difraction compensation - Page 3 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

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 January 2002, 08:51 PM   #21
Super is offline Super  United States
diyAudio Member
 
Join Date: May 2001
Location: Connecticut
Send a message via AIM to Super
Perhaps your high pass filter is set too low? Low frequencies and high SPL can do a number on tweeters...
__________________
Bryan
  Reply With Quote
Old 10th January 2002, 05:34 AM   #22
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
Nope it was 1800Hz 18db per octave. That is what scared me.

I had a MTM using a 18db octave high pass and 6db octave low pass using 2 8" Focal woofers and 1 Morel Tweeter. The Sens. was about 91-92db. It roacked. My nephew used it to blasat the neighbourhood a few times.
  Reply With Quote
Old 11th January 2002, 11:23 AM   #23
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
Default difraction

Has anyone used LSP CAD to design a crossover?
  Reply With Quote
Old 11th January 2002, 06:44 PM   #24
weeghel is offline weeghel  Netherlands
diyAudio Member
 
Join Date: Aug 2001
Location: Netherlands
Send a message via ICQ to weeghel
Default I'm using 8546'es too

Hi,

check out my speakers at my website . I also use 18w/8546'es

I can tell you that you will indeed encounter baffle diffraction. And you will also have to deal with it in the crossover. Also, the value of 6dB is a little high, but not by much.
I have had my speakers tested once, and John Murphies formula is quite correct.

I recommend that when you have finished your speakers, you should have them properly tested and design a crossover with a computersoftware package.
Not just by using the textbook formulas. (although they do give a decent startingpoint)

I still have to finish my speakertestjig, but when I've completed that, I will measure the speakers myself and start over on my crossover, using LspCad standard. (now that I know a little more about designing them)
  Reply With Quote
Old 14th January 2002, 06:12 AM   #25
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
all I have to test are my ears!

I live in India as far as I can tell I am the only DIYer here

Regards
Navin
  Reply With Quote
Old 14th January 2002, 09:44 AM   #26
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
Default difraction compensation

I see a lot of speakes today are using the side firing woofers that NHT first popularised for another system (not one that was dicussed earlier using 12" woofers and Scan Speak drivers) I am thinking of using a slightly different driver topology.

If one remembers the Allison 9 (Roy Allison had a very popular company in the 80s) it had a woofer than was a 45 deg. angle with the floor. An easy way to visualise this is imagine a front facing woofer at floor level, now if that woofer base of that cabinet was not as deep as the reat teh woofer would be firing at an angle facing somewhat downwards.

Anyway if one had a 8 or 10" woofer that was facing downwards at 45 deg. then one topped of teh cabinet wiht a 6" 2way (in my cse Vifa TC series) one could have a credible 2.5way. Using sheilded 6" 2 way one could use these speakers close to a TV fo Home Theater.

Just another idea. The concept of a 2.5 way is too simple and solves too many problems (difraction sensitivity, bass output, retaining imaging of 2 way monitors, speed of bass assuming the drivers are decent) to be ignored.

Let me know if anyone has other ideas.

Regards
Navin
  Reply With Quote
Old 14th January 2002, 12:21 PM   #27
diyAudio Member
 
Join Date: Dec 2001
Location: Milano, Italy
Hi Navin,

1 - bass section. It should be ok: with the dimension you gave, and assuming a 2 cm wall thickness, taking out 10-12 litres for the Scan subvolume, you should have about 100 lt physical volume available for the woofer.
The simulation gives a Fc of 35Hz and a Qtc of 0.79 for the box completely filled with dumping material. The Qtc seems a bit high, but this time the real world is nicer than the simulation. In effect, in case of big volumes and big qty of dumping material, the real behavior of the system is different than the simulation, and you can expect both Fc and Qtc a bit lower than expected. It is difficult to predict, but I think we can assume 33Hz and 0.75.
1b - A couple of hints: with your cabinet layout you have 2 big walls (the laterals) that are likely to vibrate. Better counter this, i.e. using some 3x3 cm fillets to connect the 2 walls, displaced irregularly to better break the resonances.
Also a good thing is to cover the interior of the walls with those tar sheets that are used by car body-repairer to silence the car plates. Over here they are available at retail car do-it-yourself shops in 50x50 cm sheets, but if you know a body repairer it should be easy to get some.
Damping material : a good compromise is that kind of Dacron used for sofa, pillows etc. Ideal is to find a craftsman or small industry who works in that business and beg him to buy some from his stock. It normally come in rolls of 2x50 mts, 2-3 cm thick.
If you buy it from a HiFi retail you will pay it 50 times more.

2 - Res of the coil. Yes, a R in series to the woofer increases the electical Q, so the Qt and the related Qtc, in addition to reduce the SPL output. These effects are taken in consideration in the simulated graph, like the diffraction effects of the panel.

About helping the Scan above 100-150Hz: I don't know if this is a good idea. The Scan is limited only by his thermal ceiling in that range, and is a very well-sounding speaker. I don't know the woofer, but looking at his TS parameters looks more a sub than a woofer, and his high moving mass does not promises a high speed and transient control. I believe it's better to leave him deal with the frequencies he is designed for!

3 - Tweeter filter topology: by putting the R between C and L you obtain a 'dumping' of the filter, so changing the slope of the system (filter+speaker+panel: never forget that your REAL slope is the one of the complete system, the electrical part is jus a component of it. And at low frequencies a big part is played also by your room and the relative positions of the walls and the woofer).
Anyway, by reviewing the project, I've found that there is a better (predicted) result by placing the R in a more conventional position (after the CL cell)

4 - By using the 10 ohm in parallel you damp the impedance, but also you create an attenuator (the 1.8 series + 10 parallel), so you change the attenuation (vs 1.8 series + infinite parallel).
See usual attached graph.
Anyway, I see no reasons to damp the impedance peak: the 9900 resonates at 500 Hz, and we are filtering it at about 2500.

5 - about the 'execution' of your Morels, it could be helpful to know their specs and the filter you used to see if there was some strange thing.
I have a suspect: which ampli you were using? By experience I've killed much more tweeters with a low power amp in clipping than with big monsters.

6 - As soon as I can (probably late in the afternoon) I will send you separately the file with graphs etc. My problem is that the only way I know to send a 'print screen' is to copy it into a word or excel file. If you can teach me a smartest way I will be grateful forever!

bye
sandro
  Reply With Quote
Old 15th January 2002, 05:14 AM   #28
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
Actually that is the only way. I can read word and excel files with Star Office so that is no problem. I had written a program in my early days to capture the screen and all that but it is licensed to Karl Heinz Fink of ALR-Jordan who gave me the drivers. Actually he gave me the drivers and did not want money. The I found out that he need a small progrm written so I wrote the program and gifted him the source code. Of late he is not doing well so I try not to disturb him.

The tweeters were Morel MDT33 circa 1988. I blew them in 1990 usig a Harmon Kardon PM 665 integrated amp that was rated at 100W / 8 ohms 150W at 4 ohms. The drivers were crossed over using a simple 18db XO. 8.2uf, 0.43mh, 24uf and was in MTM with 2 Focal 8N515 woofers. This was my 1st system (my requirements were high sensitivity 92db+, high power handling 100W, 2 way).

I know i wold have to use some bracing so effectively me box volume will reduce by 10% for bracing and driver.

yes teh scan is a far better driver than the 12" except it can produce a lot of bass so it makes snese to use the 12" only for the last octave or one and half octave.

Thanks for your help.
  Reply With Quote
Old 16th January 2002, 02:48 PM   #29
diyAudio Member
 
Join Date: Dec 2001
Location: Milano, Italy
Hi Navin,

I am just 48hr late, not too bad for an Italian!

I have sent you a wordfile with graphs, hope it's clear.

1 - Your new project
Is now simulated with the real woofer parameters, and I have corrected some of my assumptions.
Basically the mid-high x-over is a bit lower, and I have reworked the low-mid filtering the mid with a cap.
The tweeter is now attenuated with a serie/parallel net (c'mon, is 'partitore resistivo' in Italian but I have no clue about how to name it in english!). Now we have the right attenuation and the impedance peak damped (Fig.1)
Also the C and L values are changed.
Midrange has now a high pass section with a 313 Micro (a 500 cap is too big to attenuate the bass freq, it is practically useless), and the coil on the woofer is 6mH. I still recommend to try the choice with free mid + 9mH coil on woofer, maybe you can alternatively try both solutions and pick up the best sounding one.

The tweeter should work safely, since is attenuated by 10 dB at 2 kHz (Fig. 5) , see the situation of the old Morel in Fig.7 for comparison.

Also the other control simulations seems to say that the project should be correct, better refinements are up to the best existing measurement system (the ear)
Let me know if some of the graphs is not clear.

2 - Morel's death.
I have simulated the MDT33 with the data provided by www.morelusa.com, and with the filter you used (Fig.7)
Well, the MDT33+filter was surely not protected, if your nephew cranked up volume and tone control it is possible that the tweeter had more power that he could handle.
But who knows? It could also be that your speakers were from a non lucky bass, I had problems with Audax and Macrom (a Morel brand name for car hifi) burning and a super expensive Focal 10" woofer whose suspension simply detached from the cone. We 'opened' 5 woofers before Focal recognized that the batch had a glue quality problem!

By the way I well remember the MDT33, it is a great tweeter, while the HK was very good amp.

Hope I have helped you, but I want to give one last hint: try to get and learn a good CAD software, it is really impossible to design a speaker only with handwritten calculations.
The one I use is very complete and accurate, but is written in Italian. If you know somebody that can help you with the language (or you have time to invest for the translation) let me know and I'll send you a copy of mine to try before you buy your one (since the cost is only $40 it makes no sense to be a pirate)
The author is a hifi review : http://www.audioreview.it/

But most likely somebody in the list can help you in the choice of a good CAD in english!
bye
sandro
  Reply With Quote
Old 17th January 2002, 05:57 AM   #30
navin is offline navin  India
diyAudio Member
 
navin's Avatar
 
Join Date: Jan 2002
Location: Mumbai (Bombay), India
Send a message via MSN to navin Send a message via Yahoo to navin
I hope you are still talking baout the project involving the 12" Audio Concpets woofer, ScanSpeak 6" and Scan Speak tweeter.

The second project (for the small room 150 sq. ft) is using all drivers (8" Eminence woofer, 6" and 4" Vifa TC series woofers, and Vifa TC series tweeters) to which I have few if any specs. it is not critical and I will play it by ear. i still have not decided on the design.

back to the first project.

I am thinking of getting LspCad but dont know anyone who has used it. It is also not very expensive. It is often not the expense but the govt. foriegn exchange regulations that prevent me for paying for this.

I agree with teh midrange Cap. I was thinking 250uf (470uf x 2 is series + 10uf polyester + 2 uf wonder cap) but I can always add 50uf more of polyester cap. so i can use 470uf x 2 is series + 7 x 10uf polyester caps and 2 uf of wonder caps = 307uf. Actually the 10uf caps are about 10.2-10.7uf so it will add to 310-315uf. only it will be a lot of caps (10 caps). the biggest polyester caps here are 10uf but are of good quality.

will check your file and email later.

Regards
Navin
  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
Baffle step difraction JC951t Multi-Way 7 18th March 2009 12:09 PM
Gunderson compensation.. mikeks Solid State 155 25th September 2006 02:14 AM
Baffle Difraction & ralph-bway Multi-Way 36 9th January 2005 01:26 PM
Reduce difraction and distortion from the cabinet on DIY designs. Bedroc Multi-Way 1 24th February 2002 03:09 AM


New To Site? Need Help?

All times are GMT. The time now is 04:59 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