Enclosure for measured FI SP4 18 D2 - AVS Forum | Home Theater Discussions And Reviews

Forum Jump: 
 
Thread Tools
post #1 of 13 Old 06-11-2015, 01:13 PM - Thread Starter
Newbie
 
zak1976's Avatar
 
Join Date: Apr 2012
Posts: 7
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
Enclosure for measured FI SP4 18 D2

Hi togehter,

My name is Olli and I am living in germany in the near of Hamburg.
In my 5.1 cinema I need a recommandation for the subwoofer part ...
My current setup is 5.1, the front is Visaton Experience (custom made with GF200 instead of W200), the center is also a version of the Experience with 2x GF200 in a horizontal version of the visaton Experience (please look at "world wide web dot mrwoofa dot de for the vox center", this is the version i am useing).
The Rears are a small version of the front without the bass section.

All loudspeakers are powered by seperate (PA)amps with 300w RMS at 4ohms.

The Sub part is covered by a FI SP4 18D2, fully loaded (this is a beast of a sub), the actuall modell, powered by a T.Amp Proline 3000, a real (iron transformator) monster with more than 3kw power in 4ohm bridge mode.

The enclosure is actuall 375l net with an 25cm diam. port tuned to 23hz. This "relative high" tuning is because my room gains more than 10db below 25hz.

I am looking for a better sub enclosure, a tapped horn.

Below, please see my first attemp and the measured TSP (don't know why LE seems to be too low, measured with LIMP and external amp, soundcard is M-Audio 2496 in Win XP Pro)

Parameter, Value, Dim
Fs, 32.09, Hz
Re, 2.30, ohms[dc]
Qt, 0.51, -
Qes, 0.55, -
Qms, 6.37, -
Mms, 541.24, grams
Rms, 18.823193, kg/s
Cms, 0.000045, m/N
Vas, 90.80, liters
Sd, 1194.59, cm^2
Bl, 23.152895, Tm
ETA, 0.62, %
Lp(2.83V/1m), 95.43, dB
, ,
Le, 1201.90, uH
L2, 6001.31, uH
R2, 55.97, ohms
L3, 0, uH
R3, 0, ohms

Here is my first result with Hornresp:

Please help me designing a TH for this chassis, the folding of the TH is also a problem for me because of the bends and the displacement of the driver, don't know how to implement this in Google Sketchup.

Greetings from germany

Olli
Attached Thumbnails
Click image for larger version

Name:	SP4_1th.jpg
Views:	180
Size:	52.6 KB
ID:	770289   Click image for larger version

Name:	sp4_1.jpg
Views:	184
Size:	39.3 KB
ID:	770297  

Last edited by zak1976; 06-11-2015 at 01:23 PM.
zak1976 is offline  
Sponsored Links
Advertisement
 
post #2 of 13 Old 06-11-2015, 01:51 PM
AVS Addicted Member
 
Stereodude's Avatar
 
Join Date: Jan 2002
Location: Detroit Metro Area
Posts: 13,664
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 2950 Post(s)
Liked: 2112
Paging @diy speaker guy
Stereodude is offline  
post #3 of 13 Old 06-11-2015, 01:53 PM - Thread Starter
Newbie
 
zak1976's Avatar
 
Join Date: Apr 2012
Posts: 7
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
Edit: Vtc is wrong, should be 6713, but this doesn't change really much...
zak1976 is offline  
post #4 of 13 Old 06-11-2015, 03:01 PM
AVS Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,077
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked: 270
Quote:
Originally Posted by Stereodude View Post
For which part?

I'm not sure what frequencies Le and Le2 are measured at but regardless, I think this driver is going to have significant inductance issues that will considerably skew the frequency response. The SP4 measured by data-bass (not the same impedance and likely not the same configuration as OP's) measured 1.3 Re and 2.61 Le which is remarkably high, high enough to cause significant issues if it's not accounted for.

If OP wants to build with 2 parallel walls and 2 expanding walls for the flare like most bass horns made from sheet goods CON should be switched to PAR for all segments.

Compression ratio is maybe a bit high but shouldn't be much of a problem except maybe at really high spl.

Also, if there is 10 db of room gain boost below 25 hz, that should probably be incorporated into the design, otherwise it will sound very boomy if not eq'ed.

If this is OP's first attempt at designing, folding and building a horn I'd advise against it. Read a lot about horns, folding and inductance and then start with a lower inductance driver for the first time out of the gate, a driver that will actually measure the same as it sims if the fold is done right. This is not an impossible challenge but it is a big challenge for a first timer.
diy speaker guy is online now  
post #5 of 13 Old 06-11-2015, 03:42 PM
AVS Addicted Member
 
Stereodude's Avatar
 
Join Date: Jan 2002
Location: Detroit Metro Area
Posts: 13,664
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 2950 Post(s)
Liked: 2112
Quote:
Originally Posted by diy speaker guy View Post
For which part?
The warning about the inductance of the driver and how it impacts the horn's response (which you just gave).
Stereodude is offline  
post #6 of 13 Old 06-13-2015, 03:34 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 20,312
Mentioned: 379 Post(s)
Tagged: 1 Thread(s)
Quoted: 2293 Post(s)
Liked: 2487
hi olli,


by correlating the measured results of several subs to the t/s parameters, we have found that drivers with "high inductance" do not perform as predicted using the simple models in winisd and hornresp. for these drivers, the inductance serves to effectively reduce the motor strength and this can change the frequency response, in some cases rather dramatically, particularly with horns. the driver that you have likely falls into this category of driver, so the modeled hornresp response is unlikely to be what you will get if you build the horn as designed.


i'm not sure what adjustment to suggest for this particular driver.

Listen. It's All Good.
LTD02 is offline  
post #7 of 13 Old 06-18-2015, 05:56 AM - Thread Starter
Newbie
 
zak1976's Avatar
 
Join Date: Apr 2012
Posts: 7
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
I've read the "Inductance" thread, not an advantage for my design goal ...
I will try another version with adjusted parameters for the driver, I will try DSG's method by doubling RE to weaken the motor force.
Can I effectively use an equilizer to reduce the peaks caused by the high Inductance ?

Do you guys have a design suggestion with maximizes output down to 20-23 Hz for this driver?
zak1976 is offline  
post #8 of 13 Old 06-18-2015, 08:38 AM
AVS Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,077
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked: 270
Quote:
Originally Posted by zak1976 View Post
I've read the "Inductance" thread, not an advantage for my design goal ...
I will try another version with adjusted parameters for the driver, I will try DSG's method by doubling RE to weaken the motor force.
Can I effectively use an equilizer to reduce the peaks caused by the high Inductance ?

Do you guys have a design suggestion with maximizes output down to 20-23 Hz for this driver?
If you double Re you have to add more power to regain the losses incurred by the extra Le. Even so, that gives you a more accurate representation of frequency response but some of the other graphs will be off.

Yes, you can use eq to even things out, but it would have to be parametric eq or a graphic eq with many many bands to be precise enough to help.

BUT it makes a lot more sense to just design for the frequency response curve you want instead of building something so big and expensive and flawed.

Tapped horns are not very forgiving for these types of drivers, ported boxes aren't either. Sealed and front loaded horn fare much better when trying to get a decent frequency response curve out of this type of high inductance driver.

You could see if the driver sims well in LTD's Submaximus front loaded horn, that's tuned a bit lower than you are asking for but it's a lot easier than designing and folding a new horn.
diy speaker guy is online now  
post #9 of 13 Old 06-18-2015, 12:02 PM
 
Mfusick's Avatar
 
Join Date: Aug 2002
Location: Western MA
Posts: 32,084
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 608 Post(s)
Liked: 2665
Quote:
Originally Posted by Stereodude View Post
or @lilmike

(although he's probably too busy and internet-less with his new house being built)
Mfusick is offline  
post #10 of 13 Old 06-18-2015, 01:12 PM - Thread Starter
Newbie
 
zak1976's Avatar
 
Join Date: Apr 2012
Posts: 7
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
Thanks, the Submaximus seems not to fit, looks bad in Hornresp.
I know a total new design is not easy, but I would like to go this way if there is horn design that fit to this driver.
I think with your help, it is possible to make a good folding for the TH, for the build, I've experience in woodworking...

Here is new suggestion:





Not much EQ needed, what's your opinion?
Attached Thumbnails
Click image for larger version

Name:	1.jpg
Views:	135
Size:	41.5 KB
ID:	783138   Click image for larger version

Name:	2.jpg
Views:	123
Size:	54.5 KB
ID:	783146  

Last edited by zak1976; 06-18-2015 at 01:19 PM.
zak1976 is offline  
post #11 of 13 Old 06-18-2015, 02:56 PM
AVS Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,077
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked: 270
Quote:
Originally Posted by zak1976 View Post
Thanks, the Submaximus seems not to fit, looks bad in Hornresp.
I know a total new design is not easy, but I would like to go this way if there is horn design that fit to this driver.
I think with your help, it is possible to make a good folding for the TH, for the build, I've experience in woodworking...

Here is new suggestion:

Not much EQ needed, what's your opinion?
I don't have time to write a full opinion here, so here's just a few points.

1. Your measured t/s parameters in post 1 say Re is 2.3 ohms but in your most recent post the Hornresp inputs show Re is 4.6 ohms but all other t/s parameters in post 1 and the Hornresp inputs are the same. What happened here? Did you measure t/s parameters of a single coil of your dual coil driver and then just double Re in the sim? Or did you double Re in the sim to account for inductance? I'm guessing the latter. But if so, that means your driver's individual coils are 1.15 ohms each, which is pretty low for a 2 ohm nominally rated driver.

2. I don't like the frequency response curve you showed in your last post. Add a bit more Re to that to simulate the effect of power compression and the whole thing starts to fall apart really fast. Also that massive hole in response at 100 hz isn't ideal, especially with the big spike at 150 hz.

3. To be honest I don't really trust you measured t/s specs. They are just way too different than the specs measured by data-bass.com. I realize they are different drivers technically, different impedance and FI offers all kinds of different options, but they share the same name (SP4) and the same platform. And your measured specs are WAY off from data-bass's measured specs AND from the published specs on the FI website. Fs is way too high and VAS is way too low. Is this driver broken in? Are you confident in your setup and ability to measure drivers?

So I did a quick sim, Submaximus with SP4 as measured by data-bass with (red line) and without (blue line) my inductance correction method. Ignore the spl levels, this was done at a random voltage and I did not increase power to make up for the added Re losses in the red sim because it's easier to see if the red and blue lines are not right on top of each other.



As you might guess, I'd trust the inductance corrected red line sim, not the regular blue sim. And it looks pretty fantastic to me.

Just for fun I also ran the sim with your measured t/s parameters even though I really don't think they are accurate. Again, red is corrected, blue is not.



Not so fabulous, but I still like it a lot better than your most recent sim, which is horribly overdamped. And again, I don't think the t/s parameters used in this sim are accurate. But if these measured parameters are accurate I personally wouldn't put this driver in Submaximus. But to be honest I think the parameters data-bass measured are probably more accurate than the ones you measured. So I can't really provide any recommendation.

I don't have time to help you fold a horn, so if you go with a new design you are on your own.

Last edited by diy speaker guy; 06-18-2015 at 03:18 PM.
diy speaker guy is online now  
post #12 of 13 Old 06-21-2015, 01:53 PM - Thread Starter
Newbie
 
zak1976's Avatar
 
Join Date: Apr 2012
Posts: 7
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
thanks for your opinion, so what to do with this driver...

regarding your points:

1. yes, as you mentioned, RE has another value due to the correction for the "high inductance driver".
Are there other parameters to be adjusted to meet the correction you mentioned ?
Regarding RE, yes you are right, RE is pretty low for a nominal 2ohm driver, it was measured with a calibrated multimeter (Fluke) with coils in series (RE=2,3 ohm = 1,15 ohm per coil)

2. The goal was to find a design wich is within +- 6db from 80hz down to 22hz, ok, not reached because of the part around 80hz...
The peak could be equillized (minidsp is already working) easyly and the lowpass at 80hz takes down the peak at 150hz a lot, but this is only simulation ...
What do you mean by " horribly overdamped" ? Where did you see that?

3. Maybe the measured parameters are way of but i dont think so because they are consistent and the driver i use is not optimized for HT.
It uses 5 spiders (Riccis version uses only 2), that resulted in the very low cms value and thats why fs is higher and vas is lower (than Ricci s version). The rest of the parameters fit (with tolerance) to the parameters Ricci measured. The Fi Parameters, i dont trust, never said which version was measured, and there are a lot of options, even a lot that are not availible anymore and the parameters stay the same...

The driver was not extra broken in, it was used 10 month as HT sub and then measured.
But i will measure the sub again after an extra break in period the next week ( 12 hours by 10hz), to see if there are that much difference after that.
As measurement setup i use limp with arta measurement box, an old harman kardon amp and m-audio 2496 as soundcard.

Don't get me wrong, you should not do the folding for me, that is my part, i would be thankful for the assistance to make my design "ready to build", eliminate errors an so on..

Last edited by zak1976; 06-21-2015 at 02:04 PM.
zak1976 is offline  
post #13 of 13 Old 06-21-2015, 04:18 PM
AVS Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,077
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked: 270
Quote:
Originally Posted by zak1976 View Post
thanks for your opinion, so what to do with this driver...

regarding your points:

1. yes, as you mentioned, RE has another value due to the correction for the "high inductance driver".
Are there other parameters to be adjusted to meet the correction you mentioned ?
Regarding RE, yes you are right, RE is pretty low for a nominal 2ohm driver, it was measured with a calibrated multimeter (Fluke) with coils in series (RE=2,3 ohm = 1,15 ohm per coil)
Ok, you added double Re for the compensation. Got it.

Ok, your measured Re isn't far off from Fi's spec, so that's good.

Quote:
2. The goal was to find a design wich is within +- 6db from 80hz down to 22hz, ok, not reached because of the part around 80hz...
The peak could be equillized (minidsp is already working) easyly and the lowpass at 80hz takes down the peak at 150hz a lot, but this is only simulation
Sure, the lowpass will take a bite out of the 150 hz peak, but it's still there, and with that peak and the big hole at 100 hz it won't blend well with the mains.

Quote:
What do you mean by " horribly overdamped" ? Where did you see that?
There's no peak at the low knee, that usually points to overdamped. (Even if there is a peak at the low knee it can still be overdamped in front loaded horns though.) Double up Re to simulate the effect of heavy power compression and it will be massively overdamped.

Quote:
3. Maybe the measured parameters are way of but i dont think so because they are consistent and the driver i use is not optimized for HT.
It uses 5 spiders (Riccis version uses only 2), that resulted in the very low cms value and thats why fs is higher and vas is lower (than Ricci s version). The rest of the parameters fit (with tolerance) to the parameters Ricci measured. The Fi Parameters, i dont trust, never said which version was measured, and there are a lot of options, even a lot that are not availible anymore and the parameters stay the same...

The driver was not extra broken in, it was used 10 month as HT sub and then measured.
But i will measure the sub again after an extra break in period the next week ( 12 hours by 10hz), to see if there are that much difference after that.
As measurement setup i use limp with arta measurement box, an old harman kardon amp and m-audio 2496 as soundcard.

Don't get me wrong, you should not do the folding for me, that is my part, i would be thankful for the assistance to make my design "ready to build", eliminate errors an so on..
Ok, if your driver has 2.5x more spiders than Ricci's and the measured driver on the Fi webpage then it makes sense that Fs and Vas would be different. If you are confident in your measured specs then use them.

This still isn't a driver I would use as a first time horn project, but if that's what you want to do, get a sim you like, look up "advanced centerline" folding method and go to it.
diy speaker guy is online now  
Sponsored Links
Advertisement
 
Reply DIY Speakers and Subs

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

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