Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

The birth of a 15 inch woofer
The birth of a 15 inch woofer
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd August 2019, 06:36 PM   #21
Lojzek is offline Lojzek  Croatia
diyAudio Member
 
Lojzek's Avatar
 
Join Date: Feb 2012
Location: Croatia
Make them sturdy, provide evidence of the performance parameters, priced within reason and I bet you will sell a dozen, possibly more.
  Reply With Quote
Old 23rd August 2019, 11:05 PM   #22
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
So the frame I went for is not the kind you would think of for a "5000W car audio competition subwoofer", but rather a transparent one that is far more sturdy than we would ever need, and in addition has some decent openings for air to escape from under the spider. It also has the necessary depth I need from surround landing to spider landing for a 15 inch cone with 3 inch voice coil (which naturally has a deeper profile than what it would with a 4 inch coil), good height from spider landing to motor mount to allow some serious excursion before something is damaged, and last but not least, a spider landing of respectable 204mm to allow for a more linear suspension.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

This frame comes with two terminals positioned on each side, so balanced lead out wires is possible. By default it does not have any gasket or trim ring, so to make it look a bit better I plan to finish it off with a ring that looks something like this:

Click the image to open in full size.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 24th August 2019, 07:26 AM   #23
Zvu is offline Zvu  Serbia
diyAudio Member
 
Zvu's Avatar
 
Join Date: Jan 2009
Location: Belgrade
Quote:
Originally Posted by Snickers-is View Post
I am not familiar with that Beyma driver, but in what way would you regard it to have more desirable parameters?

Nothing special really. 15K200 has a little higher Qms, larger voice coil and is a bit more sensitive W/m.
__________________
Tesla

Last edited by Zvu; 24th August 2019 at 07:28 AM.
  Reply With Quote
Old 24th August 2019, 09:45 AM   #24
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
Quote:
Originally Posted by Zvu View Post
Nothing special really. 15K200 has a little higher Qms, larger voice coil and is a bit more sensitive W/m.
I checked the datasheet now. It has a stiffer suspension and a fair bit more force factor than my target parameters. It would roll off a bit early in the same box.

The coil is larger, but it has only one. In terms of cone coupling it is definitely larger, but in terms of heat it would be more relevant to say it is 2 inches smaller.

The qms is one of the hardest parameters to predict. But with relatively low loss suspension, non conductive bobbin and low air resistance it will for sure be high.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 24th August 2019, 10:15 AM   #25
Zvu is offline Zvu  Serbia
diyAudio Member
 
Zvu's Avatar
 
Join Date: Jan 2009
Location: Belgrade
Of course. Double coil, that JBL uses for quite some time, will make the heat problem less of a problem. On the other hand, if one warms up a coil in a woofer of that sensitivity and size enough to change ts parameters, he really needs a PA loudspeakers - and more than two, in my opinion.

Either way, you are doing things as you see fit and i am subscribed to this thread. I'd really like to see this project successfully finished.
__________________
Tesla
  Reply With Quote
Old 25th August 2019, 03:36 PM   #26
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
Quote:
Originally Posted by Zvu View Post
Of course. Double coil, that JBL uses for quite some time, will make the heat problem less of a problem. On the other hand, if one warms up a coil in a woofer of that sensitivity and size enough to change ts parameters, he really needs a PA loudspeakers - and more than two, in my opinion.
I think your opinion is spot on. I did not choose double coil/double gap to cope with heat, but to make it symmetrical so that I could focus more on DC components (Not the kind of DC-components associated with traditional motors, but force factor not being the the same independent of polarity).

There are some rare occations where the coils is very short, and air is not moving very much through the gap. We often focus on the coil diameter, but doubling the diameter really does not do much more to cope with heating than doubling the length of the coil. Anyway, what I meant to say is that the sensitivity does not necessarily solve this, even with a 4 inch coil, if the coil is very short.

The surface area of this coil is around 160sqcm. The Beyma 15k200 has 66sqcm. The crazy BMS 18N862 actually has 160sqcm voice coil surface, and it has 267g Mms to drag around. That one probably has some degree of forced air cooling, but it is rated at 1500W AES noise. My driver will have lower class wire to get less gap between the windings (thinner insulation), and therefore not be able to handle the same amount of heat. But using 1000W AES to simulate heating properties is probably not that far from the truth.


Quote:
Originally Posted by Zvu View Post
Either way, you are doing things as you see fit and i am subscribed to this thread. I'd really like to see this project successfully finished.
Thanks Zvu! I will post much more tech stuff here, but opinions and discussion are welcome! That kind of discussion is a significant source of knowledge.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 25th August 2019, 03:54 PM   #27
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
There are typically 4 different factors that induces distortion in a motor. All of them does it by changing the relative force factor. They are as follows:

BL(x) - Force factor as function of coil position
BL(i) - Force factor as function of coil current, typically flux modulation
Le(x) - Inductance as function of coil position
Re(T) - DC resistance as function of voice coil temperature

Re(T) is not really a direct source of distortion as it is the change in Re that causes some change in the parameters, and increased Re does not by itself cause distortion if it remains constant. It does, however, impact other parameters that causes harmonic distortion.

BL(x), also known as the large signal force factor, is often described as the main source of harmonic distortion at low frequencies. Strangely, it is not often described as a source of intermodulation distortion, but that is a far bigger problem in real life. Industry standards commonly used to describe X-max are -30% for subwoofers, and -18% for midwoofers. In the pro audio industry they often accept even more loss in force factor than that.

This is what the BL(x) (simulated) looks like for this woofer:

Click the image to open in full size.

And this is a simulation showing the response as a function of change in force factor from 0 to 8mm:

Click the image to open in full size.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 25th August 2019, 07:19 PM   #28
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
Jumping resonance treshold and suspension:

A less known non linearity effect in speaker drivers is is jumping resonance as a result of suspension stiffness increasing with amplitude.

The effect is described in this Wikipedia article:
Nonlinear resonance - Wikipedia

A good illustration of this is the foldover effect. It looks like this:
Click the image to open in full size.

The curve shows the systems tendency to be resonant. There are several things happening here. First off all, we can see that the width of the treshold decreases with amplitude. This means the system has an increasing Q with amplitude. If we make two identical drivers, except for the suspension being less compliant on one than the other, the one with the highest Fs will also have the highest Q. This is what we clearly see in this diagram.

Secondary, if wee look at the top of the curve, we can see that it ends around +500Hz. So at 300Hz, we can get a skipping effect. As the amplitude goes above ~7 the unit leaves the resonant part. This means the efficiency of the system will be slightly reduced compared to below the treshold. At an amplitude around 30, the system has an efficiency that is lower than at an amplitude of 40. The system amplitude might actually be both 30 and 40 at the same input signal.

So imagine the input signal being dynamic, varying from 0 to 40, the efficiency will vary through the period, jumping in a non linear matter. This leads to higher frequency components being added to the response.

In a real life speaker driver, the resonance shift will not be several octaves. The amount of shift will depend on how progressive the suspension is. An other interesting fact is where in the drivers working range the resonance can be found. Moving Fs in the low end of the working range will both increase amplitude around Fs, but also reduces the frequency and therefore the audibility of the harmonic components. One idea could be to reduce Fs to an octave below the low end (Fb) of the system, but it could also mean a system with reduced stability that copes less well with DC components. Moving Fs up an octave or two will increase distortion audibility, and it will add resonant behavior around 50-120Hz, leading to a less desirable impulse response. In addition to that, resonant behavior also affects higher frequencies and adds intermodulation distortion.

There are basically two different sources of capacitive effect in the simplified speaker model. One is the suspension, the other one is the internal box volume. We can not do much with the box, other than to decide its size. I could, off course set the box volume to 10 000 liters, but it wold not be a practical solution. Many people would probably say the selected 150 liters is too big as well, but that is the chosen compromise.

Then we should turn our focus to the suspension itself, and here we have some choices to make, two of them being especially important for the effect of jumping resonances:

1: The compliance of the suspension decides the drivers Fs. It also decides how well the driver controls DC components in the driver. By selecting a motor geometry that removes DC components we are left with a more free choice of Fs. We could lower it to an extremely low number, but it could create some issues with the midbass, as well as making the driver vulnerable to transients. A well balanced response is what we get when Fs is placed around 20-25Hz. However, the combined resonance is 48Hz, so the box dominates the capacitive components. So at low excursion, the suspension contributes very little to the total resonance.

2: The progressiveness of the suspension is another factor. If the suspension is very progressive, it will quickly become the dominant capacitive factor. Achieving a constant compliance over a large excursion range is therefore a crucial goal. The easiest way to achieve this is to use suspension with fairly long excursion capacity. The spider has a large outer diameter of 8 inches, and, off course, a 3 inch inner diameter. The distance from the bobbin to the edge of the suspension area on the spider is 60mm. The surround is not that wide. It has around 35mm between inner and outer fixed points, which is not bad either, but far from the 60mm of the spider. If the surround is stiffer than the spider, it will become the dominant capacitive component of the suspension. But since it is not as wide as the spider, it will also typically be more progressive. Therefore, the spider should be the stiffer suspension component, while the surround should be the softer one.

There are also many different profiles to choose from on surrounds. The so called "Double half roll" type is one of the more suitable variant for drivers that are intended to deliver both bass and midrange performance. This is the one selected so far, but I am awaiting some adjustments on this. JBL has also made some pointers on this issue. You can read about them in the right column here:

Click the image to open in full size.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 25th August 2019, 08:15 PM   #29
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
There are some more details to cover when it comes to the voice coil heating.

In typical musical performances the energy content for each frequency tends to peak between 80 and 200Hz. Music with higher energy content tends to peak at lower frequency than music with lower energy content. Independent of music type the energy does not seem to be very much different above 200Hz. At 400Hz the average level is typically 6dB lower. That means half the voltage, or 1/4 the power.

So if we feed the driver at 56,3V, the maximum power will be 100W. At 200W we have 82W, which is 82% of the energy. At 400Hz we will have 96W, but 6dB down that is just 24% of the peak energy. At 80Hz the energy is down to about 28W due to the rising impedanc towards lower frequencies. However, in bass heavy music, it can be 6dB higher than what it is at 200Hz, which means we still have 112W. At lower frequencies, the peak power will be lower as a combination of the impedance and the musical energy content, even though the level in the music could be the same down to sometimes around 40Hz.

The maximum RMS voltage needed to achieve x-max at X(f)peak, f>Fb (which is at 39Hz) is 116VRMS. This translates to 425W at 1kHz. At So this means our maximum practical power, that translates to heat, is around 1/8 of that, which means 53W. For bass heavy music, we can expect around that at 80Hz, or for more average music, we can expect around 45W maximum, occurring around 200Hz.

So how does this translate to what amplifier is needed?

Well, if we connect the two coils in parallell, the voltage should be 58VRMS. That translates to 420W in 8 ohms. A Hypex Ncore 252MP as bridged will do this with ease, with a dynamic headroom in excess of 200W).

As this typically translates to well in excess of 120dB (127,4dB using standard calculation method that could be a bit optimistic in some rooms) it should be sufficient for most applications.
__________________
Live sounds better than HiFi.
  Reply With Quote
Old 26th August 2019, 04:53 AM   #30
Snickers-is is offline Snickers-is  Norway
diyAudio Member
 
Join Date: Nov 2004
Location: Rakkestad
Send a message via MSN to Snickers-is
So I made a quick rendering, just to see what it will look like with a conical trim ring. I think that could be beneficial for the high end of the frequency range too...:

Click the image to open in full size.

...
__________________
Live sounds better than HiFi.

Last edited by Snickers-is; 26th August 2019 at 05:04 AM.
  Reply With Quote

Reply


The birth of a 15 inch wooferHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Best 8 inch woofer? acstcwrfrmn Multi-Way 95 3rd September 2019 11:43 AM
8 inch woofer for WAW Cerniu Full Range 31 25th June 2018 08:17 AM
Audax pr170mo Woofer selection... which 15 inch woofer 98db eff... youknowyou Multi-Way 15 20th June 2017 05:37 AM
Matching 12 inch woofer with 4 inch mid awdtalon21 Multi-Way 7 7th June 2017 04:43 PM
WANTED: 6-7 inch pro mid/woofer or some small box woofer Jimmy154 Swap Meet 10 18th August 2012 12:01 PM


New To Site? Need Help?

All times are GMT. The time now is 04:58 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki