kind of response you're seeing with the sub just by working your speaker and listening position setup, and you could improve it some more with some bass trapping, and you won't have to worry about the sub's placement and the affect of the slight delay in the arrival time of sound from the sub.
OK, so doing it my way requires work and all you have to do is keep the current sub settings which seem to be making you happy but while you won't get quite the same bass "weight" without the sub, you're losing very little in overall extension without the sub and I think you'll get a crisper result without it.p plus slightly clearer sound.
Don't get me wrong, I've got nothing against subs and I wouldn't do without one in my HT system, but with the results you're getting on that trace without the sub in your system, which looks better than the kind of trace I get in my audio system, I know I'd prefer to run without the sub. In fact I am, with results that aren't quite as good as yours, but which still sound preferable to my ears to the results I used to get with the sub in my audio system.
The one thing that might change my mind would be your preferred choice of music. If that's large scale symphonic works or pop with very strong and deep bass parts, I might go for the sub but for smaller scale groups, especially with acoustic instruments or with not much bass content below 35-40 Hz which is what makes up the most of what I listen to, I'd certainly run without the sub.
David Aiken
I'm learning so much as I go through this process and it will help me so much when I have a dedicated room later in the year.
Thanks a ton for the input. My musical selections do tend to favor rock and jazz - both of which benefit greatly from the added "weight" of the sub.
Just a quick note for now, I'll post more this evening.The 90 and 180 graphs showed much larger peaks than the 270 which is why I only showed the 270.
Understand, I've never done anything like this before so it's all new to me. I'm temporarily in a room that is probably a worst case scenario and am just trying to get it as good as possible
I'll post more this evening including graphs of the other phases.
Edits: 01/13/09 01/13/09
YOU WROTE:
"The blue line is in room at listening position with no sub
The red line is 0-phase w/ 40hz crossover
The green line is 270-phase w/ 40hz crossover"RG
Did you try 180 degrees out of phase?The simple test is listening to a slow sinewave sweep from 20 to 100Hz. that lasts at least 15 seconds. You'll easily hear bass booms and nulls.
Your measurements should be made with sinewave test tones spaced no more than 1/6 octave apart, and even that would result in a rough measurements.
Your current measurements using some test tones 1/3 octave apart are almost certainly NOT capturing the actual SPL peaks and troughs that would be visible with sinewave tones for each frequency: (20Hz. 21Hz. 22Hz. ... 99Hz. 100Hz.)
The test tones should be steady sinewaves. Sinewave warble tones, pink noise and white noise will smooth the measurements by failing to excite room modes much (unlike bass guitar notes and kick drum notes)
YOU WROTE:
"All measurements have been corrected for the errors in the RatShack SPL meter."
RG
The Rat Shack meter needs no correction for the bass frequencies.Most of the so-called "error corrections" are in fact a simple conversion from SPL C-Weighting to a flat SPL weighting (unweighted SPL) that would apply to all sound meters.
Rat Shack meter samples are also not consistent enough to be sure corrections for someone else's Rat Shack meter would be useful for yours.
Most Rat Shack meters read a little low for bass frequencies, so you could add +1dB to all measurements under 100Hz. and most likely inprove the accuracy.
Measurements under 30Hz. are very likely to be too low, but those frequencies are hard to hear with music playing at typical volumes, so the low readings under 30Hz. may actually help the meter correlate better with what you hear/feel at 20Hz.
Richard BassNut Greene
"The Floyd R. Turbo of Bingham Farms Michigan"
Edits: 01/14/09
it with some sweeps if they're on the Sterophile CD's I borrowed.Here are the correction tables I used in the link, it had me add +5db @25hz! This is the same table I've found all over the net when searching "Radio Shack SPL Meter Corrections":
Edits: 01/13/09
It's the treble readings over 2000Hz. that are really far from accurate, and should be ignored! Much worse than the treble error corrections shown at the link (unless the meter has been redesigned recently).The Rat Shack meter uses C-weighting.
C-weighting is very useful for average SPL's up to about 90dBA
90dBA is very loud.
Our hearing doesn't work well for bass tones unless the bass is very loud.
That's why for most listeners (who don't listen to music averaging over 90dBA), C-weighting is the best measurement to use
C-weighting assumes bass at 20Hz., for one example, will need to be +6.2dB louder than 1000Hz. to sound equally loud, because our ears have trouble hearing deep bass.
Therefore, any sound meter using C-weighting will read 6.2dB LOW at 20Hz. compared with an unweighted SPL measurement.
So, FOR ANY SOUND METER AT ANY PRICE, the conversion factor from C-weighting to an unweighted measurement at 20Hz. is +6.2dB.
It's not fair to combine a +6.2dB conversion factor that applies to ALL sound meters, with a +1.3dB error correction that applies to one sample of a Rat Shack sound meter, and then claim Rat Shack meters need a +7.5dB "error correction" at 20Hz.
That falsely makes the Rat Shack meter look bad.
In reality, a Rat Shack meter's C-weighted readings should be +/-2dB in the 30 to 100Hz. range ... and merely adding +1dB to the readings is likely to improve the accuracy to +/-1dB = very good for a meter that doesn't cost $300-$500.
Note: Measurements of sinewaves under 30Hz. may not be useful because a big portion of the sound energy may be harmonic distrortion (aka doubling) at higher frequencies which would just make a subwoofer look more powerful than it really is at 20Hz., for example.
.
.
.Richard BassNut Greene.....................................................................
The "Cliff Claven" of Audio
and the "The Floyd R. Turbo of Bingham Farms Michigan"
Edits: 01/14/09
... and I realize (and understand) it's not the most useful way of measuring, but I just switched the correction to a straight +1db below 100hz.
Just to be clear too - the sub is not corner loaded, etc. it's sitting just to the right of the left speaker. I have no choice in placement in the temporary room I'm in!
The C-weighting (+1dB for typical Rat Shack meter errors) is a decent measurement of what we hear at reasonable volumes.
At very high volumes (over 90dBA), an unweighted SPL measurement is better.
I think you said your room was 14' by 14'
The lowest room mode (standing wave) would be about 40Hz. (565/14ft)
The "modal zone" would be above 40Hz. = no standing waves below 40Hz.
Unless your room was unusually well sealed (rare), there would be no cabin gain below 40Hz., and placement in a corner would not affect the bass below 40Hz. For those huge wavelengths, any spot in the room would be "near a corner".
Your small room should provide smooth bass under 40Hz. ... but a rough bass frequency response for a few octaves above 40Hz. based only on the dimensions (aka armchair analysis).
.
.
.
Richard BassNut Greene.....................................................................
The "Cliff Claven" of Audio
and the "The Floyd R. Turbo of Bingham Farms Michigan"
All you need is a soundcard, Radio Shack meter or Mic/preamp, misc cabling, and this free program, lots of folks use it for what you are trying to do:
.
It appears to me that your sub is generating substantial output above the nominal crossover point and that it is exciting room modes at 100Hz and at 50 Hz. Your room is fairly small, right?
In addition, it appears that the output of the sub itself falls off pretty quickly below about 35 Hz. In fact, I wonder hot flat this sub's response really is, even within its operating range. My guess is that it has a peak of its own between 40 and 50 Hz -- not unexpected with a small sub that has limited bass extension. The peak will make buyers feel like they're getting something for their money.
I think your setting is about as good as you're going to get (assuming the crossover point can not be adjusted below an indicated 40 Hz) without resorting to the use of an expensive parametric equalizer like what is available from Behringer. A 4 db bump is not too bad, really.
It's a pity you didn't extend the without sub measurements to 1 kHz as well.
Over the range from 31.5 Hz o 160 Hz, I actually think the without sub trace is within a tighter envelope but I can't compare the 160 Hz level to 1 kHz so that may not be the case. If all 3 traces were normalised on 1 kHz and the blue trace remained where it currently is, I'd opt for running without a sub unless you listen to a lot of music with significant content below 40 Hz. If that were the case, then the green trace for the sub looks best.
It would also help if the plots ran in the opposite direction with 20 Hz on the left hand side as is the norm. It took me a while to realise I was reading the traces in the wrong direction.
David Aiken
I'm new to this, but it looks as though you have a 50Hz peak with a secondary harmonic at 100Hz on your 270 degree phase line. If you like that line, then a sub with one band of EQ cenered at 50Hz might be all you need for about as smooth a line as you're going to get. Might take care of that 100 Hz somewhat too. If your sub doesn't have EQ, then look into the Behringer Feedback Destroyer 1124P. It has more EQ than you need for this purpose, but may be as cheaply as you can go to add EQ to a sub anyway. It works for me.
Why did you try/measure only those two phase points? What happens in between, say at 90 and 180?
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: