New Linkwitz "LXmini" speakers

Status
Not open for further replies.
Jump to Latest
#204
The only reference I could find was this on OPLUG...


Re: Is tweeter placement really critical?
Postby kjen » Wed Dec 03, 2008 9:26 am

Stumpy is an interesting proof of how robust the Pluto design principles are, in the crucial distances it differs less from the baseline than you at first think. It is the low crossover frequency the Aurasound tweeter makes possible that is crucial to these changes working and the speakers sounding so similar.

Here are some thoughts, based on Ray's drawing and a few assumptions.

Tweeter Position

Listening distance to woofer centerline 7ft, 2.15m

Pluto Tweeter

3 inches above woofer, same level as woofer virtual point source.

Front edge:

original 2.125 inches (54mm) forward of woofer centerline
modified 2.625 inches (67mm) forward of woofer centerline (for SEAS driver)
Signal delay 328 microseconds (100mm, 4 inches)

Assume the tweeter virtual source is 33mm in front of the tweeter front edge. (SL's design clearly places it somewhere between 33 and 45mm to get tweeter and woofer coincident.)

Stumpy Tweeter

7 inches (0.175m) below woofer virtual point source (4 inches below woofer rim)
Front edge 3.75 inches (95mm) forward of woofer centerline (estimated off drawing)

Then for Stumpy the tweeter virtual source is 95 + 33 - 100 = 28mm in front of woofer centerline and the path length from the listening position is 2.129m

Difference = +14mm, 0.55 inches, +46 microseconds. This is similar in magnitude to SL's change and probably not significant. If you wanted to be a perfectionist you could modify the all pass filter to reduce the delay to 282 microseconds, the filter spreadsheet can calculate the values.
 
#206
ttan98 said:
I did, my results are "similar" to SL. It depends how you mount the driver, the resonance may not be very large.
If I have some time I will do more measurements and post them here.
Click to expand...
If you fill the back part of the tube with foam rubber, and the area right behind the driver with that stringly white synthetic fiber fill cottony stuff (I forgot what Parts Express or Madisound called it), that should work great. A gaussian arrestor for the excess un-needed energy. I think that would give a cleaner cardioid pattern. An empty tube could be bad.
 
#207
As Linkwitz designed it, it's stuffed with fiber. The tube is not empty.

Bob Richards said:
If you fill the back part of the tube with foam rubber, and the area right behind the driver with that stringly white synthetic fiber fill cottony stuff (I forgot what Parts Express or Madisound called it), that should work great. A gaussian arrestor for the excess un-needed energy. I think that would give a cleaner cardioid pattern. An empty tube could be bad.
Click to expand...
 
#209
jakja83 said:
Interesting design, however, the minidsp 2x4 seems to be a weak spot in my view. Isn't going digital - analogue - digital - analogue potentially something that could degrade the sound?
Click to expand...

Have thought about using Minisharc, i.e. (XMOS/CM6631-->I2s) interface to (I2s-->Minisharc-->I2s) interface to up to 4 DACs. Digital signals right through. This has been done before but not cheap. It is getting cheaper as the prices of DAC and XMOS/CM6631 keep coming down.
 
#212
jakja83 said:
Interesting design, however, the minidsp 2x4 seems to be a weak spot in my view. Isn't going digital - analogue - digital - analogue potentially something that could degrade the sound?
Click to expand...

Try one....they're only 80 bucks.
Configure it with no processing and insert in your signal loop somewhere with a blind A/B bypass switch and see if you can identify if it's in the loop or not.

The essence of the LXmini design is not about the ADC/DSP/DAC processing though.......you understand this, yes?

Dave.
 
#213
Davey said:
Try one....they're only 80 bucks.
Configure it with no processing and insert in your signal loop somewhere with a blind A/B bypass switch and see if you can identify if it's in the loop or not.

The essence of the LXmini design is not about the ADC/DSP/DAC processing though.......you understand this, yes?

Dave.
Click to expand...

Of course, I was just, in a very secluded fashion, asking the question:

Is there anything to be gained opting for 4x10HD instead?
 
#214
Moderately priced DSPs do seem to split opinion. Many have said that degradation is either imperceptable or at least very slight. OTOH there are those who assert serious sound quality issues. The relatively small number of actual builders that have reported so far seem pleased with the sound of the complete speaker/ digital XO system. The 4x10 is technically superior and does allow integration of some optional modules, as well as more channels that could be used for subwoofers.
 
#217
Davey said:
Features, versatility and higher voltage output. However, if the miniDSP 2x4 unit will be dedicated to LXmini operation, you have no desire for subwoofer, and your power amplifiers are suitable voltage gain then it's worth saving the $420.00. IMHO.

Dave.
Click to expand...

Ok. Yes, I was hoping for a very simple setup that I could migrate around the house easily.

Some claim there is some noise on the 2x4's outputs though and I loathe that, what is your experience?
 
Last edited:
#218
Adhanda said:
I hear that the "virtual source" of a driver is at some distance in front of it. I searched around but could only find the term "virtual source" used with reference to driver arrays, not for a single driver. I'm wondering how to calculate it for a single driver.

I thought TA was calculated from the mounting flange? How did you determine the horizontal mounting distance of the Sanyo from the axis of the Akai?

I assume SL intended to TA the virtual sources of the drivers in the LXmini, but the distance of the virtual source of the HF driver seems to be out of proportion with that of the LF - 1/4" to 3".,,unless there is some TA going on in the miniDSP...?

Dan
Click to expand...

I don't think that the audio world paid much attention to this aspect untill Stanley Lipschitz wrote a paper published in the AES Journal within the last 3 or 3 years. He found that by tracing the pressure gradients around a sealed box woofer they converged at a point around 300 odd mm in front of the driver. This probably only applies to sealed boxes and probably not dipoles. The displacement of the acoustic centre may be frequency dependant, and may vary with box dimensions so there is unlikely to be a figure that applies to any particular driver.

I have the paper but not readily to hand. It is copywrited to the AES so I cannot post it but could have a read of it if you are interested in more details.

The thing about speakers playing on You Tube videos is that you are hearing what the microphone(s) picked up which can be quite different to being in the room where the processing between the ears can discriminate between direct and reflected sounds. As others have remarked, comb filtering as measured, is largely inaudible to a pair of human ears and some brain processing.

Keith
 
#220
I think there's a 2x8 miniDSP also, cheaper than the the 2x10 but not in a box
http://www.minidsp.com/images/documents/Product Brief-miniDSP 2x8 Kit.pdf

It appears to me that the above models have 2v output (4v balanced) so should have enough output to drive even low gain amps such as Pass
models, to full output:
http://www.minidsp.com/images/documents/Product Brief-miniDSP 2x8 Kit.pdf
article:
http://www.hifizine.com/2011/06/tour-of-the-minidsp-2x8/




I believe there are also miniDSP products that don't have DACs built in so you can use your own (at possibly considerable expense as you need 4ch).

This for instance,

http://www.minidsp.com/products/minidsp-in-a-box/nanodigi-2x8-b
http://www.minidsp.com/products/minidspkits/nanodigi-2x8-k

Not sure how interchangeable the software and the LXmini files are between models.

I'm just looking at the spec sheets, I'd really appreciate Davey clarifying all this if he is able
 
Last edited:
Status
Not open for further replies.
Top Bottom