Here is the final result of the subwoofer cabinet:
Normally you do not see anything has changed at all:
Only when you "dive under" you see the underside of the woofer cabinet and when you are in "hibernation mode" late in the evening on the couch you can just see the edge of the subwoofer units.
My wife is happy (the "WAF" part of this topic) and we are both happy because of the much better and complete sound
Normally you do not see anything has changed at all:
Only when you "dive under" you see the underside of the woofer cabinet and when you are in "hibernation mode" late in the evening on the couch you can just see the edge of the subwoofer units.
My wife is happy (the "WAF" part of this topic) and we are both happy because of the much better and complete sound
Just a thought, would it be cheeper to get a new wife? Just kidding I noticed you have artsy concrete and stainless pipe speaker stands, could you make stands that are the subwoofers? They wouldn’t be as light looking but would be practical. Actually, I like your invisible subwoofer idea and it looks like a great project. As an aside, I’ve had downward firing subs for years and they didn’t sag at all. I’m planning a set of IB subs in my crawl space that would be upward firing but the q has to be .7 so maybe that won’t work. Also, all woofers are going to sag, you can’t fight gravity. It’s the law! I remember a calculator to see if a woofer would work in down firing orientation.
I found some math:
You can calculate the sag of a driver from:
Percentage of Sag = 24,849 / ( Xmax * Fs²)
where
The following is the same formula, including the relationship of acceleration and Pi in this case:
Percentage of Sag = 981,000 / (Xmax * (2 * Pi * Fs)²)
where
As a general rule of thumb, any time the sag exceeds 5% of the driver's Xmax, it's not recommended for a down-firing subwoofer application."
I noticed you have artsy concrete and stainless pipe speaker stands, could you make stands that are the subwoofers? They wouldn’t be as light looking but would be practical. Actually, I like your invisible subwoofer idea and it looks like a great project. As an aside, I’ve had downward firing subs for years and they didn’t sag at all. I’m planning a set of IB subs in my crawl space that would be upward firing but the q has to be .7 so maybe that won’t work. Also, all woofers are going to sag, you can’t fight gravity. It’s the law! I remember a calculator to see if a woofer would work in down firing orientation.I found some math:
You can calculate the sag of a driver from:
Percentage of Sag = 24,849 / ( Xmax * Fs²)
where
- 24,849 is a constant value based on the relationship of acceleration due to gravity and Pi.
- Xmax is the maximum linear excursion of a loudspeaker voice coil while remaining within the magnetic flux field (mm).
- Fs is the free-air resonant frequency of the woofer (Hz).
The following is the same formula, including the relationship of acceleration and Pi in this case:
Percentage of Sag = 981,000 / (Xmax * (2 * Pi * Fs)²)
where
- 981,000 is acceleration due to gravity (mm/S²) * 100 (for the percentage).
- Xmax is the maximum linear excursion of a loudspeaker voice coil while remaining within the magnetic flux field (mm). - Fs is the free-air resonant frequency of the woofer (Hz)
- Pi is the ratio of a circle's circumference and diameter, usually rounded to 3.14.
As a general rule of thumb, any time the sag exceeds 5% of the driver's Xmax, it's not recommended for a down-firing subwoofer application."
@Arthur Jackson
Now everything is finished you come up with another solution
Just kidding.
With using REW for the optimal intergration of the subs with the satellites and running RoomPerfect after that I have great sound.
The sag seems minimal with Fs 24,2 Hz en Xmax 14,3 mm I get a value of 0,00296 so I guess it is approx. 0,3 % ?
Now everything is finished you come up with another solution
Just kidding.
With using REW for the optimal intergration of the subs with the satellites and running RoomPerfect after that I have great sound.
The sag seems minimal with Fs 24,2 Hz en Xmax 14,3 mm I get a value of 0,00296 so I guess it is approx. 0,3 % ?
@maxolini
First I did some measurements with REW and noticed a significant peak around 43 Hz which I already smoothed with the equaliser settings in the Lyngdorf amp.
One thing that most people don't realise is that RoomPerfect calibration does not automatically correct the timing between the subs and satellites and you have to set it via the menu where you enter the measured distances from the listening (Focus) position to the different speakers.
With REW doing a sweep from 20 - 400 Hz you can clearly see the integration of the subs if you have a smooth transition in the crossover frequency.
I play around with the set distances when I get the best picture with the least big dips around the crossover frequency (this area is approx. 50% - 200% of the set crossover freq) and once I am satisfied I start RoomPerfect.
It can really make a difference if you set the x-over for the subs at another frequency and slope compared to the satellites.
This makes you have many variables to play with but some can really make a nicer x-over picture in REW than setting both at the same frequency and same slope.
Once I was really super happy with the sound, was fabulous but had the subs with a shallow slope with -6dB at 200 Hz.
This made the subs extend to much into the higher frequencies and you could locate the subs because of this.
Almost perfect sound but still something to desired
Back to the settings and tried again ... quite happy now but still think the whole 3D stage and "being there" was still better with these first settings.
I will do some more experimenting but by doing I am still learning and understand the behaviour better
First I did some measurements with REW and noticed a significant peak around 43 Hz which I already smoothed with the equaliser settings in the Lyngdorf amp.
One thing that most people don't realise is that RoomPerfect calibration does not automatically correct the timing between the subs and satellites and you have to set it via the menu where you enter the measured distances from the listening (Focus) position to the different speakers.
With REW doing a sweep from 20 - 400 Hz you can clearly see the integration of the subs if you have a smooth transition in the crossover frequency.
I play around with the set distances when I get the best picture with the least big dips around the crossover frequency (this area is approx. 50% - 200% of the set crossover freq) and once I am satisfied I start RoomPerfect.
It can really make a difference if you set the x-over for the subs at another frequency and slope compared to the satellites.
This makes you have many variables to play with but some can really make a nicer x-over picture in REW than setting both at the same frequency and same slope.
Once I was really super happy with the sound, was fabulous but had the subs with a shallow slope with -6dB at 200 Hz.
This made the subs extend to much into the higher frequencies and you could locate the subs because of this.
Almost perfect sound but still something to desired
Back to the settings and tried again ... quite happy now but still think the whole 3D stage and "being there" was still better with these first settings.
I will do some more experimenting but by doing I am still learning and understand the behaviour better
That's a very detailed response 🙂
I'm not in to HT sound but PA
One trick i learned from over the years to time align the satellites (we call them tops in PA)
Without REW and measurments Mic.
By ear and if lucky with an SPL meter
Is to play a sine wave around the xover freq
Most of the time I used 100 or 80 Hz
So you flip the phase on the subs, if your DSP processor can do it you did it from there otherwise just flip the cables on the amp feeding the subs.
So most of the time I need to delay the tops cause the subs was arriving later to the listening area, but it can be the other war around depending the layout of the stage.
Other times when stacking the tops over the subs and using a horn type sub you need to delay the tops also to time align the subs in relation to the tops.
As you know the horn have a longer path to the outside so the sub freq arrive later an that cause phase problems.
Or when mixing bass reflex with horn types you need to delay the BR boxes to be time aligned with the horns subs.
So, before flipping polarity you play a tone and more less balance the level between tops and subs to the the loudest , they will sum even with some out of phase.
Now you flip the subs polarity and then start to delay the tops with your DSP , is between 1 to 4 or 5ms , as you start adding delay and getting close to the right delay time the tone will get lower in amplitude (volume), so you sweep around those times till you get the most attenuation , you can do it by ear although not 100 accurate it will be very close and much best that not delay at all.
If you have an SPL meter use it otherwise play by ear, so when you found the very setting in that cause the lowest SPL , save the setting and call it a day.
When time aligned there is almost zero phase shift between the 2 freq bands, and if the polarity is flip it cause the most cancellation cause 2 time aligned sine waves that are out of phase 180 degrees nulls out each other. But cause reflections and room modes you still hear some , if you do that on a DAW you get perfect silence as they null out 100% inside the editing software.
So you put back your subs to normal polarity and there you go, better than have the system phase misaligned.
Sound system will sound punchier and more even.
Regarding your subs than can be localized in space is because they are spitting out to much mid lows , you mentioned 200 Hz slope that made the subs extend to much in to the higher freq , don't know what DSP you are using but that must not happen , you need to properly high pass and low pass your subs so they do not play lower to prevent driver damage by over excursion and don't play higher to prevent overlapping with the tops making nasty peaks and causing that you can localize them in space .
Subs must be felt not seen
I'm not in to HT sound but PA
One trick i learned from over the years to time align the satellites (we call them tops in PA)
Without REW and measurments Mic.
By ear and if lucky with an SPL meter
Is to play a sine wave around the xover freq
Most of the time I used 100 or 80 Hz
So you flip the phase on the subs, if your DSP processor can do it you did it from there otherwise just flip the cables on the amp feeding the subs.
So most of the time I need to delay the tops cause the subs was arriving later to the listening area, but it can be the other war around depending the layout of the stage.
Other times when stacking the tops over the subs and using a horn type sub you need to delay the tops also to time align the subs in relation to the tops.
As you know the horn have a longer path to the outside so the sub freq arrive later an that cause phase problems.
Or when mixing bass reflex with horn types you need to delay the BR boxes to be time aligned with the horns subs.
So, before flipping polarity you play a tone and more less balance the level between tops and subs to the the loudest , they will sum even with some out of phase.
