The Megaliths open baffle line array speakers

0 Members and 1 Guest are viewing this topic. Read 85197 times.

Atlplasma

  • Full Member
  • Posts: 963
  • Just off the boat
Re: The Megaliths open baffle line array speakers
« Reply #20 on: 2 May 2016, 02:04 pm »
HAL. This may seem like an odd question. But did you take an inspiration from the Siegfried Linkwitz's LX strategy? I built the LXmini a few months back and have really liked the sound this design delivers. It is very non-fatiguing. At the same time, I am a little dissatisfied by the 192/24 maximum resolution of the miniDSP 4x10. Was that a consideration for you in choosing the dspMusik 2/8? Thanks for sharing your build.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #21 on: 2 May 2016, 03:30 pm »
The only thing I am aware of is the transform for the servo subs.  More inspiration came from Danny's planar arrays I heard last year.  Wow is the best description.

The dspMusik uses better DACs than the other DSP XO's I have tried.  I like minimum phase filter DACs and this has them.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #22 on: 2 May 2016, 05:49 pm »
The nice thing about the dspMusik system is it's ability to develop generic speaker crossovers.  If the specifications for the digital crossover has been published, it can be developed on the dspMusik 6x8. 

The sound quality developed over a 4 year period of trials with improvements in the linear power supply regulation on the motherboards.  Each DAC has it's own set of regulators on the analog output stage for best sound quality.

mresseguie

  • Full Member
  • Posts: 4697
  • SW1X DAC+ D Sachs 300b + Daedalus Apollos = Heaven
Re: The Megaliths open baffle line array speakers
« Reply #23 on: 2 May 2016, 07:05 pm »
Rich,

I'm lurking in the background reading with interest. Don't have enough experience or knowledge to participate.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #24 on: 2 May 2016, 07:16 pm »
mresseguie and everyone,
Glad to see folks are watching.

Will keep posting the steps since there seems to be interest on progress.   Hope to make some serious progress this Saturday when the baffles are supposed to be ready to pickup.   :o

PDR

  • Full Member
  • Posts: 820
  • May the best man win
Re: The Megaliths open baffle line array speakers
« Reply #25 on: 2 May 2016, 07:25 pm »
Lurking here as well, very interesting..... :o

HAL

Re: The Megaliths open baffle line array speakers
« Reply #26 on: 3 May 2016, 12:03 am »
First version of the Megaliths digital crossover for the dspMusik is completed.  This is based on both the Pulsing Quasar beamformer and the BG Neo3-PDR/Neo10 open baffle designs completed earlier.  These both mated to the 6x12" H-Frame servo sub arrays, so that is known.  The gain set between the servo subs and the Neo10 arrays will need to be done from measurements. 

To measure the array, the mic needs to be at least the height of the array away from the planars.  The center listening axis is between planars 4 & 5, so that is the measurement axis.  The low frequency response of the measurement is determined by the free space measurement time before the first room reflection.  F = 1/t, where t is the free space time, and F is the lowest frequency of the measurement that is accurate.  This is the time gated frequency response measurement concept.

More once I have the baffle and get the Neo10's installed and wired to the amps hopefully this weekend.
 


HAL

Re: The Megaliths open baffle line array speakers
« Reply #27 on: 3 May 2016, 01:26 am »
As Avoosl has eluded to, the first inspiration for these speakers were when I heard the original Infinity IRS Reference III's in 1979 at an audio show in Virginia.  Those speakers to this day still set a reference standard that I have not heard matched. 

I figure if I am going to build a line array, it should be in the same range as those.  They have more power in the bass range, but these may have more finesse.  Only time will tell.


HAL

Re: The Megaliths open baffle line array speakers
« Reply #28 on: 4 May 2016, 10:23 am »
Have the Raspberry Pi3 running the latest Raspbian Linux OS.  The image contains Wolframs Mathmatica for computations.  Very powerful and flexible language in the vain of Mathcad and Matlab.  I used Mathcad for many years and this is very similar.

Since computations are needed to calculate the time delays needed for the array element drive, will attempt to make a small application to do this part of the design.  Costs less to do this on the RPi3 than getting the latest Mathcad for my PC.  Will see what happens.

Letitroll98

  • Volunteer
  • Posts: 5583
  • Too loud is just right
Re: The Megaliths open baffle line array speakers
« Reply #29 on: 4 May 2016, 11:44 am »
Lurking here as well, very interesting..... :o

Also watching your progress Rich.  Although I must admit the last couple of posts sound like, "you connect the rebostaters to the harbonators which mitigates the carboniferous interactions of the mu dependants."  Any chance you could explain it like you were talking to a ten year old?  Maybe a smart ten year old that just doesn't know anything.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #30 on: 4 May 2016, 11:53 am »
Letitroll98,
Will try harder to keep the explanations more basic.  The RPi3 thing is really just a fun use for the little computer.

Lets try this description and see if it is clear on the amp front for the speakers.

Normally to drive a segmented line array, the drivers have to be connected as series/parallel combinations to make an impedance that stock amplifiers can drive. When two drivers are connected in series, the sensitivity of the pair is the same as one driver with increased power handling capability. When two drivers are connected in parallel, the sensitivity goes up by 6dB for the pair. This is at the expense of the impedance now being 1/2 the original value.

If I were to take 8 - BG Neo10 planars and put them all in parallel, the sensitivity would go up by 18dB if they are all well matched drivers. In real life, it will be less than that due to differences. The problem is the new impedance for the amp to drive is 1.33 Ohms. Not something most amplifiers like to see as a load.

Now if each BG Neo10 that has a 6 Ohm impedance has it's own amplifier, then the sensitivity increase is the theoretical 18dB increase at the expense of 8 amps. With today's solid state Class AB amps, that is not a great expense and can have great sound quality. The amps only have to be capable of driving one driver to full potential, not the entire array. Instead of one 600w rms amp capable of 1.33 Ohm load drive, you have 8 - 75watt rms amps driving 6 ohm loads. That is the approach I am using.

S Clark

  • Full Member
  • Posts: 7356
  • a riot is the language of the unheard- Dr. King
Re: The Megaliths open baffle line array speakers
« Reply #31 on: 4 May 2016, 02:14 pm »
I'm right there with you Leti.  This video was made of Rich explaining server/dac interactions. 
https://www.youtube.com/watch?v=ss2hULhXf04

 :wink:

Actually, the mad scientist is pretty good at getting it down to my level.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #32 on: 4 May 2016, 04:42 pm »
LOL!   :thumb:

HAL

Re: The Megaliths open baffle line array speakers
« Reply #33 on: 4 May 2016, 05:36 pm »
Well at least this build is a bit simpler than this one must have been  :o



HAL

Re: The Megaliths open baffle line array speakers
« Reply #34 on: 5 May 2016, 05:53 pm »
Looks like the baffles will now be ready until some time next week.

Baffles have been in work over 1 year.  I hate waiting!!!

HAL

Re: The Megaliths open baffle line array speakers
« Reply #35 on: 8 May 2016, 11:38 am »
Good news, the baffles will be ready Wednesday and will pick them up.  Once home will install the 16 - BG Neo10 drivers.

Cables, 16 amps and dspMusik ready for trials.  The HAL MS-3 has MathAudio Room EQ software setup in Foobar2000 for measurements. Have a few more XLR balanced cables on the way to complete the setup.

This should be very interesting!  :D

HAL

Re: The Megaliths open baffle line array speakers
« Reply #36 on: 9 May 2016, 01:06 pm »
Trying out the Raspberry Pi 3 with the latest Raspbian OS, Wolfram Mathmatica is installed. 

Will be writing a notebook to calculate the time delays for an arbitrary number of drivers in a line array to calculate the data needed to use the dspMusik digital crossover at any listening position.  You just need the number of Neo10 drivers and the listening distance in feet.  The output will be the time delays needed for each channel of the digital crossover.  This is the data to put into the time delay blocks for Audio Weaver to run the speakers.

HAL

Re: The Megaliths open baffle line array speakers
« Reply #37 on: 10 May 2016, 06:45 pm »
Have most of the Mathmatica document done and getting the syntax correct since it is a bit different than Mathcad.

Decided to build insulated binding post setups for the amp modules.  Luckily PE was having a closeout on the large binding post and the simpler post connectors with the insulated plate, so was able to build all 38 binding post pairs. 



gab

  • Full Member
  • Posts: 626
Re: The Megaliths open baffle line array speakers
« Reply #38 on: 10 May 2016, 10:32 pm »
Have most of the Mathmatica document done and getting the syntax correct since it is a bit different than Mathcad.

Decided to build insulated binding post setups for the amp modules.  Luckily PE was having a closeout on the large binding post and the simpler post connectors with the insulated plate, so was able to build all 38 binding post pairs. 



HAL - whats the PE part numbers on those if you have it handy? Thanks

gab

HAL

Re: The Megaliths open baffle line array speakers
« Reply #39 on: 10 May 2016, 11:31 pm »
Here they are: