Help, need good LF 12'' (300mm) driver - Page 5

sreten · 22nd July 2006, 03:38 PM

Hi,

I understand that at maximum stuffing density effective
box volume can be up to ~ 25% bigger. Different stuffings
have different maximum increases, and you have to be
careful not to overstuff as this drastically restricts bass.

I understand you will need ~ 4lb of fibrefill per speaker.

(fibrefill = BAF/Dacron/Polyester/hollowfil and other names).

If you can monitor the drivers impedance peak you keep
stuffing until the peak starts to rise gain, maximum
stuffing is at minimum peak value and frequency.

So yes, definetely 100% stuffing, at maximum density.

/sreten.

sreten · 22nd July 2006, 03:38 PM

Hi,

I understand that at maximum stuffing density effective
box volume can be up to ~ 25% bigger. Different stuffings
have different maximum increases, and you have to be
careful not to overstuff as this drastically restricts bass.

I understand you will need ~ 4lb of fibrefill per speaker.

(fibrefill = BAF/Dacron/Polyester/hollowfil and other names).

If you can monitor the drivers impedance peak you keep
stuffing until the peak starts to rise gain, maximum
stuffing is at minimum peak value and frequency.

So yes, definetely 100% stuffing, at maximum density.

/sreten.

georgehifi · 23rd July 2006, 09:37 AM

Streten I found a distortion graph on the SV12, it looks good to me up to 400-500hz, but then this is the first distortion graph I've seen on a drive unit, how does it look to you?

Cheers George

sreten · 23rd July 2006, 11:40 AM

Hi,

THD is not that helpful, ideally you want to see the levels of each harmonic.

e.g. 2nd harmonic, at high levels also known as doubling, doesn't
sound too bad, somewhat like valve amplifier bass, whilst the same
levels of 3rd harmonic sound much worse.

The levels of distortion in the graph look high, but the testing
level is not specified, so its hard to specify its relevance.

/sreten.

sreten · 23rd July 2006, 11:51 AM

Hi,

Suppose I should also add the following :

The tool I used to model the whole thing indicated
a stupidly peaking response around 70 to 80 Hz.
The c/o components looked completely wrong.

The curves I get modelling the driver as Re + Le
with the crossover components look sensible but I
know this does not take into account the capacitive
nature of the load or the impedance peak of the driver.

(sealed box speaker is inductive at high frequencies
going to resitive around the higher minimum, then
capacitive below this then the impedance peak)

Seems to me an active crossover is the way to go,
I'd be tempted to stick a quality sub plate amplifier
in each speaker and remove all bass c/o components.

/sreten.

georgehifi · 23rd July 2006, 12:14 PM

Quote:

Originally posted by sreten
Hi,

Suppose I should also add the following :

The tool I used to model the whole thing indicated
a stupidly peaking response around 70 to 80 Hz.
The c/o components looked completely wrong.
/sreten.

Exactly what I measured a few years ago, this was the fix a freind a I developed for it and it works like a charm.

http://67.19.167.226/~tdacquis/forum...read.php?t=275

Cheers George

sreten · 23rd July 2006, 02:49 PM

Hi,

So there are extra c/o components to consider....

The apparent ineptitude in the crossover design is staggering.
(implies only Re and Le were used for design - so wrong, and
what about testing the thing before you flog it to the masses ?)

Almost certainly you will need a new parallel network for the ACI.

According to the total sim having R2 in circuit is almost mandatory
if you don't have the conjugate bass impedance circuit connected.

It indicates without the impedance circuit that bypassing L1 or L2 to
give 5mH instead of 10mH response will be far better with the ACI.
(the simulator does not allow adding the conjugate circuit)
S1 (adding R2) with 5mH then seems to work very well.

Concerns about driver distortion seem academic with low frequency
inductors involved, unless they are of utterly herculean proportions.

/sreten.

georgehifi · 23rd July 2006, 08:15 PM

Streten, I never cared much for the -3 switch position, esspecialy after the LCR was instaled, it did reduce the output but also somehow made the bass muddy.

Streten I,m not understanding the last part of your last post too well, are you saying that I should short out one of the 5mh chokes when I use the SV12?

And I should re-design LCR for a different frequency/level hump that is going to be there?

Cheers George

sreten · 24th July 2006, 09:29 AM

Hi,

Download the trial Basta! from

http://www.tolvan.com/basta/

If you ask Svante nicely enough he may include the conjugate
network in a later version. Really it should already be in the tool
for accurate modelling of impedance corrected mids/tweeters.

For the ACI I get a peaking response at 45Hz, it looks a
lot better if you change L1 from 10mH to 5mH, well it did
whilst Basta! was working, seems to have stopped.

Will have to restart my computer methinks.

Working again 5mH gives a milder peak at 55Hz, looks
about right, crossover point does look around 100Hz.

With the Eminence parameters I get a utterly stupid peak at
60Hz, adding R2 reduces it somewhat, just looks all wrong.

/Sreten.

23rd July 2006, 08:15 PM	#48
georgehifi diyAudio Member Join Date: Sep 2003 Location: Manly Australia (Jewel of the Pacific)	Streten, I never cared much for the -3 switch position, esspecialy after the LCR was instaled, it did reduce the output but also somehow made the bass muddy. Streten I,m not understanding the last part of your last post too well, are you saying that I should short out one of the 5mh chokes when I use the SV12? And I should re-design LCR for a different frequency/level hump that is going to be there? Cheers George __________________ Avatar : Production Lightspeed Attenuator www.lightspeedattenuator.com

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
This JBL driver any good ?	Foxx510	Subwoofers	11	11th December 2010 06:06 AM
good driver vs bad driver	Sujies	Multi-Way	0	14th January 2006 05:21 PM
Looking for very good bass driver	slicemaster101	Subwoofers	31	2nd September 2002 06:52 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	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

22nd July 2006, 03:38 PM	#41
sreten diyAudio Member Join Date: Nov 2003 Location: Brighton UK	Hi, I understand that at maximum stuffing density effective box volume can be up to ~ 25% bigger. Different stuffings have different maximum increases, and you have to be careful not to overstuff as this drastically restricts bass. I understand you will need ~ 4lb of fibrefill per speaker. (fibrefill = BAF/Dacron/Polyester/hollowfil and other names). If you can monitor the drivers impedance peak you keep stuffing until the peak starts to rise gain, maximum stuffing is at minimum peak value and frequency. So yes, definetely 100% stuffing, at maximum density. /sreten.

23rd July 2006, 11:40 AM	#44
sreten diyAudio Member Join Date: Nov 2003 Location: Brighton UK	Hi, THD is not that helpful, ideally you want to see the levels of each harmonic. e.g. 2nd harmonic, at high levels also known as doubling, doesn't sound too bad, somewhat like valve amplifier bass, whilst the same levels of 3rd harmonic sound much worse. The levels of distortion in the graph look high, but the testing level is not specified, so its hard to specify its relevance. /sreten.

23rd July 2006, 11:51 AM	#45
sreten diyAudio Member Join Date: Nov 2003 Location: Brighton UK	Hi, Suppose I should also add the following : The tool I used to model the whole thing indicated a stupidly peaking response around 70 to 80 Hz. The c/o components looked completely wrong. The curves I get modelling the driver as Re + Le with the crossover components look sensible but I know this does not take into account the capacitive nature of the load or the impedance peak of the driver. (sealed box speaker is inductive at high frequencies going to resitive around the higher minimum, then capacitive below this then the impedance peak) Seems to me an active crossover is the way to go, I'd be tempted to stick a quality sub plate amplifier in each speaker and remove all bass c/o components. /sreten.

23rd July 2006, 02:49 PM	#47
sreten diyAudio Member Join Date: Nov 2003 Location: Brighton UK	Hi, So there are extra c/o components to consider.... The apparent ineptitude in the crossover design is staggering. (implies only Re and Le were used for design - so wrong, and what about testing the thing before you flog it to the masses ?) Almost certainly you will need a new parallel network for the ACI. According to the total sim having R2 in circuit is almost mandatory if you don't have the conjugate bass impedance circuit connected. It indicates without the impedance circuit that bypassing L1 or L2 to give 5mH instead of 10mH response will be far better with the ACI. (the simulator does not allow adding the conjugate circuit) S1 (adding R2) with 5mH then seems to work very well. Concerns about driver distortion seem academic with low frequency inductors involved, unless they are of utterly herculean proportions. /sreten.

24th July 2006, 09:29 AM	#49
sreten diyAudio Member Join Date: Nov 2003 Location: Brighton UK	Hi, Download the trial Basta! from http://www.tolvan.com/basta/ If you ask Svante nicely enough he may include the conjugate network in a later version. Really it should already be in the tool for accurate modelling of impedance corrected mids/tweeters. For the ACI I get a peaking response at 45Hz, it looks a lot better if you change L1 from 10mH to 5mH, well it did whilst Basta! was working, seems to have stopped. Will have to restart my computer methinks. Working again 5mH gives a milder peak at 55Hz, looks about right, crossover point does look around 100Hz. With the Eminence parameters I get a utterly stupid peak at 60Hz, adding R2 reduces it somewhat, just looks all wrong. /Sreten.