What is the ideal directivity pattern for stereo speakers?

Status
Not open for further replies.
Jump to Latest
#1,603
Thanks Markus. :up: Turns out I had it, but had not installed it. Will do so and learn to use it.
BTW, what did yo do for your noise measurement? Did you run a pink noise measurement then do a second one with the amp turned off or down?
 
#1,604
DBMandrake said:
Wouldn't that only determine the direct to reflected ratio at high frequencies though due to the small gate time ?

I've used the second part of that trick before - gate out as much of the direct impulse as you can by setting the start marker just before the first early reflection and no end marker and you more or less get the room response without the direct signal.

The problem is always in obtaining the direct signal (down to a low frequency) without interference from the reflections...
Click to expand...

Yes, the free window before the first reflection will determine to what frequency you could get seperation of direct and reflected. Are you familiar with the KEF bass diminishing trick? Its the opposite of Lipshitz bass extending. Makes the impulse response decay faster.

Thanks for the impulse curves. They are about what I would have guessed.

Neat data so far. We'll have to see what the consensus is after a few more measurements.

David
 
#1,606
One interesting thing that I learned doing this is that there seems to be an optimum toe-in for my speakers.
Originally their center lines crossed well behind the listening position but I toed them in to cross right at the sofa seat for this. However, it's hard to get the mic over the sofa seat so it was placed right in front - about where my head would be if I were sitting of the edge of the sofa.

Now I find that the sound is best if I'm sitting where? On the edge of the seat, where the mic was. Need to figure out how to do the same toe-in angle back at the normal sitting position as it now is at the front edge of the sofa. Progress!
 
#1,607
Pano said:
One interesting thing that I learned doing this is that there seems to be an optimum toe-in for my speakers.
Originally their center lines crossed well behind the listening position but I toed them in to cross right at the sofa seat for this. However, it's hard to get the mic over the sofa seat so it was placed right in front - about where my head would be if I were sitting of the edge of the sofa.

Now I find that the sound is best if I'm sitting were? On the edge of the seat, where the mic was. Need to figure out how to same toe-in angle back at the normal sitting position as it now is at the front edge of the sofa. Progress!
Click to expand...
Although you've found the best toe-in for cancellation of out of phase left and right signals (correct me if I misunderstood this), I find that toeing the speakers such that the vector rays would cross just behind my head gives better stereo effect, perhaps because it reduces the interaural crosstalk a little bit. Maybe it's just that I've increased the ratio of room reflections to direct, which could create more of a sense of stereo effect laterally (?)
 
#1,608
Pano said:
There is a narrow notch in the response (about 1/12th octave) at 8.4Khz on my Altec 1005 horn + 288 driver combo.
Click to expand...

AFAIK it's a reflection in the large format driver's front chamber since it shows up on 31A, MR94B and at least one other 1005 when an alum. diaphragm with its extended HF is used.

Easy way to find out, measure the driver off the horn and high pass it around 5 kHz.

GM
 
#1,610
A couple of feet off the wall. Wish I had more room.
Humdinger. Yeah, that's what I'm finding. Crossed about 3 feet behind me should sound best. The original setup was crossed much farther back - speakers at a flatter angle.

GM. I did measure without the horn, no notch. The distance from the phase plug to the exit is a multiple of the notch wavelength.
 
#1,614
KSTR said:
Yeah, more than 10dB is not easily achieved, I learned too. Obviously I missed the center, as the diff gets even louder at the top end (graph is 1/6th-oct smoothing),

It has to be noted that this is a DSP'd system which helps for flat freq response and for part of the nulling.

- Klaus
Click to expand...

Hi Klaus,

Your time alignment could be improved a little. Your HF cancelation was weak and the blue impulse was still near full level. Either slide one speaker fore/aft slightly or sing the mic left/right until the top octave nulls as well as the lower octaves.

Nice flat direct curve!

David
 
Last edited:
#1,618
Well I've just tried the pink noise test. I'm not sure what to make of the result... Note I haven't tried to find the best null position, just did some short bursts of pink noise with the mic in the listening position without adjusting anything. (It's after midnight and I should be in bed, didn't want to push it)

I used REW with the same settings as markus, but exported and displayed using Speaker Workshop since that is what I'm familiar with, and 1/3rd octave smoothed it.

I'm really not sure what to make of this result... especially the crossing after 6K. I'll hopefully have the new crossover done tomorrow so will be interesting to compare after that.

Tony.
 

Attachments

  • pink_noise_test.png
    pink_noise_test.png
    20.7 KB · Views: 203
Last edited:
#1,620
Pano said:
GM. I did measure without the horn, no notch. The distance from the phase plug to the exit is a multiple of the notch wavelength.
Click to expand...

OK, but it shows up on different horn designs, so if not an internal reflection, then it's most likely a reflection back to the beginning of the throat due to the mismatch at the mounting flange caused by Altec's thick cardboard gasket that I've periodically talked about.

GM
 
Status
Not open for further replies.
Top Bottom