background image

10

Mackie Industrial White Paper 

 

 

 

 

 

 

 

 

                Noise  Sensing

September 2000

3.6  Factory Restore

If you nd that the EEPROM is corrupted or has not been ini-
tialized from the factory, you can perform a 

Factory Restore 

upon installing the SP-DSP1™ and running the SP-Control™ 
application.

 

A new SP-DSP© card should have all parameters 

of all presets set to default values (see Table 1).

3.7  Upload/Download From/To EEPROM

Occasionally, it is necessary to backup SP-DSP1™ presets and 
parameters or to transfer settings from one DSP to another. 
To copy the contents of the EEPROM, simply 

Connect 

to 

the DSP card and then select 

Upload From EEPROM. 

The 

EEPROM contents are stored in your Palm™. If you would 
like to copy contents of the Palm™ into another SP-DSP1™ 
card, select the 

Download To EEPROM. 

Be advised that this 

will completely erase the existing contents of the EEPROM 
and replace with what was uploaded to the Palm™. Be careful 
to do an 

Upload From EEPROM

 

before

 the 

Download To 

EEPROM

 to ensure that valid data exists in the Palm™.

4  Tips for Setting User Parameters

To clarify, setting the user parameters (

MG, GR, NT, NR, 

Attack Time and Release Time

) can be done either before or 

after the 

Auto Calibration

. These parameters have no bearing 

on the calibration and can be changed at any point. In fact, a 

Preset Save

 and 

Recall

 only affect these parameters. 

The 

Auto Calibration 

only concerns itself with the Calibration 

or 

CAL

 value and the adaptation coefcients. The user param-

eters can be set up for runaway gain, but the user would 
notice this immediately as the system would “runaway” to 
the maximum gain. During calibration, the user parameters 
are temporarily overridden to perform the calibration at a 
xed gain (easier and better results). However, the xed gain 
used during the calibration process is the Maximum Gain 
(

Minimum Gain + Gain Range

). It is probably advisable for 

the user to set up these two parameters so that he isn’t “blown 
away” by the calibration sequence if his Maximum Gain is 
louder than he intends. The Maximum Gain is used as the 
calibration value because the algorithm adapts more quickly 
when the music is loud. The calibration itself doesn’t limit 
the parameters the user can adjust in any way, it just provides 
a ‘hidden override’ when he has his 

Noise Threshold

 set too 

sensitive, which would otherwise cause runaway. You’ll notice 

that once your unit has been calibrated you can set the 

Noise 

Threshold

 as low as you want and you will still seem to get 

the same sensitivity. This is because it is being limited by the 
Noise Threshold Override (see Figure 1).

As you can see, one of the rst questions that the end-user 
needs to answer is “How much overall system gain or attenua-
tion do you want the noise sensor to provide?” By answering 
this question you have determined the setting of the 

Gain 

Range

 (

GR

). A typical good starting point for the Gai

Range 

is about 20dB. If you nd the system does not attenu-

ate enough or attenuates too much, adjust accordingly. By 
answering this rst question, you are likely to determine the 

Minimum Gain 

(

MG

) as well. Since the average user would 

want his system to attenuate from the unity point (0dB), 
which is the highest sound level he would ever want, for 

this example the 

Minimum Gain

 would be set to –20dB. As 

mentioned in Section 3.1 User Parameters (Sliders), by setting 
the 

Minimum Gain

 and the second parameter, 

Gain Range

you are actually establishing the program operating window 
within which the levels for the program material must remain. 
As in the previous example, if the user wanted his music levels 
to operate in a range 

±

10dB around -10dB down from the 

input level (unity point), he would set 

MG

=-20dB and 

GR

=20. 

Setting the 

Noise Threshold

 (

NT

) can only be done by deter-

mining the noise level at which you want the music to start 
to increase in level. This noise threshold point is best found 
through trial and error. You must be careful not to set the 

NT

 too low (i.e. more negative), because the noise sensor will 

begin “gaining up” the program material with very little noise. 
However, setting the 

NT

 too high will prevent the system from 

making any program gain increases until there is a relatively 
high noise level. There is a balance and each room will require 
a different 

NT

 setting. Keep in mind that setting this level in 

an empty room may require some “tweaks” when the room 
is full with people. For example, if you set 

NT

=-40dB in an 

empty restaurant, you may nd that the system starts gaining 
up too fast and is too loud for the occasion. In this case, you 
can reduce 

NT

 (by bringing the level closer to 0dB) by 5 to 

10dB. This would give you 

NT

=-30 to -35dB and the system 

would not start “gaining up” until the noise was 10dB “hotter” 
than it was previously.

Noise Range

 (

NR

) actually sets the amount of noise that makes 

the program gain go from the minimum setting to the maxi-
mum setting. That is, if 

NR

=40dB and 

GR

=20dB, then for 

every 2dB change in the Noise, the Program level is only 

Содержание SP-DSP1

Страница 1: ...noise The question naturally arises why can t this be done automatically Mackie Designs has invested a considerable amount of time in research and development to nd an answer to this very question In...

Страница 2: ...signal It computes its own approximation of the room response in order to cancel the music signal from the signal picked up by a room microphone 1 5 7 Room size is the most important factor determinin...

Страница 3: ...constantly adapting to the room characteristics This provides optimum performance when the room acoustics change Room acoustics can change signi cantly due to the arrangement of furnishings opening or...

Страница 4: ...from speaker to microphone including re ec tions are removed by the nLMS algorithm as there would have to be many re ections before the sound could have trav eled this far Each re ection reduces the e...

Страница 5: ...th a faster release rate will reduce the level of gain applied to the music This allows the compander to track the ambient room noise while rejecting these singular events if desired 1 6 Auto Calibrat...

Страница 6: ...ch null modem adapter cable is provided to connect a standard Palm Cradle or the HotSync Cable to the 9 pin female D Sub on the front of the SP2400 1200 After installing the application to your device...

Страница 7: ...ith the fourth param eter Noise Range actually sets the operating window of the noise source This noise window sets the level and range that the noise must be within to effect the program level Noise...

Страница 8: ...level fades 3 3 Bar Graphs Metering The main screen of the SP Control Palm application has four meters that allow the user to monitor levels during setup and normal operation see Figure 2 PI is the Pr...

Страница 9: ...ime this is not a critical require ment Once the speaker s and ambient microphone are in their xed locations and the microphone gain and input levels have been adjusted per Section 3 3 Bar Graphs Mete...

Страница 10: ...een calibrated you can set the Noise Threshold as low as you want and you will still seem to get the same sensitivity This is because it is being limited by the Noise Threshold Override see Figure 1 A...

Страница 11: ...ice he can use HyperTerminal available on any PC running Windows OS HyperTerminal can control all the parameters previously mentioned The null modem adapter is not necessary as the 9 pin female D Sub...

Страница 12: ...oolbox and a New Toolbox for Matlab Simulink He has numerous publications in IEEE and one in AES see www ece uvic ca dale cv pdf Dr Shpak is a member of the IEEE Brian Roden was born in Ragina Sask Ca...

Отзывы: