Hi PASC
Great information, thanks!
Restricting the throat is something that has been discussed in this thread. It looks like it could be needed when running these drivers at such high powers.
I am undecided which way would be the best way to restrict the throat, especially when concidering the need to let the driver reach its Xmech limits.
It might be good to retest the Keystone with the dB meter with a fresh 9V battery to check its SPL.
Kind Regards
Martin (Xoc1)
Great information, thanks!
Restricting the throat is something that has been discussed in this thread. It looks like it could be needed when running these drivers at such high powers.
I am undecided which way would be the best way to restrict the throat, especially when concidering the need to let the driver reach its Xmech limits.
It might be good to retest the Keystone with the dB meter with a fresh 9V battery to check its SPL.
Kind Regards
Martin (Xoc1)
Do we care to discuss such high SPL claims from pro boxes such as the DSL TH118?
Is it really all marketing trickery??
Peak 4 cabinets for the TH-18 should be 138.2db according to PASC's measurement for one cab. That lines up pretty well with the 140db 4 cabinet goal set by You in the beginning. Great job Martin!!
Is it really all marketing trickery??
Peak 4 cabinets for the TH-18 should be 138.2db according to PASC's measurement for one cab. That lines up pretty well with the 140db 4 cabinet goal set by You in the beginning. Great job Martin!!
Last edited:
Hi!
For better ref I attach a picture from the way I put two new slots that let the botton quiet in spite of Pe applied to 18NLW9600 during the tests.
Thinking if it helps fix similar slots to the sides from the front/mouth to the back to avoid any vibration at that points.
Could be thinner solid wood slots as teached by Weems in Designing Building and testing your Own speakers Systems
regards,
For better ref I attach a picture from the way I put two new slots that let the botton quiet in spite of Pe applied to 18NLW9600 during the tests.
An externally hosted image should be here but it was not working when we last tested it.
Thinking if it helps fix similar slots to the sides from the front/mouth to the back to avoid any vibration at that points.
Could be thinner solid wood slots as teached by Weems in Designing Building and testing your Own speakers Systems
regards,
Xoc1: "Restricting the throat is something that has been discussed in this thread. It looks like it could be needed when running these drivers at such high powers. I am undecided which way would be the best way to restrict the throat, especially when concidering the need to let the driver reach its Xmech limits."
Hi Martin,
The problem of losing control over the excursion in that last power up stage is a driver problem. The only way to suppress it, while keeping the max SPL intact, is designing a TH in such way the excursion dips fall together (just a little above) with the most important resonances in the cone. If you want to go a step further you'll have to use dynamic eq on these spots.
Although the driver will never reach its Xmech, you are right in that the cone needs to move freely to obtain max SPL with minimal change in response, at high levels.
I am undecided which way would be the best way to restrict the throat, especially when concidering the need to let the driver reach its Xmech limits."Hi Martin,
The problem of losing control over the excursion in that last power up stage is a driver problem. The only way to suppress it, while keeping the max SPL intact, is designing a TH in such way the excursion dips fall together (just a little above) with the most important resonances in the cone. If you want to go a step further you'll have to use dynamic eq on these spots.
Although the driver will never reach its Xmech, you are right in that the cone needs to move freely to obtain max SPL with minimal change in response, at high levels.
Last edited:
Xoc1: "Restricting the throat is something that has been discussed in this thread. It looks like it could be needed when running these drivers at such high powers.
Hi Martin,
The problem of losing control over the excursion in that last power up stage is a driver problem. The only way to suppress it, while keeping the max SPL intact, is designing a TH in such way the excursion dips fall together (just a little above) with the most important resonances in the cone. If you want to go a step further you'll have to use dynamic eq on these spots.
Although the driver will never reach its Xmech, you are right in that the cone needs to move freely to obtain max SPL with minimal change in response, at high levels.
Restricting the throat is likely to cause severe power compression before it helps much with cone excursion -- you really want a strong nonlinear restoring force (rapidly increasing suspension stiffness) to limit cone travel, not an almost-linear throat restriction.
If this is a real problem then you either need drivers with more Xmax/Xmech (eg. Ficar BTL N218, Xmax=28mm) or "unbottomable" ones with a very rapid decrease in compliance at large excursions -- it's been said that the B&C drivers are almost impossible to drive to Xmech even with huge input powers.
