It's interesting... I didn't experiment with it like you did, I'd love to see an impedance plot of one of those lines of yours as is. Also take a look at the group delay plot within the impedance test if taken with REW.
I could see a theoretical advantage to parallel wiring, at one point I even thought about using separate amps for each driver, but couldn't figure out how to drive 25 amps on each side and keep it manageable.
I still wonder how B&O do that with separate amps for each driver in their Beolab 90. I suspect they had a bigger budget to work with though
.Even though theory denies the effectiveness I still have my impedance compensation circuit attached. Just the part that flattens the impedance bump.
The improvement I hear from that may be all in my head though
.The measurable differences after FIR correction were minimal. The group delay plot of an impedance test did show differences though, but who ever looks at that graph?
Your STEP response looks great, how does an early waterfall plot look?
With settings like these:
Just out of curiosity....
The thing that kept me form building a focussed array was I didn't want to risk not being able to use the setup for the entire family. How does it sound outside the sweet spot?
I figured if a focussed array would be best in the sweet spot, Keely's array performs pretty good in an entire room and is bent the other way, so with some help the straight array should be 'manageable' in a wider listening area. And it was, it worked out better than expected. Even way off axis it still sounds impressive.
Thanks for the compliment... I liked the whole process, the design and ideas, the build and finally the DSP correction to tame them. But what price card could 'one' ever hang on a project like this, created by hand with hand tools...
Even if you'd consider it a paying hobby, the hours invested were borderline crazy. It really was a mad man's journey.
I've locked away every jig and template though (for save keeping), you never know! I sure am crazy enough...
I still can't think of a less invasive system (if you look at floor space occupied) with the same potential. If you ask me Sd area counts!
Last edited:
Hello wesayso,
in the www I found some ideas why parallel wiring could sound a bit better:
Two simple Current Amplifiers for low impedance loads
I coud not believe it, so I had to make a test with my own ears. Threfore I designed different wiring -constellations for comparing:
[/url]
Paul Hynes offered m a simple circuit of a transistor amp for low Impedances. But the more simple way was to take a amp with a transformer: McIntosh!
Befor I build the arrays I listened with a fullrange-fronthorn without any frequency-crossover. With the arrays I got some parts of this sound signature which I loved so much just by paralell-wiring.
Nevertheless: How does your impedance compensating look like?
This are the measurements you wanted to have, made in a distance of 2,95 meters.
You are right, my setup is not usable for more than one listener. The sound becomes dull if you move up or down more than 15 cm. tolerance in lateral direction is +-10 cm, and axial +- 40 cm.
But I expected this effect and do not have any problem with it.
Inside this sweet spot of course the sound is better than everything I ve heard.
Exactly like a one point source, but with the advantage of a big driver-surface-You know what I mean.
Btw:Crossover cancellation was a theme for me ,too.
I made som experiments with recussive HRTF base filters, but the most convincing way to improve the sound was quite simple:
[url=http://abload.de/image.php?img=p1000461g8pvj.jpg]
aktives-hoeren.de • Thema anzeigen - Ambiophonie - XTC, RACE oder auch BACCH-SP
Thank you once more for your inspiration you gave me!
Best wishes
Martin
in the www I found some ideas why parallel wiring could sound a bit better:
Two simple Current Amplifiers for low impedance loads
I coud not believe it, so I had to make a test with my own ears. Threfore I designed different wiring -constellations for comparing:
[/url]
Paul Hynes offered m a simple circuit of a transistor amp for low Impedances. But the more simple way was to take a amp with a transformer: McIntosh!
Befor I build the arrays I listened with a fullrange-fronthorn without any frequency-crossover. With the arrays I got some parts of this sound signature which I loved so much just by paralell-wiring.
Nevertheless: How does your impedance compensating look like?
This are the measurements you wanted to have, made in a distance of 2,95 meters.
You are right, my setup is not usable for more than one listener. The sound becomes dull if you move up or down more than 15 cm. tolerance in lateral direction is +-10 cm, and axial +- 40 cm.
But I expected this effect and do not have any problem with it.
Inside this sweet spot of course the sound is better than everything I ve heard.
Exactly like a one point source, but with the advantage of a big driver-surface-You know what I mean.
Btw:Crossover cancellation was a theme for me ,too.
I made som experiments with recussive HRTF base filters, but the most convincing way to improve the sound was quite simple:
[url=http://abload.de/image.php?img=p1000461g8pvj.jpg]
aktives-hoeren.de • Thema anzeigen - Ambiophonie - XTC, RACE oder auch BACCH-SP
Thank you once more for your inspiration you gave me!
Best wishes
Martin
Thanks for showing these, Martin.
The waterfall plot is very convincing! It shows there is no combing in the sweet spot. This proves that the distance of the speaker to the ear is the most dominant, and combing will occur outside that sweet spot. Nice!
This one is interesting too:
It looks much like what a single driver would show from what I've seen in my tests. It might even explain why I didn't like the compensation for the rising impedance at higher frequencies. Hmmm... more stuff to ponder about.
My circuit looked like this:
Member BYRTT helped me to figure out what and how to use the values in Bill Waslo's very useful XSim.
I fine-tuned some of this basic schematic with new impedance measurements after changing my damping to butyl between both aluminium plates/slats that form my baffle.
After listening tests I disliked the compensation for the rising impedance and removed that circuit completely. This is the left over circuit as used:
It only removes the impedance bump at Fs.
The parts in real life:
This obviously is the conjugate network in an unfinished state. I removed all high frequency compensation parts and only use the two big capacitors, one 10 ohm and one 4.7 ohm resistor and the big coil.
The waterfall plot is very convincing! It shows there is no combing in the sweet spot. This proves that the distance of the speaker to the ear is the most dominant, and combing will occur outside that sweet spot. Nice!
This one is interesting too:
It looks much like what a single driver would show from what I've seen in my tests. It might even explain why I didn't like the compensation for the rising impedance at higher frequencies. Hmmm... more stuff to ponder about
.My circuit looked like this:
Member BYRTT helped me to figure out what and how to use the values in Bill Waslo's very useful XSim.
I fine-tuned some of this basic schematic with new impedance measurements after changing my damping to butyl between both aluminium plates/slats that form my baffle.
After listening tests I disliked the compensation for the rising impedance and removed that circuit completely. This is the left over circuit as used:
It only removes the impedance bump at Fs.
The parts in real life:
This obviously is the conjugate network in an unfinished state. I removed all high frequency compensation parts and only use the two big capacitors, one 10 ohm and one 4.7 ohm resistor and the big coil.
Martin,Hello wesayso,
thank you for the compliments! Please excuse my bad english.
I am convinced that every resistor or cap in the signal path is audible and should be avoided if possible.
For better viewing once more the measurements and pictures:
best wishes
Martin
You should try the portable mohawk.
Sent from my XT1575 using Tapatalk
I totally missed out on that trend in the late 70's and early 80's. I was a bit to young and at that time didn't realise its potential . Now it's to late for me.... my hairs on top just don't have enough volume anymore . The top of my scalp is shining trough. Maybe that portable solution is something to try .
What do you for the bass notes, Martin? I doubt you use a lot of boost in the low end with the speakers placed freely like that and the transformer in the amplifier would have limits when using a lot of boost I suppose?
How low does it go by itself? I bet the way you set it up in the room helps out. Is the space behind you similar in shape? Do you get any late reflections from the back of the room?
I was a bit to young and at that time didn't realise its potential . Now it's to late for me.... my hairs on top just don't have enough volume anymore . The top of my scalp is shining trough. Maybe that portable solution is something to try .What do you for the bass notes, Martin? I doubt you use a lot of boost in the low end with the speakers placed freely like that and the transformer in the amplifier would have limits when using a lot of boost I suppose?
How low does it go by itself? I bet the way you set it up in the room helps out. Is the space behind you similar in shape? Do you get any late reflections from the back of the room?
The dimensions of our living- and listening room are 6.9 x 9.5 x 2.9 meters. The eckhorn you can see behind the rack is the most fast and dry bass- solution I could find after testing several possibilities like ripols or bass-arrays. The low pass filter- character is neville-thiele 4.th order at 110 hz. For my ears it is not possible lo localise the eckhorn. The eckhorn has its own amplifier with 300 watts althogh it needs just 5 Watts for max loudness. The first early reflections of the arrays would arrive the listening position after 10 ms but because of the different sound sources of the drivers the way of the sound is different and so the reflection is not audible. The next early reflections are more than 20 ms later on. The value for the symmetry (= good stereo) is I think the IACC which I took:
http://abload.de/img/p1000136qdqwy.jpg
More than 97% up to 80 ms. ( means 26 meters). And I am a little bit proud of this value, because I could not find a more high value in any setup in my researches in the www.
Because of the lack of early and later reflections in my setup may be that the crazy XTC-filter I built is more audible than in an environment with more reflections.
As you know it is possible to boost the array and let them play fullrange, but when I tested this option I realised more confusion and less sovereignity especially when the music became more loud and complex. You are right, one positive aspect is indeed that the amp is free of limits whe the lower frequencies are cut of. I measured a peak of 5 Watts per cannel the amplifier had to deliver. So the 200 Watts the MC 2200 shows at 0.5 ohms should be enough. Other also stabile amps with enough power I tested did not convince me. I don t know the tecnical reasons of this.
http://abload.de/img/p1000136qdqwy.jpg
More than 97% up to 80 ms. ( means 26 meters). And I am a little bit proud of this value, because I could not find a more high value in any setup in my researches in the www.
Because of the lack of early and later reflections in my setup may be that the crazy XTC-filter I built is more audible than in an environment with more reflections.
As you know it is possible to boost the array and let them play fullrange, but when I tested this option I realised more confusion and less sovereignity especially when the music became more loud and complex. You are right, one positive aspect is indeed that the amp is free of limits whe the lower frequencies are cut of. I measured a peak of 5 Watts per cannel the amplifier had to deliver. So the 200 Watts the MC 2200 shows at 0.5 ohms should be enough. Other also stabile amps with enough power I tested did not convince me. I don t know the tecnical reasons of this.
First, let me say thanks and bow my head in awe to @wesayso.
I've read this monster of a thread in the last few days and, besides just loving the design I found out how much I still have to learn. This takes the classical "Schallzeile" (=line array) to a totally new level I'd never have expected.
Be sure I will follow this thread and use it as a reference when it comes to what to study...
Greetings from Vienna
Gerald
I've read this monster of a thread in the last few days and, besides just loving the design I found out how much I still have to learn. This takes the classical "Schallzeile" (=line array) to a totally new level I'd never have expected.
Be sure I will follow this thread and use it as a reference when it comes to what to study...
Greetings from Vienna
Gerald
Those are impressive numbers in the IACC test. I wish I could figure out how to run that with my setup. As I need JRiver in my play back chain I'm not sure I could do that with the Asio drivers needed for Acourate. I assume this test is available in the measurement setup without needing to buy Acourate? I'm just a poor man....
If you don't mind, I do have a lot more questions
. If you do mind, just say so. It's just curiosity that spikes my questions. No doubt your bent array has some advantages over my straight arrays and I'd like to learn from that. The advantages I'm getting at is the lack of combing on the top end. While I try to compensate for it in FIR filtering I cannot deny there's still an effect there. I wouldn't know how that relates to listening as we haven't been able to listen to each others systems.
A first question I have is: what did your cross talk device (the passive one) do for imaging? Can you describe what changes if anything? (obviously something changed if you're prepared to wear a Mohawk like that
)A second question would be; have you tried Dr. Dr. Ulrich Brüggemann's AcourateFLOW? Did you like what it did?
I had a private talk with Mitchba about my mid/side EQ and that made him try AcourateFLOW as it's based on the same principles as my mid/side EQ processing. Just curious what it does in your setup.
And if you're up for it I'll create 2 test tracks of one song to see which one you'd prefer. I'd only mess with phase using Pano's shuffler as discussed in this thread: http://www.diyaudio.com/forums/multi-way/277519-fixing-stereo-phantom-center.html
(Not the shuffler from the start of the thread, but a later shuffler created with RePhase. I'll simply name the tracks A and B and the only job you'd have to do is listen, tell us what's different in perception (if there are perceivable differences) and if you'd choose/prefer one over the other.
Others can join in of course (Perceval, see? I got it right
) and I hope they do.I can still think of many more questions but I don't want to overload you right now
. I'm thankful for any answer you may provide. I'm just trying to learn a bit more and maybe get some ideas of useful tests to run.I bet if we all work together we could even improve our Stereo enjoyment!
Thank you for the kind words. Much appreciated. It takes dedication to get trough this enormous thread.
---------------------------------------------------------------------------------------------------------
To answer my own question from a few pages back: I guess I'm not ready to take a break yet
.
Last edited:
