Speakerplans.com Homepage
Forum Home Forum Home > General > Advanced Discussion
  New Posts New Posts RSS Feed - stack processing
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

stack processing

 Post Reply Post Reply Page  123>
Author
Message
sgarfa View Drop Down
Registered User
Registered User


Joined: 10 December 2013
Status: Offline
Points: 38
Post Options Post Options   Thanks (0) Thanks(0)   Quote sgarfa Quote  Post ReplyReply Direct Link To This Post Topic: stack processing
    Posted: 10 April 2017 at 8:24am
good morning, I have to ask advice for the setup of the stack in the picture,(the system consist in 2 of these)

the crossover points ferequenze are (24db btw); 

2x (1x21 ") sub 30-60hz 
wbin 2x15 60-250hz (approx 95cm horn, but rear reflex loaded)
tms 18 "60-250hz (100cm horn approx)
tms 10 "+1" 250-20000

for tms 4 I used the delay of Turbosound, 0 deay for the 18 "and +3,292 for the 10 + 1 with the recommended crossover points (24db lr).
for sub and the wbin I used Smaart, considering the low sections of tms as starting point, having regard to the length of the horn, then either tried to overlap the phase lines

2x (1x21 ") 4.5ms, polarity +
wbin 2x15" 60-250hz 0.750ms, polarity +
tms 18 "60-250hz 0ms, polarity +
tms 10 "+1" 3,292 polarity -

Given the length of horns, I did not expect so long especially results for subwoofers I thought of 3ms, even at the same time , I had found online a set up that talked about +7ms, - polarty for refelex sub 18" and 18" of the tms4 (david's x crossover settings)
The microphone was placed 3m on the ground and the auto delay of smaart gave me a reasonable delay 20ms delay (calcutated on tms4 18" section and the blocked) ...
Now I see me the difference between delay physically calculated (horn length), measured delay (maybe I did something wrong) and found on line delay ..
They are the first to say that the system is not the top, and has a lot of overlap, but as you know, have to do with what you have ..
at the end of the speech, find reasonable the delay results .. listening does not look bad, but you know sometimes we are wrong autoconvince



Edited by sgarfa - 10 April 2017 at 8:28am
Back to Top
toastyghost View Drop Down
Old Croc
Old Croc
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 8513
Post Options Post Options   Thanks (1) Thanks(1)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 9:11am
Autodelay is not appropriate for sub measurements. There simply isn't enough data in the FFT down low for it to get a clear impulse to lock onto - and subs will be stretched over many ms of time as well so it doesn't know where to decide.

You would be better off putting the mic further away, ideally equal distance to each speakers' mouth. Then use a tape measure / laser measure to the lowest freq enclosure to get you in the right ballpark (convert distance to ms manually) enter this as SMAART alignment delay and then see how the phase trace looks wth that cabinet solo. Ideally you want it to be flat through the intended crossover region so bump the delay time up or down a ms at a time until you achieve that. Then LEAVE IT ALONE.

Save the trace and take traces of all the next cabinet up. Save that. You want to then delay the cabinet with the GENTLEST sloping phase around the crossover region. It is likely you will need to adjust crossover points and slope types slightly and repeat the process, ideally you should have the point on magnitude trace where they meet both boxes at -6dB to sum smoothly without bad sounding peaks at crossover.

Once phase slopes overlap ideally to one octave either side of crossover point, and magnitude and slopes are set, measure both boxes together to verify it worked and traces loook good together. Then repeat with the next cabinet up, but TREAT THE BOXES YOU JUST ALIGNED AS ONE CABINET now.

You will likely need to do this a whole bunch of times before you're happy. If you can't get there then somebody more experienced may be worth calling on, measurement is a skill you need to learn over time, as the software only gives you data - you have to decide what to do with that data and there's often several approaches depending on the result you want.


Back to Top
sgarfa View Drop Down
Registered User
Registered User


Joined: 10 December 2013
Status: Offline
Points: 38
Post Options Post Options   Thanks (0) Thanks(0)   Quote sgarfa Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 9:41am
ok .. thanks a lot of info, but I find around conflicting information ... and also in the Manual Smaart spoke to set before the subwoofer and to  to help the auto delay  increasing the frequency low-pass, I thought the 250Hz low the tms4 are sufficent.
 it is the rule of the microphone adjusts to the ground ?... in this stack for the same L lenght  you have to keep it well up..
then last thing, in my case the system is mixed reflex /horn,,, subwoofer is what requires more delay, always I start with him?
thanks a lot



Edited by sgarfa - 10 April 2017 at 9:44am
Back to Top
bob4 View Drop Down
Young Croc
Young Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1020
Post Options Post Options   Thanks (1) Thanks(1)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 1:33pm
If you must use the wbin, I would measure & audition the wbin with closed/stuffed ports. 

My suggestion might sound a bit odd, but here is my reasoning: The wbin is ported in order to gain additional output around the lower cutoff, to extend useable freq response below the front loaded horn paths cutoff frequency. This is necessary if the wbin is used as the main subwoofer system without infrasub. The way you are using it in your system however, the wbin should be able to fulfill its purpose without the port. 

The benefits of closing the reflex ports are 1) decreased group delay and 2) smoother phase response. This should make it much easier to integrate the wbin into the system. Especially since you are running it in parallel with BOTH your 21" reflex (in the crossover region), and with the turbo kick (60-250). 

Actually what you are doing is a pretty bad idea in general (in my opinion, no offense!). It means that you have three different cabs playing around 60 Hz (remember they are overlapping). If you don't get it perfect, you will have cancellations and problems right in the middle of the bass range where a lot music contains the major part of its bass energy. 


My opinion on microphone placement for system tuning:
As you can see in your sketch, for a high stack like yours, measuring close up introduces big path length differences that depend on the height the microphone is placed at. Only at longer distances will the offsets become less. To overcome this problem, you have to adjust only one crossover area at once. Pick a distance from the stack, for example 1 m. Now start from the subs and wbin, put the mic 1m away from the stack, exactly at the height between the two boxes (top of the subwoofer, bottom of wbin). This way both loudspeakers should be at the same distance from the mic. Do the measurement, and note the time delay on the wbin. Move the mic up to the height between wbin top and Turbo bottom, and now measure those two. 

With this method you should be able to time align the cabs and make sure that they all radiate synchronously in the vertical. If they are not, the system may exhibit lobing. Check this rane application note for more information on lobing:  




Edited by bob4 - 10 April 2017 at 1:40pm
Back to Top
sgarfa View Drop Down
Registered User
Registered User


Joined: 10 December 2013
Status: Offline
Points: 38
Post Options Post Options   Thanks (0) Thanks(0)   Quote sgarfa Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 2:21pm
hi bob... thanks for the precision...  I am also to the idea to close the reflex ports, but this spekers they are also used alone... since they are tuned at 50hz  I thought that by cutting enough up (60 instead 40/45)  the phase response feel just minus the contribution of the reflex, in any case not cost me anything to do a test with a new panel..
good idea to do a crossover area at the time...but only 1m? (ok is for example., but it's a resonable distance?....)..


Edited by sgarfa - 10 April 2017 at 2:21pm
Back to Top
toastyghost View Drop Down
Old Croc
Old Croc
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 8513
Post Options Post Options   Thanks (0) Thanks(0)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 2:33pm
Floor bounce will be strongly apparent even at 1m, and if you align subs like that it'll be way out of time further back.

Mic should be on the floor, ideally with something absorbent in front to minimise bounce. You don't have to get exactly the same from all boxes but ideally it should be close. Sub to top alignment will also only work for that stack config when done, if you move the tops or split subs ever you will need to remeasure.

I would start with the 21", it is playing lowest and will have large group delay, but honestly it doesn't matter which really as long as you delay the cabinet with gentlest slope in phase trace.
Back to Top
bob4 View Drop Down
Young Croc
Young Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1020
Post Options Post Options   Thanks (0) Thanks(0)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 2:54pm
Originally posted by sgarfa sgarfa wrote:

hi bob... thanks for the precision...  I am also to the idea to close the reflex ports, but this spekers they are also used alone... 

You don't need to change your cabinet and permanently close the reflex ports. If you use the wbin in this configuration, you can temporarily close the reflex ports by putting something like a piece of cloth, or a big piece of plastic foam, or something similar inside them. The idea is very simple, no air can move inside and through the port. Sleepy


regarding floor bounce, toasty is right, there will be some. measuring the sub/wbin crossover on the floor will probably  achieve better results. But for the wbin/turbo alignment (they are playing the same freq range, 60 - 250, right?) I would highly recommend you use the method I desrcibed for mic placement.

Good luck Thumbs Up
Back to Top
bob4 View Drop Down
Young Croc
Young Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1020
Post Options Post Options   Thanks (0) Thanks(0)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 2:56pm
Originally posted by sgarfa sgarfa wrote:

good idea to do a crossover area at the time...but only 1m? (ok is for example., but it's a resonable distance?....)..

mic placement is always a compromise, but if you have any kind of background noise, it will force you to go closer to the speaker and make loud and short sweeps. 
Back to Top
sgarfa View Drop Down
Registered User
Registered User


Joined: 10 December 2013
Status: Offline
Points: 38
Post Options Post Options   Thanks (0) Thanks(0)   Quote sgarfa Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 3:12pm
Originally posted by bob4 bob4 wrote:

You don't need to change your cabinet and permanently close the reflex ports

yes mister!...  i'm ramirez Tongue Pig!...it's my heavy contruction and.. push right ... and I'd like to preserve the lower section of tms4 using these .. but doing everything to contribute (proper)
Back to Top
sgarfa View Drop Down
Registered User
Registered User


Joined: 10 December 2013
Status: Offline
Points: 38
Post Options Post Options   Thanks (0) Thanks(0)   Quote sgarfa Quote  Post ReplyReply Direct Link To This Post Posted: 10 April 2017 at 3:33pm
Originally posted by toastyghost toastyghost wrote:



I would start with the 21", it is playing lowest and will have large group delay, but honestly it doesn't matter which really as long as you delay the cabinet with gentlest slope in phase trace.

Ok ok ... recapping:

  • i put the mic on the floor...i measure the physical distance between mic and subwoofer, and trasform in ms, this is SMAART alignment delay (foot the mid/high i use autodelay?)
  • i see the plots of the subs, then i freeze it and i turn of subs and on the wbin
  • now i see the wbin plots and i try tu align gentlest slope (whith previously subs plot)
  • now i turn on booth subs an wbins, and see their plots like one spaker
  • move mic  between wbin and  tms4 (?)
  • do the same thing made between sub and wbin (whit new SMAART alignment delay)
Sorry for the confusion, but I think it's a useful thing to many, as I have several different speakers to align
Back to Top
U.Viktor View Drop Down
Young Croc
Young Croc
Avatar

Joined: 04 May 2010
Status: Offline
Points: 527
Post Options Post Options   Thanks (0) Thanks(0)   Quote U.Viktor Quote  Post ReplyReply Direct Link To This Post Posted: 12 April 2017 at 8:43pm
Due to the reflections and other environmental signal propagation anomalies there is just simply impossible to measure anything properly without silent room.
Back to Top
bob4 View Drop Down
Young Croc
Young Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1020
Post Options Post Options   Thanks (0) Thanks(0)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 12 April 2017 at 9:24pm
Originally posted by U.Viktor U.Viktor wrote:

Due to the reflections and other environmental signal propagation anomalies there is just simply impossible to measure anything properly without silent room.

wow this is a candidate for most useless answer of the month.

of course it depends on your definition of "properly", but if, as in your opinion, it's not worth doing acoustic measurements  unless you are in an anechoic chamber, no one would ever measure anything, and nothing would ever get done. Wacko

more constructive, practical advice please?
Back to Top
 Post Reply Post Reply Page  123>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 11.10
Copyright ©2001-2017 Web Wiz Ltd.

This page was generated in 0.078 seconds.