I got a pair of Voigt pipes for free. They are in 19mm chipboard exept for the front and back that is reinforced with an additional layer of 22 mm chipboard.
Internal measures are in mm.
Hight: 950
Depth: 310
With: 300
Hight of pipe opening:100
Length of first internal baffle:70
Length of second internal baffle: 665
Driver cutout: Diameter 190, upper edge 10mm below the internal top surface.
Front edge of the large internal baffle is 170 mm below internal top surface and 170mm behind front baffle innner surface.
The corner reflectors reach 110 mm from the corner in each direction.
The position of the 10mm felt that covers front, top and back is indicated by the masking tape. The with of the masking tape does not denote the thickness of the felt but rather reflect what tape I had at hand.
There also is a felt pad behind the driver on the internal baffle. To this I added felt to the sidewalls. Covering the sidewall back to a depth equal to the internal baffle and down to about 30 mm below the end of the felt pad on the internal baffle.
The pipe is tuned to about 45 Hz and with this amount of damping the harmoinics is quite unimpeded.
This time I will not try out many drivers and probably also make few impedance measurements. It behaves as other pipes I have tried. The drivers with low Q and a resonable Fr sounds the best. High Q driver sounds worse and high Q low Fr sounds the worst. So the plan is to use a driver that works well in pipes, optimise damping that is trying to tame the harmonics without losing to much of the fundamental resonance. 🙂
Internal measures are in mm.
Hight: 950
Depth: 310
With: 300
Hight of pipe opening:100
Length of first internal baffle:70
Length of second internal baffle: 665
Driver cutout: Diameter 190, upper edge 10mm below the internal top surface.
Front edge of the large internal baffle is 170 mm below internal top surface and 170mm behind front baffle innner surface.
The corner reflectors reach 110 mm from the corner in each direction.
The position of the 10mm felt that covers front, top and back is indicated by the masking tape. The with of the masking tape does not denote the thickness of the felt but rather reflect what tape I had at hand.
There also is a felt pad behind the driver on the internal baffle. To this I added felt to the sidewalls. Covering the sidewall back to a depth equal to the internal baffle and down to about 30 mm below the end of the felt pad on the internal baffle.
The pipe is tuned to about 45 Hz and with this amount of damping the harmoinics is quite unimpeded.
This time I will not try out many drivers and probably also make few impedance measurements. It behaves as other pipes I have tried. The drivers with low Q and a resonable Fr sounds the best. High Q driver sounds worse and high Q low Fr sounds the worst. So the plan is to use a driver that works well in pipes, optimise damping that is trying to tame the harmonics without losing to much of the fundamental resonance. 🙂
First test.
I used two pieces of felt 390x180 mm as a curtian covering from the ceiling down to the interal baffle. Stacked the felt is 30mm thick but nominal thickness is 10mm and then the fluffiness adds some more. It varies quite a lot from batch to batch.
Black is near field without felt and brown is with felt. The addition of felt seem to lower the Q and the frequency of the fundamental dip, some smoothing of the curve at 130 and 300 Hz.
Green no felt Blue felt pipe response. Very good taming of the 130 and 300 Hz peaks. The flattening of pipe output around the fundamental is consistent with the lowered Q of the driver near field dip.
That persistent 200 Hz peak was around with the Decca as well, I will try to find out if it is an artifact or something in common with Voigt pipes folded once...
I used two pieces of felt 390x180 mm as a curtian covering from the ceiling down to the interal baffle. Stacked the felt is 30mm thick but nominal thickness is 10mm and then the fluffiness adds some more. It varies quite a lot from batch to batch.
Black is near field without felt and brown is with felt. The addition of felt seem to lower the Q and the frequency of the fundamental dip, some smoothing of the curve at 130 and 300 Hz.
Green no felt Blue felt pipe response. Very good taming of the 130 and 300 Hz peaks. The flattening of pipe output around the fundamental is consistent with the lowered Q of the driver near field dip.
That persistent 200 Hz peak was around with the Decca as well, I will try to find out if it is an artifact or something in common with Voigt pipes folded once...
The influence of the room and position is minimal
Black:Office
Brown: Office loudspeaker upside down
Green: Living room with the speaker upside down on a chair.
Black:Office
Brown: Office loudspeaker upside down
Green: Living room with the speaker upside down on a chair.
In theory that is easy to deal with the 5th harmonic at 200 Hz. Resistive damping is most effective at velocity maximas and for that harmonic that is at 1/5 2/5 3/5 4/5 5/5 of the lenght of the pipe and. By stuffing the first 40-50% lenght of the pipe the damping material should have two velocity maxima to tame the harmonics.
At least with the Decca the 5th harmoic was rather more recalcitrant than expected.
So the tentative plan is to fill the pipe stepwise
1. Closed end to lower edge of driver
2. Closed end to the top of the interal baffle and back behind the driver to depth of the edge of the baffle board
3. Closed en of the pipe filling past the first bend all the way to the back wall
4. Filling the whole pipe.
The last option create a closed box with a resistive leak and that is not the point of a quarter wave pipe at all. I will try to keep the dip at 40-45 Hz as much as possible and try to reduce the dip/peak at 200 Hz as much as I can.
There was a similar pipe in L'Audiophile for a Triangle 6.5" years ago, I might take a look at what they did.
At least with the Decca the 5th harmoic was rather more recalcitrant than expected.
So the tentative plan is to fill the pipe stepwise
1. Closed end to lower edge of driver
2. Closed end to the top of the interal baffle and back behind the driver to depth of the edge of the baffle board
3. Closed en of the pipe filling past the first bend all the way to the back wall
4. Filling the whole pipe.
The last option create a closed box with a resistive leak and that is not the point of a quarter wave pipe at all. I will try to keep the dip at 40-45 Hz as much as possible and try to reduce the dip/peak at 200 Hz as much as I can.
There was a similar pipe in L'Audiophile for a Triangle 6.5" years ago, I might take a look at what they did.
I changed the driver to a OEM driver with a light paper cone (12 grams) as the effect of resonances shows up more the lighter the cone is.
I tried felt damping using two pieces of felt 170x390x10mm. They were placed across the pipe at either as indicated by the blue line separating A and B compartments or the B and C dito.
Lets start with the pipe output
1. No felt = black line
2. Both at AB= Blue
3. One at AB one at BC = Green
4. Both at CB= Orange
Results
1. No felt = black line
2. Both at AB= Blue, least effective for the 130 Hz peak and the peaks above 250 Hz, but most effective for the 200 Hz peak
3. One at AB one at BC = Green (intermediate effect of 2 and 4)
4. Both at CB= Orange, the most effective at 130 Hz and above 250 Hz but the least effective at 200 Hz. Note the reduced cone output below 50 Hz.
Below the pipe fundamental the opening is shortening the driver felt curtains before the driver does nothing (1 and 2). Felt between the cone and the pipe opening add some resistive damping to the flapping of the cone ( 3 and 4).
The pipe results suggest that damping between the closed and and the driver is most effective for the 200 Hz peak but does very little for the higher harmonics as well as the 130 Hz peak.
Lets turn to the nearfield response at the fundamental
1. All felt dampings reduce the fundamental dip by 4-5 dB, there is no free lunch.
2. At 30 Hz two layers of felt behind the cone (orange) reduce the cone output by 2 dB. This is consistent with the pipe results.
Lastly nearfield response of harmonics (note the change in scale)
Blue (two layers of felt before the driver in the pipe at AB) is the best in supressing the 200 Hz peak and two layers behind the driver at BC the worst. While the latter showed far better damping of the 130 Hz peak and the higher harmonics in pipe output there is no real difference in near field response. The black line of the undamped pipe show som peaks and dips 250-400Hz but either felt combination will supress this effect.
My conclusion would be to concentrate on damping material between the driver and the closed end. With that set try out some final tweeking with material between the driver and open end.
Time to try out some bonded fiber stuffing!
I tried felt damping using two pieces of felt 170x390x10mm. They were placed across the pipe at either as indicated by the blue line separating A and B compartments or the B and C dito.
Lets start with the pipe output
1. No felt = black line
2. Both at AB= Blue
3. One at AB one at BC = Green
4. Both at CB= Orange
Results
1. No felt = black line
2. Both at AB= Blue, least effective for the 130 Hz peak and the peaks above 250 Hz, but most effective for the 200 Hz peak
3. One at AB one at BC = Green (intermediate effect of 2 and 4)
4. Both at CB= Orange, the most effective at 130 Hz and above 250 Hz but the least effective at 200 Hz. Note the reduced cone output below 50 Hz.
Below the pipe fundamental the opening is shortening the driver felt curtains before the driver does nothing (1 and 2). Felt between the cone and the pipe opening add some resistive damping to the flapping of the cone ( 3 and 4).
The pipe results suggest that damping between the closed and and the driver is most effective for the 200 Hz peak but does very little for the higher harmonics as well as the 130 Hz peak.
Lets turn to the nearfield response at the fundamental
1. All felt dampings reduce the fundamental dip by 4-5 dB, there is no free lunch.
2. At 30 Hz two layers of felt behind the cone (orange) reduce the cone output by 2 dB. This is consistent with the pipe results.
Lastly nearfield response of harmonics (note the change in scale)
Blue (two layers of felt before the driver in the pipe at AB) is the best in supressing the 200 Hz peak and two layers behind the driver at BC the worst. While the latter showed far better damping of the 130 Hz peak and the higher harmonics in pipe output there is no real difference in near field response. The black line of the undamped pipe show som peaks and dips 250-400Hz but either felt combination will supress this effect.
My conclusion would be to concentrate on damping material between the driver and the closed end. With that set try out some final tweeking with material between the driver and open end.
Time to try out some bonded fiber stuffing!
bjorno posted this in fullrange forum, http://www.diyaudio.com/forums/full-range/249977-another-voigt-pipe.html#post3798000
Am doing similar testing atm with an MLTL. 🙂
Am doing similar testing atm with an MLTL. 🙂
I tried 100grams of BAF in the closed end of the pipe.
Black trace=No damping at all (exept the felt as indicated in previous posts).
Brown= stuffing crammed into the first 20 cm of the pipe
Red= distributet up to 40 cm from closed end.
Yellow= spread evenly along the dividning board all the way up to the upper edge of the internal baffle, about 60 cm from the closed end.
Maximal damping should occur with damping material at the velociy maxima of the resonances so for the 5th harmonic about 35-40 cm from the closed end and the other ones closer that that. Lets start with pipe output:
The undamped pipe has severe peak and dips from 100 to 500 Hz (black)
The 20cm stuffing (Brown) helps in evening out the peaks and dips.
The 40 cm variant (Red) looks even better.
Then the 60 cm (yellow) is much worse?
How about the influcence at the fundamental resonance?
Here the 20 and 40 cm stuffing reduce the fundamental by 2 dB but the 60 cm by about 5 dB.
Then turning to harmonics
Here the differences are less clearcut. The 40 and 20 cm looks very similar and the 60 cm looks worse again but not by such a large margin.The huge 3rd harmoic at 130 Hz does little to the driver due to its location. The 200 Hz peak is the major influence and to a lesser degree the 300 and 400 Hz peaks.
The results are not what I expected perhaps it is not damping per se that is important but causing phase shifts that moves velocity minima away from the position of the driver?
With all resonances having pressure maxima/velocity minima at the very end of the closed end of the pipe, why have any damping material there at all? Should it not start 10-15 cm out from the start the ninth harmonic shuould have its first veolocity maxima about 20cm from the end and the lower harmonics even further out...
Black trace=No damping at all (exept the felt as indicated in previous posts).
Brown= stuffing crammed into the first 20 cm of the pipe
Red= distributet up to 40 cm from closed end.
Yellow= spread evenly along the dividning board all the way up to the upper edge of the internal baffle, about 60 cm from the closed end.
Maximal damping should occur with damping material at the velociy maxima of the resonances so for the 5th harmonic about 35-40 cm from the closed end and the other ones closer that that. Lets start with pipe output:
The undamped pipe has severe peak and dips from 100 to 500 Hz (black)
The 20cm stuffing (Brown) helps in evening out the peaks and dips.
The 40 cm variant (Red) looks even better.
Then the 60 cm (yellow) is much worse?
How about the influcence at the fundamental resonance?
Here the 20 and 40 cm stuffing reduce the fundamental by 2 dB but the 60 cm by about 5 dB.
Then turning to harmonics
Here the differences are less clearcut. The 40 and 20 cm looks very similar and the 60 cm looks worse again but not by such a large margin.The huge 3rd harmoic at 130 Hz does little to the driver due to its location. The 200 Hz peak is the major influence and to a lesser degree the 300 and 400 Hz peaks.
The results are not what I expected perhaps it is not damping per se that is important but causing phase shifts that moves velocity minima away from the position of the driver?
With all resonances having pressure maxima/velocity minima at the very end of the closed end of the pipe, why have any damping material there at all? Should it not start 10-15 cm out from the start the ninth harmonic shuould have its first veolocity maxima about 20cm from the end and the lower harmonics even further out...
Last edited:
- Status
- Not open for further replies.