Hi mRgSr,
Marketing maybe, but as long people 'forget' to read the additional info, such as measured in 1/2Pi, it's not Danley who you should blame.
It's actually similar to what happens here on DIY. Members around here compare TH's with different drivers, based on 2,83V instead of the 1W/1m. That means the driver with the highest Re will show a lower sensitivity response compared to a driver with the lowest Re.
Hi Iand,
I agree in most part. Yes, restricting the mouth or the path will result in extra (Dynamic) Power Compression. But you can design a TH in such way to restore one or max 2dB's (not over the entire bandwidth but at 'problem spots') that will be lost by Dynamic Power Compression in non-optimised TH's.
Last edited:
I had a debate with Tom about this when he started to use lower impedances; his view was that so long as the spec was clear (e.g. 105dB for 2.82V, 4 ohms impedance) this was not an issue, especially given that most (all?) amplifiers are voltage-limited -- you just need to know how many you can drive off one amplifier.
Since some of his subs are 2 ohms nominal they appear even more "efficient" if you don't read the details. But it would be more honest if he included impedance in the comparison tables like this as well as the headline figures...
Danley | Tapped Horn Comparison
All that means is you've lost 1-2dBs off a figure that is higher to start with -- either way you're getting lower output than you could get.
It's still the wrong approach in my view, you want something that doesn't reduce output with normal drive but limits excursion under extreme drive, and throat restriction will never be able to do this, it's down to the driver.
Hi Iand,
With "design in such way" I didn't mean "throat restricting" but designing in such way that the excursion dips fall together with the most important cone resonances.
Last edited:
I'd hope you don't have any inband "cone resonances" apart from the fundamental driver resonance Fs !
For tapped horns the ideal driver Fs is higher than the 3dB point of the speaker, usually by at least 50% -- not only does this reduce excursion below cutoff, it acts like reactance cancellation in a FLH (small rear chamber) and gives a little bit more extension.
You can see this if you change the driver T-S physical parameters to just change compliance (also alters Fs, Vas, Qes).
Unfortunately most PA drivers don't have stiff enough suspension to meet this requirement, this would typically mean Fs of 50-60Hz (or even higher) would be ideal for TH with 33-40Hz cutoffs.
Maybe it would be interesting to google for "cone losses" and "Diagonal nodal and Circle Nodal" resonances in the cone. These are all Fs related that is true. You can even measure them since these 'spots' normally turn up as THD peaks in your measurements.
Indeed, but if any of these happen in the "sub" range I'd be very worried, this would mean the driver cone is far too flexible for a high-power TH application...
'They' happen in every driver and in any type of loading. Tapped Horns however suffer most from them. Danley warned from the beginning, TH's don't have an 'air-spring' like the closed back of a traditional FLH. Therefore the suspension has to deal with all non linearity's. Diagonal nodal and Circle nodal (not all but some of them) cause the biggest non-linear behaviour in TH's. The suspension of PA drivers, more importantly the internal damping of the cone, can deal with them relative good accept in the last stage towards max SPL (relative good because even at low excursion you can find higher THD figures at these spots).
Tighten the suspension not only raises the Fs of the driver but also drops the internal damping values of the cone. In that case you have to strengthen the cone which means automatically higher its weight. That will result in a drop of the Fs and sensitivity. Again you have to compensate that with a stronger motor and so on....
The problem is that when you start changing just one figure/value in a driver, the whole system of the driver becomes unbalanced. It's all a matter of balance.
One last thing, if you higher the Fs of the driver it will probably end exactly on the excursion peak. Try to imagine what will happen, max excursion that falls together with the fundamental of a driver...
Edit: Iand, it's not personally meant but all I get in general are only two counter arguments: 'hope not' and 'don't believe'. Of course I might be completely wrong which I'm willing to accept but think about this, resonances in material don't go from 0% to 100% all of the sudden. It's not a gradual linear process either. But if someone agrees they can happen, he also have to admit they start earlier, before they destroy a cone.
Last edited:
The free-air Fs of a driver and whether it coincides with the excursion peak or not really makes no difference, it's the loaded Fs (with air load) in a TH that matters, and this is always below the horn cutoff.
The cone travel doesn't suddenly go crazy at Fs, it's just that this is where all the damping comes from back EMF (this happens at all the impedance peaks) -- unless the driver has a tiny magnet (high Qes) it doesn't really matter where Fs is, except to control excursion below cutoff.
No sub driver with a cone rigid enough to be driven to high PA-type levels in a TH should have any cone resonances within the passband, you'd expect such drivers to behave as pistons up to at least 300Hz or so.
Tom's comment about not having an air spring is exactly what I referred to; in a FLH a small sealed rear chamber is normally used to reduce the compliance and give reactance cancellation, with a TH you need a driver with higher Fs to do the same job -- the stiffer suspension has the same effect as a small rear chamber.
Of course the suspension now provides all the restoring force (no air spring) so it has to be reasonably linear up to at least Xmax, this makes the driver suspension design harder especially if it also has to limit cone excursion to Xmech.
Thanks Iand for taking your time and I’ll try to answer.
"The free-air Fs of a driver and whether it coincides with the excursion peak or not really makes no difference, it's the loaded Fs (with air load) in a TH that matters, and this is always below the horn cutoff."
I agree, until the point where the cone is no longer able to damp the energy coming from circle nodal and diagonal nodal resonances or to transfer it into the suspension.
"No sub driver with a cone rigid enough to be driven to high PA-type levels in a TH should have any cone resonances within the passband, you'd expect such drivers to behave as pistons up to at least 300Hz or so."
Pro LF PA drivers should be dealing with these non-linearity’s up till at least Xmax and they do! Put them in a basreflex and you will see they do up to at least their AES power rating figures (Xmax is a different story).
"Tom's comment about not having an air spring is exactly what I referred to; in a FLH a small sealed rear chamber is normally used to reduce the compliance and give reactance cancellation, with a TH you need a driver with higher Fs to do the same job -- the stiffer suspension has the same effect as a small rear chamber."
I have looked it up but I’m not sure if you are referring to this passage of Danley: "With the Tapped horn, one has no rear volume so to get the lowest low corner, one needs the driver’s Fs to be somewhat above the low cutoff. For example, it wouldn’t be unusual to have an Fs of 45 or 50Hz with a low corner of 30Hz."
I make quiet some mistakes in English so correct me if my interpretation is wrong. "it wouldn’t be unusual" doesn’t mean it has to be. Actually I think it’s more of an indication of what is possible with a TH or maybe even likely. Also the statement seems to be quiet old and I actually wondering what Danley thinks of this subject today in relation to TH’s optimised for PA. Especially after the TH118 development stages.
If Danley really favour higher suspension values over higher internal cone damping values why doesn’t he use them in his TH118? The B&C18SW115, with its 33Hz fundamental resonance, can’t be considered as a typical high Fs driver optimised like you suggested. The -3dB point is above the driver Fs which is another contradiction. The B&C18SW115 seem to have almost no competition from any other true PA driver. In my view, and again this is my personal view, the 18SW115 is the outstanding example of a balance in favour of higher internal cone damping values. If that is true, doesn't that proof that cone damping factor is more important than higher suspension values for TH’s optimised for PA?
I think the best competition so far comes from 18Sound, 18NLW9600 and the 18NLW9600c. The C stands for carbon. 18Sound deliberately choose carbon fibre to improve cone damping values. Again, I don’t consider its 35Hz fundamental (and 34Hz for 9600 serie) very high in PA terms. For the rest, both drivers are pretty much equal in design. If cone resonances wouldn’t be important to suppress there wouldn’t be much difference between the two. It would also be very uneconomic to invest in this research and specialised machinery to develop these carbon cones. However, the carbon cone version seems to be doing what it is designed for.
To make no misunderstanding, I’m trying to find the right answer for what happens in the last stage of powering up PA drivers in a TH. Where you can 'light' them up till max AES power ratings in basreflex cabs, you can’t in TH’s (although very near). I my view you can blame it on the circle nodal and diagonal nodal resonances because that is how cones usually are getting destroyed. The physical evidence can been found on these destroyed cones while there is no physical evidence of over-excursion.
"The free-air Fs of a driver and whether it coincides with the excursion peak or not really makes no difference, it's the loaded Fs (with air load) in a TH that matters, and this is always below the horn cutoff."
I agree, until the point where the cone is no longer able to damp the energy coming from circle nodal and diagonal nodal resonances or to transfer it into the suspension.
"No sub driver with a cone rigid enough to be driven to high PA-type levels in a TH should have any cone resonances within the passband, you'd expect such drivers to behave as pistons up to at least 300Hz or so."
Pro LF PA drivers should be dealing with these non-linearity’s up till at least Xmax and they do! Put them in a basreflex and you will see they do up to at least their AES power rating figures (Xmax is a different story).
"Tom's comment about not having an air spring is exactly what I referred to; in a FLH a small sealed rear chamber is normally used to reduce the compliance and give reactance cancellation, with a TH you need a driver with higher Fs to do the same job -- the stiffer suspension has the same effect as a small rear chamber."
I have looked it up but I’m not sure if you are referring to this passage of Danley: "With the Tapped horn, one has no rear volume so to get the lowest low corner, one needs the driver’s Fs to be somewhat above the low cutoff. For example, it wouldn’t be unusual to have an Fs of 45 or 50Hz with a low corner of 30Hz."
I make quiet some mistakes in English so correct me if my interpretation is wrong. "it wouldn’t be unusual" doesn’t mean it has to be. Actually I think it’s more of an indication of what is possible with a TH or maybe even likely. Also the statement seems to be quiet old and I actually wondering what Danley thinks of this subject today in relation to TH’s optimised for PA. Especially after the TH118 development stages.
If Danley really favour higher suspension values over higher internal cone damping values why doesn’t he use them in his TH118? The B&C18SW115, with its 33Hz fundamental resonance, can’t be considered as a typical high Fs driver optimised like you suggested. The -3dB point is above the driver Fs which is another contradiction. The B&C18SW115 seem to have almost no competition from any other true PA driver. In my view, and again this is my personal view, the 18SW115 is the outstanding example of a balance in favour of higher internal cone damping values. If that is true, doesn't that proof that cone damping factor is more important than higher suspension values for TH’s optimised for PA?
I think the best competition so far comes from 18Sound, 18NLW9600 and the 18NLW9600c. The C stands for carbon. 18Sound deliberately choose carbon fibre to improve cone damping values. Again, I don’t consider its 35Hz fundamental (and 34Hz for 9600 serie) very high in PA terms. For the rest, both drivers are pretty much equal in design. If cone resonances wouldn’t be important to suppress there wouldn’t be much difference between the two. It would also be very uneconomic to invest in this research and specialised machinery to develop these carbon cones. However, the carbon cone version seems to be doing what it is designed for.
To make no misunderstanding, I’m trying to find the right answer for what happens in the last stage of powering up PA drivers in a TH. Where you can 'light' them up till max AES power ratings in basreflex cabs, you can’t in TH’s (although very near). I my view you can blame it on the circle nodal and diagonal nodal resonances because that is how cones usually are getting destroyed. The physical evidence can been found on these destroyed cones while there is no physical evidence of over-excursion.
Last edited:
One small correction before I get accused of something, with balance I don’t mean the suspension is less valuable in this whole story. It's more like, what is it worth having high value suspension if the cone is not able to resist the forces from internal resonances.
The 18SW115 is very balanced in that way and its not the first time. About 15 years ago Monacor (I know
) released a driver under the name SPH-450TC. It’s cone capabilities were abnormal at that time and are able to resist almost every force. Back then there were not many drivers with those specific physical capabilities. I still wonder about the true origin of this ‘old’ beast since Monacor doesn't produce drivers themselves. Although it has low sensitivity figures over the entire bandpass it still beats many drivers down low. That's what I meant with the right balance in my view and they all seem to perform excellent, if not outstanding, in TH's for PA purposes.
of something, with balance I don’t mean the suspension is less valuable in this whole story. It's more like, what is it worth having high value suspension if the cone is not able to resist the forces from internal resonances.The 18SW115 is very balanced in that way and its not the first time. About 15 years ago Monacor (I know
) released a driver under the name SPH-450TC. It’s cone capabilities were abnormal at that time and are able to resist almost every force. Back then there were not many drivers with those specific physical capabilities. I still wonder about the true origin of this ‘old’ beast since Monacor doesn't produce drivers themselves. Although it has low sensitivity figures over the entire bandpass it still beats many drivers down low. That's what I meant with the right balance in my view and they all seem to perform excellent, if not outstanding, in TH's for PA purposes.
Last edited:
