Real world tuning @ a theatre   Leave a comment

Just the overview of the room to start with .

The models used in this venue are :

Meyer sound UPA : left and right.

Meyer sound UPM : left and right infill underneath the UPAs.

Meyer sound UPJ : 3 x delay speaker (not visible in the picture).

Meyer sound Mina : 5 cab line array divided into 2 (3 for far field 2 for mid/near field)

Meyer sound HP500 : flown at the centre cluster position in a point source configuration (not yet done in the picture below) .

Meyer sound MM4 : infill system for the first row .

2 Renkens-Heinz 15” horn loaded sub woofers left and right underneath the stage (not visible in the picture).

Meyer sound Galileo 6-16 system processor.

Strt overview of the room

Mina as a centre cluster has to be delayed to match with the UPA lft/rght . So i synchronize via an IR measurement on the UPA and turn it off . Then turn on the Mina cluster and delay it until the the phase of the high-end of this system matches the UPA .

So this is the response of the mina centre cluster .

0001

And this is the response of the UPA . The high-end of both systems matches but up to 1kHz there will be a problem if the UPA is not matched in phase to the mina system via an all-pass filter .

0002

The screen shot below shows that the Mina system has an extra phase wrap which will cause a dip in the frequency response between 500Hz and 1kHz because somewhere in this range there’s a phase difference of 180 °.

0003

So as said before if i some both systems together like the screen shot below there are a few hints of the 180° phase difference :

The area of interest (500Hz to 1kHz) shows a big jump in the phase display .

Coherence (the red trace ay the top of the magnitude screen) tumbles down indicating there’s a problem .

The magnitude response shows a rather large dip in the frequency response . 

0004

So by applying a 2nd order allpass filter on the UPA @ in this case 581Hz the response of the UPA matches with the Mina array . (i hope i have that trace saved but it’s not saved on this computer yet so….not yet maybe later).

For the rest of this post I’ll be using smoothed data for visibility reasons but remember : the real data looks a lot different from these screen shots from smaart which is at a 12 oct view from here on .

By applying the allpass filter @ 581Hz the dip at the area from 500Hz to 1kHz disappears when both system are summed together . In the bottom screen shot the blue trace is the sum of Mina and the UPA(no allpass filter). The other trace is the sum of Mina and the UPA(with allpass filter) and shows the dip has gone to analyzer heaven . 

0005

So as a infill system mounted below the UPA is a UPM . Since the UPA’s phase response has been altered to match with the response of the Mina system i also have to alter the response of the UPM . So the same procedure applies here . I have to delay the UPM to the UPA and after that i need to apply an other allpass filter to match the UPM to the UPA thus matching them both to the MINA system .

So i synchronize on the UPA with the mic @ the onax position of the UPM . The phase trace of the UPM when soloed will have a lot of phase wraps with the angle of the trace going up (sorry did not save that trace) meaning in order to get the high end to match again i have to use a bit of delay on the UPM .

0006

So UPM is delayed by 1.94ms and matched @ the high end of the system but again i need to correct the area between 500Hz and 1kHz .

0007

The sum of UPA and UPM(no allpass yet) again shows a large area of cancelation @ the area between 250Hz up to 1kHz due to a 180° phase difference and about equal level somewhere in this area .

0008

So the screen shot below shows the combination of UPA and the UPM with delay and a allpass filter in red compared with the same combination but no allpass filter on UPM . Again the dip has been solved and shows addition in the area of interest (about +6db).

0009

So still i had 3 delay’s to phase align to the mina array and again the same procedure applies . Sinq in on the mina array with a mic placed on axis of the first UPJ .

0010

Adjust the delay on the UPJ to match the phase response of UPJ to mina @ the high end of the system . Again clearly visible The UPJ(blue-ish) needs an allpass filter in the same region as the UPA & UPM before so around 500Hz .

0011

So the phase response of UPJ has been altered via an allpass filter to match with the mina array . Sorry to say i did not save a measurement of the sum of both pre & post allpass alignment but the idea should be clear to you now .

0013

A lot of data will not be shown here because i save a lot of it just for comparison . The next challenge in this theatre was the configuration of the 2 HP500 sub woofers . Behind them is a wall which causes cancels because of reflections from it . To sort overcome this the subs are reversed so the speakers are pointed to the wall giving a bit more level . The distance between the 2 subs causes phase differences at the side of the array so cancels from the side walls are somewhat avoided . The disadvantage is that it causes less level on the sides of the area to be covered .

So i decided to try something else which is a point source configuration . First the old configuration when i walked in to the venue .

Old cofig for sub and centre cluster

 And the measurement data from this set-up . There is a challenge because the reflections of the back wall are somewhat solved but probably the ceiling is still causing problems .

0014

So this is the new configuration . The subs are now in a “point source” configuration . The reflections from the back wall are still there but the different arrival time between the subs and the reflections of them is small enough to not cause cancels . The ceiling is a different story all together but i thought of it during writing this post so that’s for an other day in the future .

Fulco new config 02

Fulco new config 01

The measurement data from the new configuration (red-ish) shows an improvement compared to the data from the old configuration .

0015

So finally this is the result . The dark red measurement is the whole set-up so all speakers are on with the addition of 2 Renkens-Heinz sub woofers which are located on the floor underneath the stage . The trace that’s not high-lighted is all on with just the HP500 subs .

0016

Geef een reactie

Vul je gegevens in of klik op een icoon om in te loggen.

WordPress.com logo

Je reageert onder je WordPress.com account. Log uit / Bijwerken )

Twitter-afbeelding

Je reageert onder je Twitter account. Log uit / Bijwerken )

Facebook foto

Je reageert onder je Facebook account. Log uit / Bijwerken )

Google+ photo

Je reageert onder je Google+ account. Log uit / Bijwerken )

Verbinden met %s

%d bloggers liken dit: