Todd's AG Trio G3 System -- it's about time
- Thread starter sbnx
- Start date
Yes. In theory. But i will still have to align the speaker with itself and the left speaker. We can not measure accurately enough with a tape measure. For example, my floor is not perfectly level. I will need to adjust the rake to compensate for this.
Hello everyone. A little update on the system progress.
I was somewhat in the weeds on the bass bin. I had taken measurements earlier when I was trying to find the rough area that I wanted to position the speaker and I know that with the bass bin anywhere in the 2nd side panel is a very good spot from a frequency response perspective. You can visit one of my previous videos on this and see the REW graph. However, aligning the bass bin to the mains was proving to be quite the challenge. This might be easier if I was willing to put them between the speakers but for me that would have to be a last ditch effort.
The weeds are pretty tall and easy to get lost in. There are six degrees of freedom on the bass bin: front/back, left/right, up/down, Toe-in, Rake and Azimuth. But wait there are three more: Crossover point, Volume and Phase. Ouch!
Based on the measurements I took had I set the crossover to 158Hz (18 dB/Oct Slope), set the volume to -5.0 and set the phase to 0ms.
I had raised the height of the woofer a good bit already by putting Track Audio feet on it and was experimenting with placing wood blocks under them. I had cut shims at 1/16, 1/8 and 1/4 inch thicknesses and was experimenting to some success with those. I had matched the rake angle of the main horns. I originally had matched the toe-in angle but had toed-in the bass bin more in an attempt to correct some detachment. I had experimented with lateral spacing to address coupling/decoupling the bass. It always seemed like I wanted to go in to make them more coherent but when I did that other bad stuff happened. I could easily time align the bass bin by bumping it forward or backward using some symphonic music. But, I just could not seem to decode this mass of semi-independent variables. Maybe another few months and I would have had a matrix of mappings that would give me the exact adjustment(s) to each musical queue. But it was getting pretty maddening.
The most maddening thing was that I could get simple things, such as upright bass (plucked or bowed), to sound good. But the acid test was Shostakovich Symphony 8 movements 3 and 4. The end of the 3rd (Allegro non troppo) has the percussion beating the hell out of the timpani while the horns and other instruments are going full tilt as well. Then the beginning of the 4th is the crescendo. This is very exciting music and when played correctly always gives me goose bumps. But hard as I might try it was always laughably bad. The timpani were a hot mess with the timing being all over the place and the bass wave of the crescendo was arriving so late. Dmitri would probably tell me to stop playing his symphony if that is what it was going to sound like.
Time to phone a friend. I called Stirling Trayle to see if he could come over and spend a single day with me. Not to setup up the bass bin but to go over with me how he approaches subwoofer/woofer integration and work on it together. As always, I learned a ton. This is that story but reader beware. There are some shocking and contentious things below. I will just report what we did and that it worked.
We started out listening to a few things and he made a few small adjustments to the main horns as he was starting to get to know the Trios. No X,Y movements. Mostly raking the speaker back a little more and then the corresponding tweeter adjustment. I used my laser device and by raking the speaker back we were pointing the tweeter more and more at the ear canal. Nice to know I wasn’t crazy on that one. Then we moved on to the business of the bass bin.
We started the bass bin work with Stirling showing me how much an impact the bass has on the high frequencies. We were using an oboe part for this. It is shocking just how much you can change the tone of an oboe by making very, very small adjustments to the position of the bass bin or by adjusting the phase in the DSP. The vibrato on the highest note played could go from non-existent to shrill to speaker induced to pure oboe vibrato. The tone could be shifted up to the point it sounded more like a kazoo than an oboe. The dramatic pause where the oboe and symphony stop together could be completely ruined. The supporting musicians could sound like they are trying to push the oboist out of the way. The expressiveness of the oboe could go from blah to tear inducing beauty. That is a lot of oboe control for that huge behemoth bass bin. You can imagine the Tuba, bass and bass drum players all with little puppet strings on the oboe player saying “Ha! Ha! I have you now. Come to the dark side.”
I learned how to detect the sonic signature of phase problems and what to do with phase adjustments in the DSP to make the oboe sound natural. We started with big phase changes of 1.5 ms (this would be about 90 degrees at the crossover point). But quickly realized this was too much Then we started reducing the phase by 0.1ms and then finally we were adjusting in 0.001ms (1 microsecond) increments. We ended up at 0.142 ms. This was quite easy to do as I could just have the laptop in my lap and type in the box 0.145, 0.144, 0.143, 0.142, 0.141, etc. As we worked our way down and around the phase a change of 1 microsecond was very easy to detect. When experimenting and going down from 0.142 (which was the good spot) to say 0.138 it was getting worse. Then jump back to 0.142 and it was a wow moment when all of the musicality of the oboe returned.
For those of you who are wondering how far sound travel in 1 microsecond. That is a distance of 0.014 inch or 0.35 mm. Perhaps this gives pause to those who have tried integrating a subwoofer with the phase knob on the back.
The way I thought about phase before was not like this. I simply thought of phase an electronic way of altering the relative position of the woofers. In other words, a way of time aligning the subwoofer/woofer to the main array. When we had this discussion on phase vs. position it was like a light bulb went off in my head. There is the woofer position in the room and that corresponding interaction. Then there is the phase “knob”. If you like the position of the woofer then you can use phase to adjust. If you like the phase you can adjust the physical position of the woofer. The two combined let you both time align the woofer and dial in the tone. Who knew?! Well I guess Stirling knew.
We spent the rest of the morning basically experimenting with small movements as he was just getting to know the system. I had the luxury of having spent the last several months getting to know it. We played with height and I cut some thicker shims as we kept moving up. But it seemed that height wasn’t really doing it alone. The bass would simply not release and was stuck on the floor. Since the bass was on the floor and I had the woofer raked forward 2.6 degrees to match the mains we decided it made logical sense to see what would happen with raking the bass bin back. So we zero’d out the speaker and return it to level rake. That was the right direction. We ended up raking the woofer back 0.489 degrees when all was done. But of course it was a journey to get to that rake angle.
Now that we had the phase in the ballpark and figured out how to get the bass to release with rake and be articulate with height we moved on to the business of getting the woofer integrated with the main array. We were working on the overlap of the subwoofer with the main speaker. We stated with lateral movements. But was it out or in. It seemed there was a disconnect in the music. Some signals pointed to out and some to in. We assessed the azimuth at some point and by chance it was at 0.084 degrees out. We decided to set the azimuth to zero and proceed with lateral adjustments. So we got it to zero and started moving the woofer out one bump at a time and then 1mm at a time but at about the 5th millimeter it was a time-out moment. Why aren’t things improving to the point we had them earlier in the day? Then upon reflection we realized that things had never been better than when we had started with the azimuth 0.084. We put the azimuth back and things dramatically improved. Then the epiphany hit me. The bass bin is tall and it is interacting with the top horn. So now we had the key -- lateral movement and azimuth. We started adjsutin azimuth out more and laterally moving the bass bin. The sonic signature of the azimuth change became more and more obvious as did the sonic signature of the lateral movement. We ended up with an azimuth of 0.141 degrees in the “out” direction.
We ended the day with some small bumps front/back to time align and some more experimenting with phase until we were happy. A few very small azimuth adjustments to lock things in. And viola, the oboe now has oodles of musical expression with perfect vibrato. Bass plucks are very nice and bouncy. The acid test passes with flying colors. The Timpani in Shostakovich 8 is now perfect with a massive crescendo that washes over me. I think Dmitri would be proud.
If you are having issues getting your speakers to work then don’t be afraid to ask for help. It certainly saved me a lot of time and I am not sure I would have ever figured out the phase thing.
I was somewhat in the weeds on the bass bin. I had taken measurements earlier when I was trying to find the rough area that I wanted to position the speaker and I know that with the bass bin anywhere in the 2nd side panel is a very good spot from a frequency response perspective. You can visit one of my previous videos on this and see the REW graph. However, aligning the bass bin to the mains was proving to be quite the challenge. This might be easier if I was willing to put them between the speakers but for me that would have to be a last ditch effort.
The weeds are pretty tall and easy to get lost in. There are six degrees of freedom on the bass bin: front/back, left/right, up/down, Toe-in, Rake and Azimuth. But wait there are three more: Crossover point, Volume and Phase. Ouch!
Based on the measurements I took had I set the crossover to 158Hz (18 dB/Oct Slope), set the volume to -5.0 and set the phase to 0ms.
I had raised the height of the woofer a good bit already by putting Track Audio feet on it and was experimenting with placing wood blocks under them. I had cut shims at 1/16, 1/8 and 1/4 inch thicknesses and was experimenting to some success with those. I had matched the rake angle of the main horns. I originally had matched the toe-in angle but had toed-in the bass bin more in an attempt to correct some detachment. I had experimented with lateral spacing to address coupling/decoupling the bass. It always seemed like I wanted to go in to make them more coherent but when I did that other bad stuff happened. I could easily time align the bass bin by bumping it forward or backward using some symphonic music. But, I just could not seem to decode this mass of semi-independent variables. Maybe another few months and I would have had a matrix of mappings that would give me the exact adjustment(s) to each musical queue. But it was getting pretty maddening.
The most maddening thing was that I could get simple things, such as upright bass (plucked or bowed), to sound good. But the acid test was Shostakovich Symphony 8 movements 3 and 4. The end of the 3rd (Allegro non troppo) has the percussion beating the hell out of the timpani while the horns and other instruments are going full tilt as well. Then the beginning of the 4th is the crescendo. This is very exciting music and when played correctly always gives me goose bumps. But hard as I might try it was always laughably bad. The timpani were a hot mess with the timing being all over the place and the bass wave of the crescendo was arriving so late. Dmitri would probably tell me to stop playing his symphony if that is what it was going to sound like.
Time to phone a friend. I called Stirling Trayle to see if he could come over and spend a single day with me. Not to setup up the bass bin but to go over with me how he approaches subwoofer/woofer integration and work on it together. As always, I learned a ton. This is that story but reader beware. There are some shocking and contentious things below. I will just report what we did and that it worked.
We started out listening to a few things and he made a few small adjustments to the main horns as he was starting to get to know the Trios. No X,Y movements. Mostly raking the speaker back a little more and then the corresponding tweeter adjustment. I used my laser device and by raking the speaker back we were pointing the tweeter more and more at the ear canal. Nice to know I wasn’t crazy on that one. Then we moved on to the business of the bass bin.
We started the bass bin work with Stirling showing me how much an impact the bass has on the high frequencies. We were using an oboe part for this. It is shocking just how much you can change the tone of an oboe by making very, very small adjustments to the position of the bass bin or by adjusting the phase in the DSP. The vibrato on the highest note played could go from non-existent to shrill to speaker induced to pure oboe vibrato. The tone could be shifted up to the point it sounded more like a kazoo than an oboe. The dramatic pause where the oboe and symphony stop together could be completely ruined. The supporting musicians could sound like they are trying to push the oboist out of the way. The expressiveness of the oboe could go from blah to tear inducing beauty. That is a lot of oboe control for that huge behemoth bass bin. You can imagine the Tuba, bass and bass drum players all with little puppet strings on the oboe player saying “Ha! Ha! I have you now. Come to the dark side.”
I learned how to detect the sonic signature of phase problems and what to do with phase adjustments in the DSP to make the oboe sound natural. We started with big phase changes of 1.5 ms (this would be about 90 degrees at the crossover point). But quickly realized this was too much Then we started reducing the phase by 0.1ms and then finally we were adjusting in 0.001ms (1 microsecond) increments. We ended up at 0.142 ms. This was quite easy to do as I could just have the laptop in my lap and type in the box 0.145, 0.144, 0.143, 0.142, 0.141, etc. As we worked our way down and around the phase a change of 1 microsecond was very easy to detect. When experimenting and going down from 0.142 (which was the good spot) to say 0.138 it was getting worse. Then jump back to 0.142 and it was a wow moment when all of the musicality of the oboe returned.
For those of you who are wondering how far sound travel in 1 microsecond. That is a distance of 0.014 inch or 0.35 mm. Perhaps this gives pause to those who have tried integrating a subwoofer with the phase knob on the back.
The way I thought about phase before was not like this. I simply thought of phase an electronic way of altering the relative position of the woofers. In other words, a way of time aligning the subwoofer/woofer to the main array. When we had this discussion on phase vs. position it was like a light bulb went off in my head. There is the woofer position in the room and that corresponding interaction. Then there is the phase “knob”. If you like the position of the woofer then you can use phase to adjust. If you like the phase you can adjust the physical position of the woofer. The two combined let you both time align the woofer and dial in the tone. Who knew?! Well I guess Stirling knew.
We spent the rest of the morning basically experimenting with small movements as he was just getting to know the system. I had the luxury of having spent the last several months getting to know it. We played with height and I cut some thicker shims as we kept moving up. But it seemed that height wasn’t really doing it alone. The bass would simply not release and was stuck on the floor. Since the bass was on the floor and I had the woofer raked forward 2.6 degrees to match the mains we decided it made logical sense to see what would happen with raking the bass bin back. So we zero’d out the speaker and return it to level rake. That was the right direction. We ended up raking the woofer back 0.489 degrees when all was done. But of course it was a journey to get to that rake angle.
Now that we had the phase in the ballpark and figured out how to get the bass to release with rake and be articulate with height we moved on to the business of getting the woofer integrated with the main array. We were working on the overlap of the subwoofer with the main speaker. We stated with lateral movements. But was it out or in. It seemed there was a disconnect in the music. Some signals pointed to out and some to in. We assessed the azimuth at some point and by chance it was at 0.084 degrees out. We decided to set the azimuth to zero and proceed with lateral adjustments. So we got it to zero and started moving the woofer out one bump at a time and then 1mm at a time but at about the 5th millimeter it was a time-out moment. Why aren’t things improving to the point we had them earlier in the day? Then upon reflection we realized that things had never been better than when we had started with the azimuth 0.084. We put the azimuth back and things dramatically improved. Then the epiphany hit me. The bass bin is tall and it is interacting with the top horn. So now we had the key -- lateral movement and azimuth. We started adjsutin azimuth out more and laterally moving the bass bin. The sonic signature of the azimuth change became more and more obvious as did the sonic signature of the lateral movement. We ended up with an azimuth of 0.141 degrees in the “out” direction.
We ended the day with some small bumps front/back to time align and some more experimenting with phase until we were happy. A few very small azimuth adjustments to lock things in. And viola, the oboe now has oodles of musical expression with perfect vibrato. Bass plucks are very nice and bouncy. The acid test passes with flying colors. The Timpani in Shostakovich 8 is now perfect with a massive crescendo that washes over me. I think Dmitri would be proud.
If you are having issues getting your speakers to work then don’t be afraid to ask for help. It certainly saved me a lot of time and I am not sure I would have ever figured out the phase thing.
So after all these optimization it must sound phenomenal. Share a video and let us all hear a glimpse of the stellar results of your efforts.
You mean the video that has almost no bass information?
My videos have bass information. One of the guys even claimed that one of the videos of one of my systems was a sea of bass. Don’t be shy. If it sounds great, I’m sure that the video will demonstrate it. You posted videos of your system before. What’s there to be scared of? I like to hear the results of the work that you and Stirling put in.
I am certainly not afraid to post a video. I posted videos earlier in this thread of the piano as I was comparing with and without the ground box and preamp gain. I will see what I can do. I thought about seeing if we can hear the change in the oboe through the iphone with making small phase adjustments to the bass bin.
I don't know how you have a sea of bass through an iphone mic.
I don't know how you have a sea of bass through an iphone mic.
Check out some of my most recent videos of my WAAR system with the Remastering process and judge for yourself when you get a chance. It may also provide a good reference for comparison to you as well.
iPhones can use external mic’s also. Try it. Don’t believe me? Listen to my videos and judge for yourself.
So you mean to tell me that I’m the only one able to take great videos? Come man, have some confidence.
Todd, this is a great journey and I love following you on your way to perfection.
What I really don't understand is why the crossover of your sub is 158hz.
The big horn goes down to around 105hz (when I am not mistaken)
My Unos (from 2004) x-over at 195hz.
Happy to see where you are heading, I have to visit Avantgarde factory again to hear the G3 series.
Regards, Remco
What I really don't understand is why the crossover of your sub is 158hz.
The big horn goes down to around 105hz (when I am not mistaken)
My Unos (from 2004) x-over at 195hz.
Happy to see where you are heading, I have to visit Avantgarde factory again to hear the G3 series.
Regards, Remco
Hello and thank you for the kind comments.
When we started working on the bass bin the crossover was at 158Hz. When we finished it was lowered to 135Hz and that is where it is now. The crossover slope is 18dB/Oct as a point of reference. So by 270 Hz the woofer is down 18dB. This is still quite high in frequency and thus the challenge of getting it to integrate well. It makes integrating a subwoofer down at 35 or 40Hz seem simple.
I measured the output of the main horn and the spacehorn using REW back when I started the process of finding a candidate area in the room to set up the speakers. The bass from the big main horn drops at 24 dB/Oct starting at 95Hz. However, The output of the horn starts dropping below 200 Hz. I am trying to fill some of that in by raising the crossover point. Yes, there is definitely overlap between the two from a frequency perspective.
Alright, it was 95hz actually. Even as the horn slowly drops from 200hz, I think 135hz is really to high. What does Avantgarde say about x-over?
I can't remember if you showed us the horn response from your Trio.
Your way of finding the way to hornheaven is something to behold
I live in Germany and was invited to hear the Trio G3 introduction at the factory. Sadly I didn't get to it...
I can't remember if you showed us the horn response from your Trio.
Your way of finding the way to hornheaven is something to behold
I live in Germany and was invited to hear the Trio G3 introduction at the factory. Sadly I didn't get to it...
Similar threads
