Hi Djim,
I don't have much to add to what jbell has already said (e.g.: Post #157), for me the description of screamersusa's Th, system and use (or abuse) is sufficient to at least look more deeply into this. Maybe it would help if we wouldn't call it a compression chamber, but a coupling chamber, a device to couple the large diaphragm of the driver to the reduced cross-sectional area of the horn @ S2; then it also becomes more obvious, that the shape of the opening will work together with the shape of the horn to accomplish that coupling without creating uneven loading or pressure induced oscillation of the cone. It would also make it more obvious when there is not enough room for the driver to reach maximum excursion. The other thing, in this particular case, is that a compression ration of more than 4:1 may just be a bit on the high side for the cone, considering the maximum power this driver's motor can handle. So you have high compression, uneven loading, possible physical contact...(Just pondering along
.).
Regards,
I don't have much to add to what jbell has already said (e.g.: Post #157), for me the description of screamersusa's Th, system and use (or abuse) is sufficient to at least look more deeply into this. Maybe it would help if we wouldn't call it a compression chamber, but a coupling chamber, a device to couple the large diaphragm of the driver to the reduced cross-sectional area of the horn @ S2; then it also becomes more obvious, that the shape of the opening will work together with the shape of the horn to accomplish that coupling without creating uneven loading or pressure induced oscillation of the cone. It would also make it more obvious when there is not enough room for the driver to reach maximum excursion. The other thing, in this particular case, is that a compression ration of more than 4:1 may just be a bit on the high side for the cone, considering the maximum power this driver's motor can handle. So you have high compression, uneven loading, possible physical contact...(Just pondering along
.).Regards,
I once had a set of frazier bass bins (early 1980s) that had rear firing woofers in a sort of suspended chamber, where the sound went toward the back of the cab and then made its way to the front, sord of a wierd folded horn design. Each woofer had a shallow "cone" in front of it (mounted at the rear of the cabinet, outside the chamber, facing the woofers) that dispersed the sound coming from the woofers (JBL E140s). Sounded great for the day. The cones were equidistant from the speakers and on the same plane (not tilted toward one edge or the other). I'm a believer in symmetry in all things engineering, where you are dealing with distributing force over an area. So I think Josh's design could be improved by angling the plate to be perfectly equidistant to the speaker over its area (and not tilted in relation to the speaker) as the drawing shows.
Hi Oliver,
The possible physical contact is not the first thing that comes in my mind, looking at the kind of damage that is done to the cone. This folding style was already known for its uneven loading long before tapped horns. The bigger the diameter of the driver the bigger the problem becomes.
If someone makes a detailed horizontal roll-out that includes the extra volume of the cone plus baffle cut out you could easily see the problem. When the distance between the driver baffle and its opposite panel becomes relative small compared to the diameter of the driver you automatically change the position of S2.
I agree that a coupling chamber (maybe we should call it a coupling baffle since the chamber isn’t calculated but the result of physical limitations) could help in case of the Othorn to cover for both issues. I would keep the ratio as low as possible, just enough to fix the problems and have a minimum effect on the system.
I know I always get resistance when I bring up the Fs of the driver, but in case of the Othorn the driver’s Fs is relative high with its 32Hz. It is within the usable range of the Othorn. If you look to Josh’s post #4 (on page 1), you can see the Akabak predictions of the Forces and Pressure. Together with the knowledge that the measured impedance is shifted a few Hz higher as predicted, the peaks in pressure and the forces seem to fall almost together with the Fs of the driver. People might have another opinion but I would think twice to use such combination for PA or high power situations if you care for the durability of the driver.
The possible physical contact is not the first thing that comes in my mind, looking at the kind of damage that is done to the cone. This folding style was already known for its uneven loading long before tapped horns. The bigger the diameter of the driver the bigger the problem becomes.
If someone makes a detailed horizontal roll-out that includes the extra volume of the cone plus baffle cut out you could easily see the problem. When the distance between the driver baffle and its opposite panel becomes relative small compared to the diameter of the driver you automatically change the position of S2.
I agree that a coupling chamber (maybe we should call it a coupling baffle since the chamber isn’t calculated but the result of physical limitations) could help in case of the Othorn to cover for both issues. I would keep the ratio as low as possible, just enough to fix the problems and have a minimum effect on the system.
I know I always get resistance when I bring up the Fs of the driver, but in case of the Othorn the driver’s Fs is relative high with its 32Hz. It is within the usable range of the Othorn. If you look to Josh’s post #4 (on page 1), you can see the Akabak predictions of the Forces and Pressure. Together with the knowledge that the measured impedance is shifted a few Hz higher as predicted, the peaks in pressure and the forces seem to fall almost together with the Fs of the driver. People might have another opinion but I would think twice to use such combination for PA or high power situations if you care for the durability of the driver.
Let's not lose proper perspective here. When I design a cab the first order of business after it is built is to find the limits and see what happens at high power. I care about what happens at war volumes and in worst case scenarios not at a few volts where most any system will be comfortable. In this case I deviated from the plan by adding an unknown quantity into the system and let's not forget that the signal inputs that caused the damage were likely well in excess of 200 volts which is far beyond the design goal or what will be used on a single cab. I chose to see what would happen and we see the result. It will be tested again with the plate removed...If something goes wrong there then further modification will be done until it is right. At which point they will then be put out into the wild for real world testing I'm still convinced that this is going to be a great bass cab at the end of the day. You have to break a few eggs to make an omelet.
Djim,
I personally don't think the driver resonance is of much concern at all. The driver FS is near 32Hz and the cabinet impedance measurement came in with the peaks and saddle points very close to simulated below 100Hz which puts maximum in band driver excursion near 40Hz with some variation at higher drive levels. This is 1/3rd octave. The unloaded driver FS is closer to the minimum excursion point at 27Hz than the maximum. Loading the driver in the cabinet changes the driver FS to the system FS and I would expect to see some sort of ringing, build up of energy, etc if there were truly an area of uncontrolled resonance at play. Often there will be some indications of this at various points in the response for higher order systems and undersized bass horns. I do not see this indicated in the measurements for the Othorn, which if anything seem to indicate a very well damped system comparatively. Since the driver FS is modified by loading it into a cabinet I don't follow why the free air resonance would matter? Isn't one of the conventional wisdoms of TH design to chose a driver with an FS slightly higher than your cabinet corner?
Looking at post #4 way back on page #1 what I take away from it is that there are pressure and force maximums very close to 63Hz which also correspond with a minimum in driver excursion. Further constricting the throat area only served to increase these forces as does bumping the supplied voltage from 120V in the simulation to almost twice that.
I'm still convinced that this is going to be a great bass cab at the end of the day. You have to break a few eggs to make an omelet.Djim,
I personally don't think the driver resonance is of much concern at all. The driver FS is near 32Hz and the cabinet impedance measurement came in with the peaks and saddle points very close to simulated below 100Hz which puts maximum in band driver excursion near 40Hz with some variation at higher drive levels. This is 1/3rd octave. The unloaded driver FS is closer to the minimum excursion point at 27Hz than the maximum. Loading the driver in the cabinet changes the driver FS to the system FS and I would expect to see some sort of ringing, build up of energy, etc if there were truly an area of uncontrolled resonance at play. Often there will be some indications of this at various points in the response for higher order systems and undersized bass horns. I do not see this indicated in the measurements for the Othorn, which if anything seem to indicate a very well damped system comparatively. Since the driver FS is modified by loading it into a cabinet I don't follow why the free air resonance would matter? Isn't one of the conventional wisdoms of TH design to chose a driver with an FS slightly higher than your cabinet corner?
Looking at post #4 way back on page #1 what I take away from it is that there are pressure and force maximums very close to 63Hz which also correspond with a minimum in driver excursion. Further constricting the throat area only served to increase these forces as does bumping the supplied voltage from 120V in the simulation to almost twice that.
Last edited:
Hi Josh,
You are right about the guideline from "conventional wisdoms of TH design". It should be slightly higher than the "cabinet corner". The words "cabinet corner" can be confusing. That’s why I prefer to use the words: The Fs of the driver should be slightly above the quarter wavelength resonance of the system.
I can make up from your HornResp inputs that the total path length is 3.85 metres. That means its 1/4WL = 22.3Hz. You can discuss if the Fs of 32Hz for the 21SW115 falls within the concept of "slightly above" or not. That's why I have been stating for a long time that:
The Fs of a driver in tapped horn can be best chosen between the quarter wavelength resonance of the system and the first excursion dip (or 1/3WL if you prefer).
The reasons for this Fs guideline are simple, the mechanisms behind are more complex. If the Fs is lower than its 1/4WL it ends in lower sensitivity. If the Fs is higher than the first excursion dip, the low knee (or low frequency roll-off if preferred) shifts upwards in the frequency response.
This is why your low knee starts to roll off earlier compared to its prediction. In return, the slope becomes less steep compared to its prediction. That counts for many designs here on DIY. Cone correction will not much help in that area.
The main mechanism behind it is caused by the capacitance value of air. The capacitance will increase as frequency drops. The Fs is the result of the capacitance of air in a free environment. The relation between the resonance in free air and all types of enclosures is that at point zero (= starting point = no cone movement) the capacitance value is the same. That’s why the free air resonance of drivers matters in loudspeaker design in general.
But like you said, we need to see things in perspective and you seem very pleased with the results. That's what matters in the end.
Last edited:
Testing of the Othorn outdoors is complete as of last Saturday. The good news is that the driver survived without issue which tells me that the throat pyramid that I added without securing very well caused the damage/contact to the cone. I assumed this was the case but you never know. Anyway I re ran it with the previously damaged driver which I had repaired the cone on with rubberized CA glue. I will not be needing a recone as I'm not paying a couple of hundred dollars and spending a few hours on it for the cosmetic improvement. So far the repaired driver has survived an all day festival, quite a few hours of listening to it in my PA and outdoor maximum output measurements with various burst and sweep test signals at over 120volts. I'd say it is good and long term reliability shouldn't be an issue. I think this system has a really good sound to it with the large peaks present in most tapped horns pushed up past 100Hz and the group delay and energy decay being much better than I expected. I suspect that this may be in part due to the incredible damping of the 21sw152-4 driver. I'm happy with the combination of cab size, sensitivity, output capabilities and useful extension. 25Hz is represented with real power (>122dB @ 2 meters, ground plane) and peak output potential 30-120Hz is substantial to say the least. I will have the measurements results on Data-Bass soon. Also in case no one heard I now have the input voltages used to generate the long term output compression sweeps for all passive cabs listed.
More later...I'm rather happy with how these measured and turned out overall. I've known they sound good for awhile. Measurements and reams of data will be forthcoming.
More later...I'm rather happy with how these measured and turned out overall. I've known they sound good for awhile.
Measurements and reams of data will be forthcoming.
Last edited:
Here is a teaser...I should have all of the measurements public on Data-Bass by the end of the day tomorrow. This is a pants flapper...Output capabilities 25-30Hz are what I hoped they would be and the distortion performance is very good as the distortion sweeps will later show. Dropping the output 3 to 5 dB dramatically drops THD levels. At 100 and 125Hz my sound card input was being clipped a little.
CEA-2010 burst output measured from 2 meters, ground plane, outdoors. (Roughly equivalent to 1m anechoic or full space).
CEA-2010 burst output measured from 2 meters, ground plane, outdoors. (Roughly equivalent to 1m anechoic or full space).
Last edited:
Art,
Thanks. The final long term output compression sweep measurement was at 121 volts. There is about 3 to 4 dB of compression at this level.
I do record the voltage for the compression sweeps but I never do for the bursts. I will start if I remember. Near as I can tell the burst outputs involve this much voltage at absolute minimum extrapolating from a 38.3V sweep with no HPF where there is virtually no compression occuring. There is likely to be a good amount of compression involved with the power levels used for the bursts so the voltages levels are probably quite a bit higher in reality.
16Hz 62v
20Hz 71v
25Hz 153v
31.5Hz 138v
40Hz 165v
50Hz 160v
63Hz 228v
80Hz 218v
100Hz 166v
125Hz 203v
Base sensitivity is presented here.
Judging from everything so far if I were rating this cabinet it would be something like this...
Frequency Response 27.5-93Hz +/-3dB. -10dB at 24Hz (2m ground plane, quasi anechoic)
Frequency Response with Recommended Processing 30-128Hz +/-3dB. -10dB at 25Hz (2m ground plane, quasi anechoic)
1w/1m sensitivity (1.86v) 99dB 34-93Hz
Maximum output @1m: 142dB @ 125Hz, 139dB @ 63Hz, 134dB @ 31.5Hz, 128dB @ 25Hz
Min impedance 3.4 ohms
Recommended maximum amplifier power: 6000w program (140 to 145 volts)
Recommended Processing: 24Hz Butterworth 18dB/Octave high pass, BP EQ 107Hz Q10.0 -7.5dB, BP EQ 84Hz Q2.20 +2.5dB, BP EQ 44Hz Q2.80 -1.5dB
Dimensions 24x36x36"
Weight 160 lbs
Here is what the FR looks like with 3 bands of EQ a 24Hz BW 3rd order high pass and a 4th order LR 120Hz low pass.
Thanks. The final long term output compression sweep measurement was at 121 volts. There is about 3 to 4 dB of compression at this level.
I do record the voltage for the compression sweeps but I never do for the bursts. I will start if I remember. Near as I can tell the burst outputs involve this much voltage at absolute minimum extrapolating from a 38.3V sweep with no HPF where there is virtually no compression occuring. There is likely to be a good amount of compression involved with the power levels used for the bursts so the voltages levels are probably quite a bit higher in reality.
16Hz 62v
20Hz 71v
25Hz 153v
31.5Hz 138v
40Hz 165v
50Hz 160v
63Hz 228v
80Hz 218v
100Hz 166v
125Hz 203v
Base sensitivity is presented here.
Judging from everything so far if I were rating this cabinet it would be something like this...
Frequency Response 27.5-93Hz +/-3dB. -10dB at 24Hz (2m ground plane, quasi anechoic)
Frequency Response with Recommended Processing 30-128Hz +/-3dB. -10dB at 25Hz (2m ground plane, quasi anechoic)
1w/1m sensitivity (1.86v) 99dB 34-93Hz
Maximum output @1m: 142dB @ 125Hz, 139dB @ 63Hz, 134dB @ 31.5Hz, 128dB @ 25Hz
Min impedance 3.4 ohms
Recommended maximum amplifier power: 6000w program (140 to 145 volts)
Recommended Processing: 24Hz Butterworth 18dB/Octave high pass, BP EQ 107Hz Q10.0 -7.5dB, BP EQ 84Hz Q2.20 +2.5dB, BP EQ 44Hz Q2.80 -1.5dB
Dimensions 24x36x36"
Weight 160 lbs
Here is what the FR looks like with 3 bands of EQ a 24Hz BW 3rd order high pass and a 4th order LR 120Hz low pass.
Last edited:
So I'm curious how the 21" Horn loaded Cerwin Vega cabinet claims 144db...from
A single cabinet. This cabinet should be somewhat comparable to the Cerwin cab which is 45x24x42.
The Cerwin cab is -3db at 35hz, but is it possible to attain that much more efficiency, or just marketing BS?
I'm betting on marketing BS considering no test conditions were given for the 144db...
TS-42 CERWIN VEGA FOLDED HORN SERIES SPEAKERS SUBWOOFERS
A single cabinet. This cabinet should be somewhat comparable to the Cerwin cab which is 45x24x42.
The Cerwin cab is -3db at 35hz, but is it possible to attain that much more efficiency, or just marketing BS?
I'm betting on marketing BS considering no test conditions were given for the 144db...
TS-42 CERWIN VEGA FOLDED HORN SERIES SPEAKERS SUBWOOFERS
I don't know...Don't really pay much attention to what they are claiming. Maybe it will maybe it won't. I do know this most of those peak SPL numbers are just extrapolated data usually with some fudge in there for good measure and not measurements. These are numbers from actual measurements, granted the burst output numbers require a nuclear power plant to reach which isn't really feasible to have strapped to a single cab in actual prolonged use without launching a cone or melting something...But at least they are actual measurements of what you can get short term if you do.
Thanks Djim. I have some more going up today. It often takes me weeks to get the stuff cleaned up and added to the site after doing the measurements.
I do have impedance measurements of this cab with about 1.5 volts input. I have not done the high power impedance measurements in some time. I have the LinearX VI box but not the LinearX software and the only program that I currently have that I have been able to get to work is REW but it requires 2 measurements for each voltage input and this results in even more heating of the drivers. Also I find that there are some strange shifting of the measurements that occurs depending on how hot the input level is. The outdoor sweeps may span from 2 volts to 150v in level which is a huge difference in signal strength. This requires adjusting of input gain etc... every couple of sweeps. Additionally this is yet another test rig to setup which requires another day of testing. I may still do the more interesting cabs at some point like this one but I haven't messed with it in a while.
I do have impedance measurements of this cab with about 1.5 volts input. I have not done the high power impedance measurements in some time. I have the LinearX VI box but not the LinearX software and the only program that I currently have that I have been able to get to work is REW but it requires 2 measurements for each voltage input and this results in even more heating of the drivers. Also I find that there are some strange shifting of the measurements that occurs depending on how hot the input level is. The outdoor sweeps may span from 2 volts to 150v in level which is a huge difference in signal strength. This requires adjusting of input gain etc... every couple of sweeps. Additionally this is yet another test rig to setup which requires another day of testing. I may still do the more interesting cabs at some point like this one but I haven't messed with it in a while.
Thanks Josh,
I saw the high power impedance measurements of the Gjallerhorn a while back (and your surprised notion of the 'shifts' in another post) so I was curious if you had made them of the Othorn. I agree, archiving isn’t the most inspiring job;
If 2V to 150V is giving too much variation in output level measurements why don't you use the 28,3V/10m instead? That usually keeps you within the 40dB range.
I saw the high power impedance measurements of the Gjallerhorn a while back (and your surprised notion of the 'shifts' in another post) so I was curious if you had made them of the Othorn. I agree, archiving isn’t the most inspiring job;
If 2V to 150V is giving too much variation in output level measurements why don't you use the 28,3V/10m instead? That usually keeps you within the 40dB range.
Last edited: