Design check of a simple <100€ sub - Page 2 - diyAudio
Go Back   Home > Forums > Loudspeakers > Subwoofers

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 9th December 2011, 02:00 AM   #11
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
Some enlightening hours later...

Thanks again for the sim this was a great help, so some new questions :-)

1. Shouldn't L12 and L34 always be the same size to prevent overlapping of the enclosure?

2. In the loudspeaker-wizard, I try to calculate cone excursion but somehow when using Pmax instead of Eg, the cone excursion starts to change its shape rather than its amplitude. As Eg is dependant on Pmax and impedance I don't quite understand why it is doing this. Also Pmax to reach maximum cone excursion is really low... Is Pmax the acoustical output power???

cheers!
hurtz

P.s. I expected L23 to be much more critical for frequency response than the simulation told me, but I guess this is a result of the low frequencys. Also it seemed quite impossible to to raise the higher cutoff frequency to 100Hz without drastically loosing on low frequency response.
  Reply With Quote
Old 9th December 2011, 07:58 AM   #12
bjorno is offline bjorno  Sweden
diyAudio Member
 
Join Date: Mar 2005
Location: Jacobsmountain
Send a message via MSN to bjorno
Quote:
Originally Posted by hurtz View Post
1. Shouldn't L12 and L34 always be the same size to prevent overlapping of the enclosure?
No

Quote:
2. In the loudspeaker-wizard, I try to calculate cone excursion but somehow when using Pmax instead of Eg, the cone excursion starts to change its shape rather than its amplitude. As Eg is dependant on Pmax and impedance I don't quite understand why it is doing this. Also Pmax to reach maximum cone excursion is really low... Is Pmax the acoustical output power???
There in no reason to ever calculate Pmax as the T-TQWT is displacement limited only for low frequencies.

Pmax is dependent on Eg and the impedance at the specific frequency used when calculating, not the opposite!


Quote:
P.s. I expected L23 to be much more critical for frequency response than the simulation told me, but I guess this is a result of the low frequencys. Also it seemed quite impossible to to raise the higher cutoff frequency to 100Hz without drastically loosing on low frequency response.

Impossible? No, But you always trade Power-handling with BW.Look at the submitted example:

b
Attached Images
File Type: jpg Mivoc-AW3000_OD_T-TQWT.JPG (657.8 KB, 140 views)
  Reply With Quote
Old 9th December 2011, 05:14 PM   #13
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
Hmm I see, thanks...

I actually had the wrong design in mind.

I'd like to know what you think of the attached design, SPL is a bit higher at 100HZ, but not too much and the response is a little more flat. Eg is at 13.5 => 46W in 4Ohm

A question on cone excursion vs. simulation though, my research suggests that the response will be a bit flatter than the simulation due to resonance of the enclosure (and general not ideal physics). Now to keep the driver safe, will it be the same with the excursion? Or how should the safety margin be chosen?
I wanted to include a simple vu/power-meter to tell me when the maximum power exceeds driver capabilities. Also due to the low power demands of the T-TQWT I think I will rather use a LM3886 with 80VA toroidal than the symasym.

Thanks and cheers!
hurtz
Attached Files
File Type: txt mivoc_v1.txt (409 Bytes, 8 views)
  Reply With Quote
Old 13th December 2011, 02:00 AM   #14
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
Finally got time to make the screenshots,
I've changed some more and sacrificed some SPL for flatness of frequency response. Since I'm "only" using it for movies and a party a few times a year an average above 100db should be more than enaugh I think.

So let me know what you think of the design and if I missed anything vital!

cheers!
hurtz
Attached Images
File Type: jpeg acimp.jpeg (46.0 KB, 108 views)
File Type: jpeg delay.jpeg (49.3 KB, 95 views)
File Type: jpeg displ.jpeg (38.0 KB, 95 views)
File Type: jpeg elimp.jpeg (37.9 KB, 91 views)
File Type: jpeg impresp.jpeg (39.0 KB, 2 views)
File Type: jpeg impspec.jpeg (50.2 KB, 6 views)
File Type: jpeg main.jpeg (63.2 KB, 12 views)
File Type: jpeg phase.jpeg (55.6 KB, 11 views)
File Type: jpeg schem.jpeg (27.8 KB, 10 views)
Attached Files
File Type: txt mivoc_v2.txt (408 Bytes, 2 views)
  Reply With Quote
Old 13th December 2011, 11:06 PM   #15
bjorno is offline bjorno  Sweden
diyAudio Member
 
Join Date: Mar 2005
Location: Jacobsmountain
Send a message via MSN to bjorno
post#13:
Quote:
..my research suggests that the response will be a bit flatter..
See the two * here-below.( Advises to a T-TQWT design using HR TH modeling recommendations (observations) would IMO lend to an apples to oranges comparison)

HR cannot predict the superior resulting performance(IR,Gldy FR smoothness and so on....) of a stuffed T-TQWT/T-QWP when comparing with a TH, but this can easily partly be shown, by using quarter-wave programs like MJK:s.

Your first HR simulation in post#13 would work but with a 2 dB lower in-band SPL if compared with the one I posted.

I targeted a single folded T-TQWT with your driver and to use a smaller enclosure with an unusual(larger) compression ratio,that has to be stuffed with less amount of damping material than a normal T-TQWT would be:

Therefore the 'non peaking' of the lower FR corner as I always try to use an optimum amount of damping material but always fill ~ half of the enclosure counted from the largest CSA.



*You've probably read that in HR shown peaks doesn't matter but I have an entirely different opinion:

Based on measurement and listening to single folded TH:s + T-TQWT:s and T-QWP:s:
A single folded TH that isn't stuffed with damping materials will preserve the in HR shown peaks to a larger degree than if multi-folded* but would still sound IME very boxy/resonant and is not suitable for home HT or music unless an advanced parametric EQ + Signal delay units(often needed) are in use but may be more successful for outdoor/ large venues Live/PA reinforcement as the 'Low-Fidelity' artifacts would be more or less masked.


Quote:
..Now to keep the driver safe, will it be the same with the excursion? Or how should the safety margin be chosen?..
Your design show a maximum input of 43.5 W before exceeding x-max at ~ 28.5 Hz that if provided a suitable HPF is in use to protect otherwise evidently over-excursion that would occur at lower frequencies.

Then a maximum operating power level of -3dBW is a good choice to not exceed for SQ reasons, that is ~22W. 22W at 4 Ohm → 9.38V Rms=Peak_13.27 V could be used for an comparator input trigger voltage to indicate beginning of an overload or a minimum voltage acting on a soft limiting circuit that goes hard limiting or into complete mute mode if reaching ~18.65 V peak.

When using a single LM3886 and a typical 4 Ohm load: Choose a transformer with a secondary RMS Voltage <= 2x20.


Quote:
..I've changed some more and sacrificed some SPL for flatness of frequency response. Since I'm "only" using it for movies and a party a few times a year an average above 100db should be more than enaugh I think..
Agree !00dB/(2x Pi) or if the sub is placed near a room corner the corresponding SPL is more than 110 dB or max. ~161 dB internal SPL at the enclosure section within the S2-S4 areas, a quite easy load for a T-TQWT to cope with as this occurs only at a very low frequency (~19 Hz).

I noticed a problem conserning the very high Sd/ port ratio, that in my example is around 3:1(~optimal if not larger ratio) but for your last simulation the ratio is ~9:1.

For sure this will lend to annoying port chuff's at higher SPL.

b
Attached Images
File Type: jpg Mivoc-AW3000-comp.jpg (401.5 KB, 23 views)

Last edited by bjorno; 13th December 2011 at 11:13 PM.
  Reply With Quote
Old 14th December 2011, 07:31 AM   #16
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
Hmm I see, thanks for the explanations!

I have the distinct feeling that your design in post 2# is as close as it gets to the optimum for this driver...
The peak at 85Hz seems to be dealt with by using a filter, what program did you use to simulate the LR-filtering on the HR response?

Also I did think about the multi-folding, the form factor seems a bit more convenient. I also looked at some tutorials for this, it doesn't seem overly complicated once the basic design parameters are known.

Speaking of the basic design parameters, can you just do them by hand, lets say 40cm width for easy mounting of the driver and then just calculate the rest? Or does it require more math

cheers!
hurtz
  Reply With Quote
Old 14th December 2011, 12:41 PM   #17
bjorno is offline bjorno  Sweden
diyAudio Member
 
Join Date: Mar 2005
Location: Jacobsmountain
Send a message via MSN to bjorno
Quote:
Originally Posted by hurtz View Post
..I have the distinct feeling that your design in post 2# is as close as it gets to the optimum for this driver...The peak at 85Hz seems to be dealt with by using a filter, what program did you use to simulate the LR-filtering on the HR response?..
Thats the one I would follow too..You're right if the peak is small it will be swamped by the LPF.I always try to use tools of the trade to show manipulated HR exported data: Here the easy to use freeware HOLMimpulse program.

Quote:
..Also I did think about the multi-folding, the form factor seems a bit more convenient. I also looked at some tutorials for this, it doesn't seem overly complicated once the basic design parameters are known...
It's up to your intension but sometimes a tedious work that I assign to the DIY'r to do and the outcome of FR from using multi-folds can be quite different as a series of equivalent LPF is introduced from the bends causing unpredictable in-band/off-band response. A POC is the only way to know..

Quote:
..Speaking of the basic design parameters, can you just do them by hand, lets say 40cm width for easy mounting of the driver and then just calculate the rest? Or does it require more math..
Of course, I usually start with the driver frame width(+ a few mm:s) and calculate the required CSA that can accommodate the driver with a suitable rear clearance behind the driver magnet for the CSA= S2 in HR.

If to close to the Wood in use, I start to increase the Driver baffle thickness before alter other enclosure dimensions...

b
  Reply With Quote
Old 15th December 2011, 12:56 AM   #18
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
The S2 seems to narrowly fit.

Driver frame width is 31.5cm and has a depth of 12.4cm. So a total width of 35cm for the enclosure makes that [S2=489cm²] 489cm²/35cm=13.97cm.

When drawing it, the cone of the driver is "overlapping" the opening at S4, that doesn't seeem right. But since L34 is only 10cm..

Also I think I have to go with a multi-fold or smaler design since there is now way I can fit in the ~145*35*19cm (the 145cm especially...)

Sadly holmimpulse doesn't work in linux so I have to delay the filter part until I can grab hold of a windows machine for a few hours...

thanks & cheers!
hurtz
  Reply With Quote
Old 15th December 2011, 01:41 AM   #19
hurtz is offline hurtz  Germany
diyAudio Member
 
Join Date: Jun 2010
P.s.
Ok erase the drawing part.. made an error >.<
  Reply With Quote
Old 15th December 2011, 03:01 PM   #20
bjorno is offline bjorno  Sweden
diyAudio Member
 
Join Date: Mar 2005
Location: Jacobsmountain
Send a message via MSN to bjorno
Quote:
Originally Posted by hurtz View Post
P.s.
Ok erase the drawing part.. made an error >.<
If there is a problem with driver clearance, never add more than a few mm to the driver frame width when calculating the resulting section depth from the suggested S2 CSA.

Step by step:

b
Attached Images
File Type: gif OD-T-TQWT_fold.GIF (48.6 KB, 74 views)
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
10" Subwoofer < $100 5toes Subwoofers 8 25th April 2011 11:12 AM
Lafayette LT-100 and 6AQ8 tubes rickl Tubes / Valves 5 17th April 2008 07:54 PM
subwoofer <$100? gychang Subwoofers 16 3rd February 2007 02:49 AM
Sub (plate) amp <100€ in Europe Cordraconis Subwoofers 5 29th July 2005 08:19 PM


New To Site? Need Help?

All times are GMT. The time now is 08:47 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2