New 3-way, 18 inch bass, Beyma TPL150 AMT Horn

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello Derek,

What do you think of a raw 12" OB with Eq for 200 -around 800 ? Has it according to you better transcient, good snap than a sealed ?

Some good unit at Beyma for this, but high Xmax, I'm not aware if it's better to use very low Qts as well as it's not really a bass driver ! My feeling, despite the fact there are huge moves in an OB with EQ is the driver need the most important electrical damping, so better a low Qts and high BL ?

Any inputs please about that !
 
OB Vs sealed....?

Hi Eldam,

Good questions and a few caveats are needed to answer these.
In a few words, sealed is good but open baffle is best!

But only if you do it carefully and you will need a small baffle, I have used both the 10 inch Volt 2500.4 and the attached Beyma 12 inch driver.
Both needed small baffles ( VPL) and both sounded great.

The smaller 10 inch can be crossed over higher (up to about 1,200Hz and the 12 inch around 800Hz) before the off axis response starts to drop too far below the on axis response.
Some guys push up these crossover points but I think these are as high as you can go.

Re low Qts and Mms / Bl.
My preference is always for a linen surround, low Qts, low Mms / high Bl driver.
With open baffle, it is not the Qts that defines the low frequency limit, it is Fs!
As long as the driver has a good 4 inch single layer voice coil (3 inch is enough on a 10 inch, but it will have a lower max SPL) one can Eq and power it all the way down to Fs....Qts is irrelevant....Most guys dont understand this but its true!
So if you want killer 200Hz to 800Hz in a small open baffle you wont find a better driver than the attached Beyma.

Hope this helps and all the best
Derek.
PS, My attachment might be of the old 12 Nd series as its been quite a few years since I used it....The new ones are very similar I believe so will still be fantastic, maybe even better?!
 

Attachments

  • 12P80Nd Best Beyma 12 inch.pdf
    280 KB · Views: 141
Thank you Derek,

Yes it helps a lot . According to me once your source is good enough this is not the highs which are worying but more the first harmonics below 1000 Hz. I surmise the OB design to winn below the frequencies where the room is dictating its behavior (around and below 400 Hz to....the first harmonic of 99% of the reccordings):rolleyes:

Basicly it's hard to decide the trade offs here : with sufisant aera I believe I would go for direct horn, maybe Something à a VOTT with a lower XO !

But what made Juhazi member inspired by the Gradient design with this time a raw 12" for 180-600 with eq in the low iirc is maybe a good solution. But of course the effcienty for tubes users. A Beyma 12" around 100 dB - 6 dB due to the dipole and less in the low is a chalenge for transcient and damping control of the cone on smallest signals : but the lack of enclosure seems to be a better trade off !

btw i'm asking mysel what could be an OB synergy when talking about the mid-bass firering in a port of a horn with no box behind these drivers ! in theory, the length between the front port and the bottom of the driver will be longer than the same driver raw... so having a wider usuable frequencies ! or the front bafle with its little firering port become a load and most of the rear driver becomes the emissiv surface ?

Although I'm not aware of the off axis behavior of the OBs, but it's an other subject and off topic.
 
Last edited:
Lewinski

Thanks. I guess I answered that question too quickly.

Q does play an important role...and each driver has nominal preferred parameters....and I consider those.

For bass, our fellow forum member is correct, it's often cleaner/easier to stick with sealed and avoid the complexity of porting. It's also academic when you're running 800 watt amp with a huge 18 inch driver in a domestic setting. Any gain I would have made with porting, is easily compensated by the driver efficiency and big amp and EQ.

(In contrast, my own bass system is Altec 420A 15 inch in 250 liter ported cab. The porting was critical in this instance to maximise bass depth...as I was running passive xover and lower powered valve amp.)

Cheers.
 

Attachments

  • finished pair no grills 2.jpg
    finished pair no grills 2.jpg
    95.9 KB · Views: 907
The 10 inch is a wee sweetie too!
Cheers
D.
PS I only ever use the one Qtc, 0.707.

Nice!
Dual Beyma 10G40 in sealed boxes is what I'm going with for 80 to 400Hz :D
Actively amped by Hypex UcD400, digital xo, time aligned, linearized, etc.

I'm also designing for Qtc=0.707, but I guess I struggle with the small box this driver needs to be at 0.707. Vb=9.1 liters is small.

I seem to be on a good path...I hope! :)
 
10.7 litres plus 7.3 litres for dampening and driver volume

Hi Lewinsky,

Yes he Beyma 10G40 is a really nice driver, I suggest a total cabinet volume of 18 litres will give you a 0.707 Q.

I get the simulated Vb to be 10.7 litres.
Add 2.7 litres to allow for the air volume which the driver will displace when front mounted on an 18mm thick baffle.

Then add approx 5 litres for internal cabinet wall dampening ie I usually line the inside of the cabinet with 9mm of bitumen or self adhesive underfloor insulation material.

If you are using twin Beyma's per side I suggest two separate 18 litre cabinets, not one shared 36 litre cabinet.
A pair of these will be fantastic covering the 80Hz to 1,000Hz.

Even without a sub, they will have superb bass, -3dB @ approx 35Hz and still good output at 25Hz.
This assumes you want to be able to reach a MAX or 105dB continuous SPL with 110dB peaks @ 1 meter. Your room / ears / neighbours may vary!
Adding a good sealed box sub will obviously give even more bass, but more importantly it will free up the 10G40 drivers from all the heavy lifting and extended cone travel required to reproduce bass.....

In otherwords the biggest improvement when you add a sub is better midrange!!

Hope this helps and all the best.
Derek.
 
Last edited:
Take a good 15 inch Pro driver ( Beyma , Precision Devices, 18 Sound etc) with a good 4 or 5 inch single layer voice coil, look for a low Qts and Qts, good Xmax (8mm to 10mm will do) and a powerful motor ( Bl 20 to 25) and you are done....The rest is just a big amp and good Eq.

Exact box volume is almost irrelevant, well more a by product of the Qtc you want.

Re low Qts and Mms / Bl.
My preference is always for a linen surround, low Qts, low Mms / high Bl driver.

With open baffle, it is not the Qts that defines the low frequency limit, it is Fs!

As long as the driver has a good 4 inch single layer voice coil (3 inch is enough on a 10 inch, but it will have a lower max SPL) one can Eq and power it all the way down to Fs....Qts is irrelevant....Most guys dont understand this but its true!

I agree with Overkill Audio that you can shape the response (including roll-off) to however you like it with EQ (rather than box volume), which is what Qtc describes. It may not be the most efficient way of doing it but as long as your driver can take it (without adding distortion, thermal compression or bottoming out) and your amp can handle it, I'd say it is a better way of doing it since it allows you to use drivers with strong motors (and therefore better control) in sealed enclosures with all the benefits hey have inherently over resonant systems. Also your simulated Qtc describes an idealized anechoic response without regard to your room, so the unequalized response you will hear in your LP will represent a totally different Q than what you were designing for. Better then to equalize that response, in the relevant room, into the response you desire (taking your taste into account, whether a 0.5, 0.577 or 0.707 type of roll-off etc).

I don't agree with him though, given the same context of EQ and powerful amps, that Fs defines the LF limit in OB, instead it is Xmax. At the LF limit, the woofer is operating in a pistonic fashion and the only defining factor is how much air the woofer can move (SPL), i.e. Xmax * Sd. This is well illustrated by the Pistonic Excurison Calculator, where Fs is not even a factor in solving for SPL at a given frequency (in this case, the desired LF limit):

Piston Excursion calculator

Sure, with a higher Fs you have a steeper roll-off below to fight but again as long as it is within your Xmax...

Let me illustrate this with two WBCD simulations of two 18" woofers in open air (Vb=99,999 litres), one pro woofer (BMS 18N862) and one dipole woofer (AE Dipole18). Xmax is 18 and 12 mm while Fs is 28 and 23 Hz, respectively. Each woofer's response has been EQ'd (quick-n-dirty using the filter section available in the spreadsheet) to be similar and in both cases to have 12mm excursion between 10-30Hz.


Frequency Response - AE Dipole18: HP BW1 38Hz + two peak filters 5-15Hz for gentle lift up to Xmax

An externally hosted image should be here but it was not working when we last tested it.



Frequency Response - BMS 18N862: LWT F0=28Hz, Q0=0.30; Fp=38Hz, Qp=0.81

An externally hosted image should be here but it was not working when we last tested it.




Excursion - AE Dipole18: Xmax=12mm

An externally hosted image should be here but it was not working when we last tested it.


Excursion - BMS 18N862: Dotted line at 12mm for comparison but Xmax=18mm

An externally hosted image should be here but it was not working when we last tested it.



Impulse Response - AE Dipole18

An externally hosted image should be here but it was not working when we last tested it.



Impulse Response - BMS 18N862

An externally hosted image should be here but it was not working when we last tested it.



Group Delay - AE Dipole18

An externally hosted image should be here but it was not working when we last tested it.




Group Delay - BMS 18N862

An externally hosted image should be here but it was not working when we last tested it.



Consumed Amplifier Power - AE Dipole18

An externally hosted image should be here but it was not working when we last tested it.



Consumed Amplifier Power - BMS 18N862

An externally hosted image should be here but it was not working when we last tested it.


As can be seen, the response at 12 mm excursion below about 30Hz is more or less identical with SPL 115dB, F3=34Hz, F6=27Hz and F10=22Hz. The major difference is the amount of power to get there: Dipole18=165W vs 18N862=450W.

However the 18N862 can take a bit more than a doubling of that power (100W) mechanically (thermally even more) before hitting Xmax of 19mm for another 3dB of SPL!

When designing a sealed enclosure that is going to be used with EQ, a good starting point is deciding the lower LF limit and the desired SPL at that frequency (presonally I think a good target is about 115dB @ 27-30Hz). The Piston Exursion Calculator will then give you a quick idea of the sort of driver(s) you need to achieve this (number, size, Xmax). Then it is a balancing act as usual between power, Fb and low-end efficiency using a simulator like WBCD for example.

To find extremes, simulate feeding the driver its Pe (or even double for 3dB peaks) in a large box but reduce Fb until excursion is reigned in at Xmax (or +15%) at your target LF limit. If the box is too small then (like it would be for some strong drivers like the 12P80Nd), increase Fb to a more logical/practical size and reduce power instead. Power greater than your amp? Then reduce power to the amp you're going to use while increasing Fb (low-end sensitivity) to see if you can hit your Xmax. Box too large for your living space? Then reduce to acceptable size and upgrade amp or accept lower SPL max output. Try different HP filters (Hz, type, order etc) to limit excursion below the desired LP limit while allowing higher SPL in the passband. LWT is a quick way of flatting a response for simulation purposes over the desired passband. Worth keeping an eye on the impulse response too because there is typically a volume (Vb) at which the IR cleans up signficantly going from small to larger.

In terms of Qtc, as a starting point before EQ, I normally find that my sealed enclosures range from about 0.4 (strong drivers) to 0.62.

When you have settled on a suitable volume and built the box, you will know from your simulation what sort of max power you can feed and the HP filter needed to protect your driver. Implement accordingly and then EQ at LP to flat (or desired house curve e.g. 3-6 dB greater below 100Hz :D ) to compensate for room, SBIR, baffle step etc.

As said, not the most efficient, or perhaps elegant, way of doing it as requires plenty of Vd and W... but gives great results!

EDIT: Sorry the rubbish images. Hate Tinypic :(
 
Last edited:
All good points....

Hi Inotin,

Very detailed answer, phew!
On the money in what you say and I agree with most of it, but just to clarify my thinking on a couple of point:

The reason I say Fs defines lower limit in OB designs and not Qts is that you cant drive a driver below Fs regardless of its Qts.

On the Xmax front:
I only ever use big Pro drivers with large high power voice coils (like Beyma), so say the driver has a 10mm Xmax, you can Eq the bass to stay within Xmax for say, 85% of the time (or 100% or whatever%) but as the driver can travel say plus / minus 20mm or 25mm before damage you can overload it for short periods of time (or parties!) .....

So Xmax limits are never brick wall limits only guide lines for best performance.

Fs, on the otherhand are brick wall limits....Go below Fs and the results are a VC train wreck!

Hope this helps and all the best
Derek.
 
The reason I say Fs defines lower limit in OB designs and not Qts is that you cant drive a driver below Fs regardless of its Qts.

On the Xmax front:
I only ever use big Pro drivers with large high power voice coils (like Beyma), so say the driver has a 10mm Xmax, you can Eq the bass to stay within Xmax for say, 85% of the time (or 100% or whatever%) but as the driver can travel say plus / minus 20mm or 25mm before damage you can overload it for short periods of time (or parties!) .....

So Xmax limits are never brick wall limits only guide lines for best performance.

Fs, on the otherhand are brick wall limits....Go below Fs and the results are a VC train wreck!

Hi Derek,

Sure, you can take the driver into the nonlinear territority that is between Xmax and Xmech for a bit of extra headroom (read 'fun'), especially down low where our hearing is not as sensitive, and I design/dimension like that too sometimes if Vd is too SPL limiting - although sometimes a driver becomes too noisy mechanically (vent chuffing etc) for it to be useful in practise.

Not with you on the Fs as a brick wall limit for the VC though. Could you elaborate why this should be the case, preferably providing some online reference for further reading? In my mind, this is just another audio myth... :smirk:
 
Last edited:
4 times driver travel for each octave lower in frequency.

Hi Derek,

Not with you on the Fs as a brick wall limit for the VC though. Could you elaborate why this should be the case, preferably providing some online reference for further reading? In my mind, this is just another audio myth... :smirk:

Hi Inotin,

I have tried to push 3 drivers below Fs in open baffle designs and its been an expensive learning curve!
The first time was with with audiophile drivers (twin Seas Excel) and they crumbled very quickly.
The last time was with the old spec ( attached) Beyma SM115K which is a very good value driver and can handle 500 watts with 7.5mm Xmax and Fs of 27 Hz.

I was getting great bass down to 30Hz but I wanted to get the last bottom octave, or at least flat to 20Hz with a -3dB of around 17Hz.
I had plenty of power in reserve (Spectron Musician 2 power amps) so I started to boost the Eq into the "unknown territory"....

All looked and sounded good measuring with the DEQX and Earthworks mic having Eq'd an 11dB boost at 20Hz with a Q of 1.4, this gave me flat to 22Hz with a -3 of around 19Hz, not bad so time for some music.....

I only got about 5 seconds in before the loud "crack" stopped the bass player on Maxwells Embrya track dead in his tracks!
Even although I was still using exactly the same master volume settings which gave "loud but safe" ( this was -14dB on the DEQX ) and it was the same test track, the law of driver travel killed the deal.

This is why I say Fs is a brick wall:
"For each octave you go down, you require 4 times the driver travel to maintain the same SPL"

So, for open baffle designs, when you are already down at Fs ( free air resonance, not box resonance, as that is different!) I dont know of any driver that can be pushed that hard if its already working at real world SPLs to reach its Fs.

I hope this makes sense and all the best
Derek.
 

Attachments

  • Beyma 15 inch Open Baffle SM-115 K.pdf
    121.7 KB · Views: 107
The last time was with the old spec Beyma SM115K which ...can handle 500 watts with 7.5mm Xmax and Fs of 27 Hz.

I wanted to get the last bottom octave... flat to 20Hz with a -3dB of...17Hz.
I had plenty of power in reserve (Spectron Musician 2 power amps)... having Eq'd an 11dB boost at 20Hz with a Q of 1.4 ....gave me flat to 22Hz with a -3 of around 19Hz... only got about 5 seconds... of the bass player on Maxwells Embrya track... The law of driver travel killed the deal

This is why I say Fs is a brick wall: "For each octave you go down, you require 4 times the driver travel to maintain the same SPL"

Hi Derek,

Hhmm… still not convinced. Drivers fail in one of two ways - mechanically (too much cone excursion) or thermally (shorted, melted or burnt voice coil). A majority of failures are mechanical, without having stressed the VC thermally, for the very reason you mention that excursions must quadruple for every halving of frequency to produce the same acoustic output. Running a woofer down to 20 Hz in free air like you did, without the benefit of an enclosure to add stiffness to the suspension, is very severe mechanically for the driver.

Here is a simulation of the SM115/K in free air with 11dB boost at 20Hz with Q=1.4 when fed 140W. F3=17Hz coincides with your own setup (the grey line is max SPL):

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.


As you can see, 140W is enough to exceed the driver’s mechanical excursion limit (X damage = +/-15mm). The mechanical SPL limit of the driver before damage would have been 102dB. Your Spectron amp could potentially output 700W into the SM115/K’s voice coil with Re=5.9 ohm, perhaps more at higher distortion levels. Maxwell’s track ‘Embrya’ features heavy basslines.

I would therefore argue that your SM115/K failed for mechanical rather than thermal
reasons. Again, the defining factor for mechanical failure is Xmech and has nothing to do with Fs, the point at which the driver’s moving parts resonate (nor do thermal failure for that matter). Like you said yourself, “the law of driver travel killed the deal”.
 
Maintaining high SPL's below Fs wont work....

Hi Derek,

I would therefore argue that your SM115/K failed for mechanical rather than thermal ]

I think that is in fact my point....?
Due to the driver travel law :
"For each octave you go down, you require 4 times the driver travel to maintain the same SPL"

One will induce massive mechanical failure if trying to maintain high SPL's below Fs.

You can call it anything you want and describe related events, the bottom line is wont work! Thats my point.

Cheers
D.
 
Today I:

Trimmed the mid bass baffles.

Glued in the T-nuts for the midd bass drivers.

Added the vertical braces to the mid bass.

Cut the plinths for the bass using one inch Birch Ply, mitred, screw holes frilled, counter-sunk drilled.

Pic of dry fit only, ignore odd angles.
 

Attachments

  • bass cab plinths cut.jpg
    bass cab plinths cut.jpg
    88.3 KB · Views: 894
  • mid bass baffle dry run.jpg
    mid bass baffle dry run.jpg
    103.5 KB · Views: 893
Hello atilsley,
What kind of bass are you expecting from that driver without EQ? Personally, I actually tend to be more fond of sealed bass but my issue with sealing big audio drivers has been a relatively high F3 and little to no low bass. Now I know that you plan to use EQ but distortion does begin to increase, sometimes quite substantially, when your EQ'ing a driver and box with a F3 of 65hz to somewhere in the 30s.
Thanks and great build btw,
Clashing
 
Hello atilsley,
What kind of bass are you expecting from that driver without EQ? Personally, I actually tend to be more fond of sealed bass but my issue with sealing big audio drivers has been a relatively high F3 and little to no low bass. Now I know that you plan to use EQ but distortion does begin to increase, sometimes quite substantially, when your EQ'ing a driver and box with a F3 of 65hz to somewhere in the 30s.
Thanks and great build btw,
Clashing

Thanks.

My goal with building a $10,000 killer system is to end up with no bass and distortion...!

I've used large drivers for previous assignments, eg EV 18 inch, plus a mix of Faital Pro drivers.

I understand that with these pro drivers, output comes out ahead over bass depth.

As such, I wouldn't be recommending these drivers for passive systems. For example, my own personal rig uses a vintage Altec 420A 15 inch driver in large 250 liter ported cab. With 96dB, I can get good output from 28-30Hz using valve amp. That suits me fine for my largish listening room.

With the Faital Pro 18, we'll get 122dB in sealed cab at 30Hz. That's pretty large. Rooms usually complain way before distortion comes into play. F3 is 61Hz.

We'll run a low pass filter around 80Hz....where the 12 inch driver will cross over.

Cheers.

Andrew
 
ELF

With the Faital Pro 18, we'll get 122dB in sealed cab at 30Hz. That's pretty large. Rooms usually complain way before distortion comes into play. F3 is 61Hz.

Andrew

Wow...122dB at 30 Hz....Nice!
How big was the box?

I like Elliot sound pages guy, he has a killer compact sub design
and this guy explains it in detail here Linkwitz Transform Subwoofer Equaliser

Extended low frequency subwoofer

Its a great way to get low distortion high SPL out of a small easy to build sealed box / big driver design.
All you need is a high power amplifier and a driver that can take the power boost when you Eq.

Alex.
 
Not bad.
I´ve been taken out all from my BMS 862´s in sealed and there´s quite sharp point on volume where i have to put 30hz high pass on to get more clear sound (or switch low boost off)

of course we are talkin about hearing loss levels
Single 862 can put out 128db @60hz outdoor groundplane

An externally hosted image should be here but it was not working when we last tested it.

in this room @30hz pair would give 130-135db
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.