Thanks guys. Much appreciated. Now that I have one speaker done with passive xover in place, I'm going to pause completion of the other speaker so that I can do some A/B monophonic listening comparison tests. One speaker will be driven by a full range signal going into the passive xover. The other speaker will be driven by the convoluted DSP xover channels coming from Equalizer APO. I should be able to switch back and forth fairly quickly. All I need to do is figure out how to add the right and left channels in Equalizer APO to produce a mono signal before convolution. This, I would think, should be a fairly easy thing to set up. If it works, I can simply use the Equalizer APO full range signal going to one of the plate amplifiers to drive the passive crossover.
@Steve_Lee said:
Can you run OmniMic on a separate PC from the one playing/processing the digital audio, 4thtry?
(this is what I do to reduce latency/processor overhead).
Thanks again for the suggestion, Steve. I gave this a try, but it did not work. I still get a huge suck out in the 500 to 800Hz region when using two PC's to make the measurements when Equalizer APO is in the measurement timing loop. Equalizer APO is somehow conflicting with the OmniMic Whoop-Whoop-Whoop process. When I do an OmniMic validation measurement of the passive crossover alone, without Equalizer APO in the loop, I get a nice, smooth flat response with no 500 to 800Hz suck out.
I've been doing A/B listening tests for several days now, comparing the convoluted Equalizer APO crossover to the passive crossover. I have everything set up and working properly to quickly switch a mono signal back and forth between the left and right speakers, one speaker using the digital xover and the other using the passive xover.
When I first started switching back and forth I could easily hear differences in tonal balance between the passive and digital xovers. But, as it turns out, this was due to set up problems. It is very difficult to level match the two speakers because I am using 3 separate stereo amplifiers. The preamplifier and all power amplifier gain levels have to be very carefully matched, otherwise one channel will sound more vibrant, punchy, or brighter than the other. And the comparison needs to be made using a mono signal to driver both speakers. Initially, I had the stereo to mono conversion programming set up incorrectly in Equalizer APO. The right and left channels were simply being swapped instead of being converted to mono. But once I got the level matching and the mono conversion problems fixed, the two speakers, digital verses passive, sounded remarkably similar.
If you are interested in attempting this type of setup, here are a few screen shots with some comments and one important cautionary note:
I used the Soundblaster X3 mixer balance controls to flip back and forth between speakers. The convoluted digital crossover was on the left channel and the passive crossover was on the right channel. Flipping back and forth took about 10 seconds because I had to do 4 mouse clicks to slide the 3 digital sliders from zero to max and then slide the 1 passive slider to minimum. And then another 4 mouse clicks to slide the 3 digital sliders from max to zero and then slide the 1 passive slider to max.
Here is the Equalizer APO rack listing. The preamplifier level was very carefully adjusted, together with power amplifier gain settings, to balance things out. After the preamplifier I used two "Copy channel" commands to convert the L and R stereo channels to mono. Next, I used a third "Copy channel" command to copy the L and R mono channels to the side and back channels. The tweeter and mid channels were then selected and convoluted with the wav file that I had previously exported from VituixCAD. The "driver in D5" is the tweeter crossover and the "driver in D4" is the midrange crossover. The woofers are simply sent straight through the L and R channels without convolution because they receive the full range signal going to the plate amplifiers, which, in turn, provide the low pass filter at 175Hz.
Hopefully, this information is useful for others that might consider setting up such a system. I followed @dcibel 's basic instructions for most of the setup (link below).
If you have any questions or suggestions, fire away.
One word of caution on the setup. Because of the large number of wires going out to the speakers and because of the somewhat complicated APO interface, it is very easy to send the amplified midrange or woofer signal to the tweeter by accident! I did this twice during setup. Luckily, my mistake was made at very low playback levels, so no tweeter damage occurred. After making this mistake twice, I replaced the tweeter during the setup process with a Dayton 4" carbon fiber woofer (CF120-4). Then, once I had confirmed that only high frequencies were coming from the 4" woofer, I re-connected the speaker wires to the tweeter.
Now that the xovers are complete and the weather is improving, I've decided to turn my attention to the aesthetics. When my garage gets up to about 60 degrees or so, I may be able to start the bondo, sanding, painting process. I put togther a three color scheme in sketchup. A mix of black, burgundy, and gray. What do you think?
What we all hope for is the sound overwhelms any differences on appearance. Though wonder if rendition perception changes once add in the bass stands. Is the gray rings on front implying trim rings on the tectonic mids? Lastly, will there still be exterior hardware still holding the the alignment? Though I understand the intent is to get fundamental visual
Thanks for the feedback. I think you are right, John. Too many colors. How about this one? Cut it down to just two colors, black and brown. The baffles are all brown and should contrast against the black driver flanges. Traw, yes, the gray rings around the mids were for a diy style painted trim ring. I plan to carefully mask off the baffles to do this. I darkened them a bit to make them blend a little better. There will still be hardware holding the alignment; it was just too difficult to position all these brackets in Sketchup, so I gave up. The brackets in the finished speaker will probably be aluminum instead of masonite, so that will probably add just a bit more visual interest to the side profile.
@4thtry said:
Thanks for the feedback. I think you are right, John. Too many colors. How about this one? Cut it down to just two colors, black and brown. The baffles are all brown and should contrast against the black driver flanges. Traw, yes, the gray rings around the mids were for a diy style painted trim ring. I plan to carefully mask off the baffles to do this. I darkened them a bit to make them blend a little better. There will still be hardware holding the alignment; it was just too difficult to position all these brackets in Sketchup, so I gave up. The brackets in the finished speaker will probably be aluminum instead of masonite, so that will probably add just a bit more visual interest to the side profile.
I do like this version with the lighter fronts. I'm nit picking, but my preference would be for the upright stand to also be black. I don't know what your preference is but the black should be flat black. I think the shape is the highlight of the design and I would guard against anything that took away from seeing the shape.
The color scheme is nice. The shape is a bit stark. I'm wondering what the look would be like with a roundover on the top, bottom and back edges, as well as the base and pillar. That could be a can o' worms with no end though, but...
@Turn2 said:
The color scheme is nice. The shape is a bit stark. I'm wondering what the look would be like with a roundover on the top, bottom and back edges, as well as the base and pillar. That could be a can o' worms with no end though, but...
Thanks for the feedback. I will have 1/2" radius roundovers on the top, bottom, base, pillar, and sides of the tweeter section. They are not on the drawings because I'm still on the low end of the Sketchup learning curve, and I haven't figured out how to do roundovers yet.
@traw said:
What we all hope for is the sound overwhelms any differences on appearance. Though wonder if rendition perception changes once add in the bass stands. Is the gray rings on front implying trim rings on the tectonic mids? Lastly, will there still be exterior hardware still holding the the alignment? Though I understand the intent is to get fundamental visual
Traw, here is a rendering with the bass cabinets added. I think if I use the same colors, the upper cabs should blend fairly well with the subs.
@ugly_woofer said:
What about the masonite pieces in the sides?
Haven't decided what to do with these yet. They could be painted black, along with the fender washers and screws. Or I could paint them a contrasting color to make them stand out. Or I could replace them with aluminum bar stock type brackets of the same dimensions and then use stainless steel finishing washers, which I think look a little better than the big honking fender washers that I am using now.
I was thinking T-tracks might be a good option if you could rabbet them into the sides. But that might mean making new enclosures, which would be a major pain.
I would think they would be best painted black to match the enclosures. They're really not a feature I would want to stand out. It would actually be cool if you could figure out a way of making some triangular wedges that fit between the sections, and painted to match, so that you didn't need anything on the sides at all. I know, easier said than done.
How about a [recessed] piano hinge at the front edge of the baffle(s) with threaded rod/aluminum angle bracket(s) at the rear to adjust the angle/aim of each driver?
No visible hardware . . . from the listening position . . . assemble with all enclosures on their side . . .
@Tom_S said:
I was thinking T-tracks might be a good option if you could rabbet them into the sides. But that might mean making new enclosures, which would be a major pain.
Maybe not. I think T-tracks could be mounted flush to a depth of 3/8" and held in place with chamfered screws and nuts on the existing 3/4" thick cabs. I simply used cheap masonite brackets so that I could quickly set things up for "proof of concept" measurements.
@Steve_Lee said:
How about a [recessed] piano hinge at the front edge of the baffle(s) with threaded rod/aluminum angle bracket(s) at the rear to adjust the angle/aim of each driver?
No visible hardware . . . from the listening position . . . assemble with all enclosures on their side . . .
Just a thought.
This is probably the best solution and would probably work well in terms of what I am doing. But it would take some very good woodworking skills to accomplish this, getting all the parts to fit together precisely. Right now, I have all the cabinets carefully positioned so that the acoustic centers of all drivers converge at a distance of 9 feet on axis to the tweeter.
@ugly_woofer said:
I vote for aluminum or brass bar stock.
I am leaning toward aluminum.
@PWRRYD said:
I would think they would be best painted black to match the enclosures. They're really not a feature I would want to stand out. It would actually be cool if you could figure out a way of making some triangular wedges that fit between the sections, and painted to match, so that you didn't need anything on the sides at all. I know, easier said than done.
Triangular wedges would certainly work. After Indy, my plan is to completely disassemble all cabs and prep them for painting. At that time, I could easily install and glue all the cabs together with wedges instead of brackets. Maybe this would give the cabs kind of a "floating in air" type of effect. People viewing the speakers from an angle would be left wondering how all the cabs can remain suspended "in air" without completely falling apart!
I decided to double check my acoustic center offset measurements before making a set of small wedges to fit between the cabs (for time alignment purposes). Really glad I did, because, as it turns out, my initial measurements were in error. I don't know exactly what happened; bad measurements, misinterpretation, software setup problems, etc. For my initial set of offset measurements, I used the three measurement process at a distance of 9 feet, using XSim, OmniMic, and the latest version of the software (6.05). For the double checking process, I switched back to OmniMic version 4.68 and used the three measurement process again. But this time I repeated the set of three measurements at four different distances: 28.5", 48", 84", and 108", all on axis with the tweeter. This eliminated the possibility of misinterpreting the plus and minus signs on XSim's "mod delay" entry.
Since I still have plenty of time before Indy, I think I will "triple" check my offsets yet again using my XLR microphone as well. One of the problems on my first set of measurements might have been that the phase curve tends to bounce around a little during the "whoop, whoop" process and, depending on when you pause the curve to save the FRD, a small amount of phase error can creep in.
Just found out that the Tectonic TEBM46C20N-4B BMR's, that I am using in this design, are NLA. PE cleared them out and they are all gone. Sad. I checked a couple months ago and there were hundreds in stock. I have a few backups, but I was planning on picking up a few more at the SDC in August. Too late now.
Comments
Thanks guys. Much appreciated. Now that I have one speaker done with passive xover in place, I'm going to pause completion of the other speaker so that I can do some A/B monophonic listening comparison tests. One speaker will be driven by a full range signal going into the passive xover. The other speaker will be driven by the convoluted DSP xover channels coming from Equalizer APO. I should be able to switch back and forth fairly quickly. All I need to do is figure out how to add the right and left channels in Equalizer APO to produce a mono signal before convolution. This, I would think, should be a fairly easy thing to set up. If it works, I can simply use the Equalizer APO full range signal going to one of the plate amplifiers to drive the passive crossover.
Thanks again for the suggestion, Steve. I gave this a try, but it did not work. I still get a huge suck out in the 500 to 800Hz region when using two PC's to make the measurements when Equalizer APO is in the measurement timing loop. Equalizer APO is somehow conflicting with the OmniMic Whoop-Whoop-Whoop process. When I do an OmniMic validation measurement of the passive crossover alone, without Equalizer APO in the loop, I get a nice, smooth flat response with no 500 to 800Hz suck out.
I've been doing A/B listening tests for several days now, comparing the convoluted Equalizer APO crossover to the passive crossover. I have everything set up and working properly to quickly switch a mono signal back and forth between the left and right speakers, one speaker using the digital xover and the other using the passive xover.
When I first started switching back and forth I could easily hear differences in tonal balance between the passive and digital xovers. But, as it turns out, this was due to set up problems. It is very difficult to level match the two speakers because I am using 3 separate stereo amplifiers. The preamplifier and all power amplifier gain levels have to be very carefully matched, otherwise one channel will sound more vibrant, punchy, or brighter than the other. And the comparison needs to be made using a mono signal to driver both speakers. Initially, I had the stereo to mono conversion programming set up incorrectly in Equalizer APO. The right and left channels were simply being swapped instead of being converted to mono. But once I got the level matching and the mono conversion problems fixed, the two speakers, digital verses passive, sounded remarkably similar.
If you are interested in attempting this type of setup, here are a few screen shots with some comments and one important cautionary note:
I used the Soundblaster X3 mixer balance controls to flip back and forth between speakers. The convoluted digital crossover was on the left channel and the passive crossover was on the right channel. Flipping back and forth took about 10 seconds because I had to do 4 mouse clicks to slide the 3 digital sliders from zero to max and then slide the 1 passive slider to minimum. And then another 4 mouse clicks to slide the 3 digital sliders from max to zero and then slide the 1 passive slider to max.
Here is the Equalizer APO rack listing. The preamplifier level was very carefully adjusted, together with power amplifier gain settings, to balance things out. After the preamplifier I used two "Copy channel" commands to convert the L and R stereo channels to mono. Next, I used a third "Copy channel" command to copy the L and R mono channels to the side and back channels. The tweeter and mid channels were then selected and convoluted with the wav file that I had previously exported from VituixCAD. The "driver in D5" is the tweeter crossover and the "driver in D4" is the midrange crossover. The woofers are simply sent straight through the L and R channels without convolution because they receive the full range signal going to the plate amplifiers, which, in turn, provide the low pass filter at 175Hz.
Hopefully, this information is useful for others that might consider setting up such a system. I followed @dcibel 's basic instructions for most of the setup (link below).
https://www.htguide.com/forum/articles/do-it-yourself-diy/927389-how-to-filter-simulation-with-vituixcad-and-eq-apo
If you have any questions or suggestions, fire away.
One word of caution on the setup. Because of the large number of wires going out to the speakers and because of the somewhat complicated APO interface, it is very easy to send the amplified midrange or woofer signal to the tweeter by accident! I did this twice during setup.
Luckily, my mistake was made at very low playback levels, so no tweeter damage occurred. After making this mistake twice, I replaced the tweeter during the setup process with a Dayton 4" carbon fiber woofer (CF120-4). Then, once I had confirmed that only high frequencies were coming from the 4" woofer, I re-connected the speaker wires to the tweeter.
Now that the xovers are complete and the weather is improving, I've decided to turn my attention to the aesthetics. When my garage gets up to about 60 degrees or so, I may be able to start the bondo, sanding, painting process. I put togther a three color scheme in sketchup. A mix of black, burgundy, and gray. What do you think?
I likey
The different colors look to busy imo. The shapes are interesting enough.
Make the fronts all one color and the sides another?
What we all hope for is the sound overwhelms any differences on appearance. Though wonder if rendition perception changes once add in the bass stands. Is the gray rings on front implying trim rings on the tectonic mids? Lastly, will there still be exterior hardware still holding the the alignment? Though I understand the intent is to get fundamental visual
Thanks for the feedback. I think you are right, John. Too many colors. How about this one? Cut it down to just two colors, black and brown. The baffles are all brown and should contrast against the black driver flanges. Traw, yes, the gray rings around the mids were for a diy style painted trim ring. I plan to carefully mask off the baffles to do this. I darkened them a bit to make them blend a little better. There will still be hardware holding the alignment; it was just too difficult to position all these brackets in Sketchup, so I gave up. The brackets in the finished speaker will probably be aluminum instead of masonite, so that will probably add just a bit more visual interest to the side profile.
I like this rendering.
Yup, better. If that was walnut and black it'd be even more pimp. Looking forward to hearing these units regardless of color.
I do like this version with the lighter fronts. I'm nit picking, but my preference would be for the upright stand to also be black. I don't know what your preference is but the black should be flat black. I think the shape is the highlight of the design and I would guard against anything that took away from seeing the shape.
The color scheme is nice. The shape is a bit stark. I'm wondering what the look would be like with a roundover on the top, bottom and back edges, as well as the base and pillar. That could be a can o' worms with no end though, but...
Thanks for the feedback. I will have 1/2" radius roundovers on the top, bottom, base, pillar, and sides of the tweeter section. They are not on the drawings because I'm still on the low end of the Sketchup learning curve, and I haven't figured out how to do roundovers yet.
Traw, here is a rendering with the bass cabinets added. I think if I use the same colors, the upper cabs should blend fairly well with the subs.
Maybe just me but I think they would look better if the subs had the same brown baffles and black cabinets.
Like this?
Really like that better!
What about the masonite pieces in the sides?
Haven't decided what to do with these yet. They could be painted black, along with the fender washers and screws. Or I could paint them a contrasting color to make them stand out. Or I could replace them with aluminum bar stock type brackets of the same dimensions and then use stainless steel finishing washers, which I think look a little better than the big honking fender washers that I am using now.
https://www.homedepot.com/p/Hillman-6-Stainless-Steel-Finish-Washer-40-Pack-2905/204775397
I was thinking T-tracks might be a good option if you could rabbet them into the sides. But that might mean making new enclosures, which would be a major pain.
I would think they would be best painted black to match the enclosures. They're really not a feature I would want to stand out. It would actually be cool if you could figure out a way of making some triangular wedges that fit between the sections, and painted to match, so that you didn't need anything on the sides at all. I know, easier said than done.
I vote for aluminum or brass bar stock.
How about a [recessed] piano hinge at the front edge of the baffle(s) with threaded rod/aluminum angle bracket(s) at the rear to adjust the angle/aim of each driver?
No visible hardware . . . from the listening position . . . assemble with all enclosures on their side . . .
Just a thought.
Maybe not. I think T-tracks could be mounted flush to a depth of 3/8" and held in place with chamfered screws and nuts on the existing 3/4" thick cabs. I simply used cheap masonite brackets so that I could quickly set things up for "proof of concept" measurements.
This is probably the best solution and would probably work well in terms of what I am doing. But it would take some very good woodworking skills to accomplish this, getting all the parts to fit together precisely. Right now, I have all the cabinets carefully positioned so that the acoustic centers of all drivers converge at a distance of 9 feet on axis to the tweeter.
I am leaning toward aluminum.
Triangular wedges would certainly work. After Indy, my plan is to completely disassemble all cabs and prep them for painting. At that time, I could easily install and glue all the cabs together with wedges instead of brackets. Maybe this would give the cabs kind of a "floating in air" type of effect. People viewing the speakers from an angle would be left wondering how all the cabs can remain suspended "in air" without completely falling apart!
Suggestion - go ahead and measure/fit the wedges now so that they will be available and set to the correct angle before you dismantle them . . .
I decided to double check my acoustic center offset measurements before making a set of small wedges to fit between the cabs (for time alignment purposes). Really glad I did, because, as it turns out, my initial measurements were in error. I don't know exactly what happened; bad measurements, misinterpretation, software setup problems, etc. For my initial set of offset measurements, I used the three measurement process at a distance of 9 feet, using XSim, OmniMic, and the latest version of the software (6.05). For the double checking process, I switched back to OmniMic version 4.68 and used the three measurement process again. But this time I repeated the set of three measurements at four different distances: 28.5", 48", 84", and 108", all on axis with the tweeter. This eliminated the possibility of misinterpreting the plus and minus signs on XSim's "mod delay" entry.
Since I still have plenty of time before Indy, I think I will "triple" check my offsets yet again using my XLR microphone as well. One of the problems on my first set of measurements might have been that the phase curve tends to bounce around a little during the "whoop, whoop" process and, depending on when you pause the curve to save the FRD, a small amount of phase error can creep in.
Just found out that the Tectonic TEBM46C20N-4B BMR's, that I am using in this design, are NLA. PE cleared them out and they are all gone. Sad. I checked a couple months ago and there were hundreds in stock. I have a few backups, but I was planning on picking up a few more at the SDC in August. Too late now.
This is just an opportunity to locate and test some newer drivers, me thinx . . .
I hope Dennis Murphy stocked up on those. If not, he'll have to design a new BMR Monitor.