Now you flip the subs polarity and then start to delay the tops with your DSP , is between 1 to 4 or 5ms , as you start adding delay and getting close to the right delay time the tone will get lower in amplitude (volume), so you sweep around those times till you get the most attenuation , you can do it by ear although not 100 accurate it will be very close and much best that not delay at all.
If you have an SPL meter use it otherwise play by ear, so when you found the very setting in that cause the lowest SPL , save the setting and call it a day.
When time aligned there is almost zero phase shift between the 2 freq bands, and if the polarity is flip it cause the most cancellation cause 2 time aligned sine waves that are out of phase 180 degrees nulls out each other. But cause reflections and room modes you still hear some , if you do that on a DAW you get perfect silence as they null out 100% inside the editing software.
So you put back your subs to normal polarity and there you go, better than have the system phase misaligned.
Sound system will sound punchier and more even.
Regarding your subs than can be localized in space is because they are spitting out to much mid lows , you mentioned 200 Hz slope that made the subs extend to much in to the higher freq , don't know what DSP you are using but that must not happen , you need to properly high pass and low pass your subs so they do not play lower to prevent driver damage by over excursion and don't play higher to prevent overlapping with the tops making nasty peaks and causing that you can localize them in space .
Subs must be felt not seen
I know the inversing polarity trick to find the correct delay settings.
Depending on the set x-over frequency you can get completly different outcome in REW as depending on what (sub or satellites) is creating the sound at a cretain frequency will give other side effects, some nasty, some not.
With reversing polarity for the lowest volume you will never be aware of the effects at slightly higher frequencies for example.
If I set x-over at 60 Hz for example 80 Hz will largely come from the satellites and with x-over at 100 Hz the subs will mostly take care of 80 Hz.
Both situations will create a different frequency response with other peaks and dips depending on the reflections of the dominant driver At this 80 Hz.
Lyngdorf allows different x-over frequency and slope for the subs and for the satellites.
You couldd use a steep slope for the subs at 80 Hz and a shallow slope at 100 Hz for the satellites for example.
It is worth to experiment but with all those variables it can take some time
That the subs were cut off too high in frequency was not due to DSP as I have to set the x-over in advance.
Calibration will only take care that the summing due to mutual frequencies in the subs and satellites are smoothed out for a straight frequency response.
It was my own fault setting the x-over frequency in combination with the slope extending too far but only noticed it after the RoomPerfect calibration when I started listening to music.
Depending on the set x-over frequency you can get completly different outcome in REW as depending on what (sub or satellites) is creating the sound at a cretain frequency will give other side effects, some nasty, some not.
With reversing polarity for the lowest volume you will never be aware of the effects at slightly higher frequencies for example.
If I set x-over at 60 Hz for example 80 Hz will largely come from the satellites and with x-over at 100 Hz the subs will mostly take care of 80 Hz.
Both situations will create a different frequency response with other peaks and dips depending on the reflections of the dominant driver At this 80 Hz.
Lyngdorf allows different x-over frequency and slope for the subs and for the satellites.
You couldd use a steep slope for the subs at 80 Hz and a shallow slope at 100 Hz for the satellites for example.
It is worth to experiment but with all those variables it can take some time
That the subs were cut off too high in frequency was not due to DSP as I have to set the x-over in advance.
Calibration will only take care that the summing due to mutual frequencies in the subs and satellites are smoothed out for a straight frequency response.
It was my own fault setting the x-over frequency in combination with the slope extending too far but only noticed it after the RoomPerfect calibration when I started listening to music.
Well I don't have none or almost none of the room messed up interactions as mostly is live PAI know the inversing polarity trick to find the correct delay settings.
Depending on the set x-over frequency you can get completly different outcome in REW as depending on what (sub or satellites) is creating the sound at a cretain frequency will give other side effects, some nasty, some not.
With reversing polarity for the lowest volume you will never be aware of the effects at slightly higher frequencies for example.
If I set x-over at 60 Hz for example 80 Hz will largely come from the satellites and with x-over at 100 Hz the subs will mostly take care of 80 Hz.
Both situations will create a different frequency response with other peaks and dips depending on the reflections of the dominant driver At this 80 Hz.
Lyngdorf allows different x-over frequency and slope for the subs and for the satellites.
You couldd use a steep slope for the subs at 80 Hz and a shallow slope at 100 Hz for the satellites for example.
It is worth to experiment but with all those variables it can take some time
That the subs were cut off too high in frequency was not due to DSP as I have to set the x-over in advance.
Calibration will only take care that the summing due to mutual frequencies in the subs and satellites are smoothed out for a straight frequency response.
It was my own fault setting the x-over frequency in combination with the slope extending too far but only noticed it after the RoomPerfect calibration when I started listening to music.