BillFitzmaurice.info

Cheers Brent.

Won't hitting the limiter (regardless of release time) have the same effect though? ie the pulsing volume pattern?

edit:
From http://www.doctorproaudio.com/doctor/te ... s_en.shtml

"If it is the crossover unit that is taking care of the limiting, in practice we have a multiband compressor and, if compression attack and release times are user selectable, we will need to chose faster timer for the high frequencies and slower ones for the low frequencies, thus optimizing the compromise between protection and audibility."

Perhaps 407msec is to long, but couldn't faster release times also be a problem dependant upon frequency?

Grant Bunter wrote: Won't hitting the limiter (regardless of release time) have the same effect though? ie the pulsing volume pattern?

With a short release time, you're just cutting off the damaging voltage peaks without otherwise affecting the source material. You really don't hear the limiter unless you get into it hard. With a longer release time, the effect is far more audible becasue the overall level drops noticeably, and an inexperienced operator will get hard into the limiters, then respond by turning up.

From http://www.doctorproaudio.com/doctor/te ... s_en.shtml

"If it is the crossover unit that is taking care of the limiting, in practice we have a multiband compressor and, if compression attack and release times are user selectable, we will need to chose faster timer for the high frequencies and slower ones for the low frequencies, thus optimizing the compromise between protection and audibility."

This is a bit different. We typically don't have multiband compressors at our disposal in live situations. Even with a broadband compressor, though, you would typically select long attack and release times to do a form of RMS limiting. It's a good option in some situations, but it can create problems too, as you then have a situation where gain can be raised without your direct adjustment.

Perhaps 407msec is to long, but couldn't faster release times also be a problem dependant upon frequency?

Release times aren't usually a problem. With a brickwall limiter, you can usually have an almost instant release and have no problems. Attack times are a different story.. but brickwall limiters have an attack time of 0, regardless.

BrentEvans wrote: This is a bit different. We typically don't have multiband compressors at our disposal in live situations. Even with a broadband compressor, though, you would typically select long attack and release times to do a form of RMS limiting. It's a good option in some situations, but it can create problems too, as you then have a situation where gain can be raised without your direct adjustment.

Perhaps 407msec is to long, but couldn't faster release times also be a problem dependant upon frequency?

Release times aren't usually a problem. With a brickwall limiter, you can usually have an almost instant release and have no problems. Attack times are a different story.. but brickwall limiters have an attack time of 0, regardless.

Ok, I understand what you're saying I think lol

The DCX2496 being a crossover etc etc is multiband though right? eg output 1 = subs with it's crossover settings and brick wall limiting, output 2 = mid high with it's crossover and so on.
After all, a compressor becomes a limiter at high ratio's.


My earlier qoute then suggests different release times on the sub and mid/high outputs should be applied.

As far as being hard into the limiter, I have made myself watch the metering almost constantly when I'm pushing my system (not to the point of ignoring whatever else I need to do, just regular quick glances). I know you can't ingrain that into everyone, but wouldn't that be best practise?

Grant Bunter wrote: The DCX2496 being a crossover etc etc is multiband though right? eg output 1 = subs with it's crossover settings and brick wall limiting, output 2 = mid high with it's crossover and so on.
After all, a compressor becomes a limiter at high ratio's.

A multiband compressor operates on a full range signal by using bandpass filters. It's typically a mastering effect... very hard to use and apply correctly in a live situation. The closest application would be adjusting for the equal loudness curves.

My earlier qoute then suggests different release times on the sub and mid/high outputs should be applied.

Whatever sounds good, I suppose... the release time doesn't have much to do with stopping the signal. The shorter the release time, the better, I would think. Any release time at all is reducing signal that shouldn't be reduced.

As far as being hard into the limiter, I have made myself watch the metering almost constantly when I'm pushing my system (not to the point of ignoring whatever else I need to do, just regular quick glances). I know you can't ingrain that into everyone, but wouldn't that be best practise?

You can't always see your limiters, particularly in PA setups. Knowing where on your output meters the limiter engages helps. Having a real compressor with long attack and release set just below and placed just before the limiter in the signal chain helps too.

When you operate your own system, and you know how it responds, you can handle problems. When your system is under the control of someone else, it's best to have these things in check, filed under the header of "You Can't Fix Stupid."

Cheers Brent, thanks for your help.

Hope the discussion hasn't befuddled to many people...

Grant Bunter wrote:DJTrumptight: how did you decide that 20msec should be your setting?

If 407ms is the default factory setting and mine was set at 20ms then it must have been because of research on this forum.Interesting discussion,im sure alot of us can use this.Thanks for the thread,the study and the research Grant and Brent.

So, after all that lol.
You've been given enough information to make your own decision about what to do regarding release time.

I'll leave it up to the indiviual...

Time to stop being slack.

EQing using the DCX.

I will say right up front, this is not the only way to EQ a system!
It is a very simple overview that will get most people started.

For the purposes of this thread, I need to choose some cab examples.
So, I will use DR250's and T39's as those choices. Substitute your own cabs.

This works for both mono and stereo inputs. Obviously, with mono you only have to EQ one input, but for stereo, you will have to:

Press Input A, set EQ, then press input B and repeat.

Initially, you are trying to tune the system to relatively flat ie, no significant peaks or drops.
We are gong to "Mirror Image" the SPL charts to achieve that.
The goal with EQing is to try to cut frequencies whenever possible, rather than boost, sometimes boosting is the only option.

Step one is to go to your cabs SPL charts.
T39: http://billfitzmaurice.info/forum/viewt ... f=12&t=487
DR250: http://billfitzmaurice.info/forum/viewt ... =12&t=1099

Let' start off with the LF
Pick a chart that roughly equates to your cab size if there is one.
I pick 20" (my cab size), so look at the chart at the top of the T39 SPL chart page.

Print the chart.
With a ruler and pencil, draw a line across the page that is roughly halfway between the lowest point in the bandpass (bandpass in this case = your HP setting and your crossover point to tops/mains) and the highest.
In this case the line will be just below 100dB.

We are going to try and increase from 40 to 55 Hz and decrease 70 to 100Hz

Press The Channel A button, it will turn green.
Press PAGE --> twice.
Press PARAM --> until EQ is highlighted. Rotate large dial and turn EQ to ON
Press PARAM --> so NR is higlighted. If it's not there, use large dial to make it NR1.
Press PARAM --> so TYPE is highlighted. Rotate large dial to get BP
Press PARAM --> so FREQ is highlighted.
Turn the large dial until FREQ says 45-47Hz.
Press PARAM--> to highlight GAIN.
Turn the large dial to get gain to about 3.5db (this = the "deficit" below your pencil line)
Press PARAM --> to highlight Q
Turn the large dial to about 7.
You will see this makes the EQ curve a higher bump. The lower the Q, the wider the effect on frequency band. We only want affect a small width, so raise Q.

Press PARAM--> three times to highlight NR
Turn large dial to select 2.
Press PARAM--> twice to highlight frequency.
Turn large dial to select frequency of somewhere in the high 80's, say 88Hz.
Press PARAM-->
Set gain to - 3db (note the minus).
Press PARAM--> to highlight Q
Turn large dial on Q to about 5.

Now on your EQ, you see one bump up and one bump down. Your LF has been flattened.
You can mess about with these settings with cabs attached to see what changing things does.
Just remember to keep your gain increases as low as possible when "boosting" (making bumps that go up), but there is minimal concern with large decreases in gain when "cutting" (making bumps that go down).

HF next post...

HF
Again, print of your mains/tops chart and draw a pencil line across it with a ruler.
The first chart on the DR250 page is with ASD's, so is a bit sharper in HF from about 3KHz to 9 Khz.
The second chart looks more like my melded array, so I'll use that.

I will put my pencil line at about 105dB. You can see some peaks above the line, and some troughs below it as you go across from 100Hz to 20Khz. Note that about 800Hz to 2.5KHz is on my line!

So, we have to:
Bring down the bump seen from 100 to 220Hz
Try to fix the dip centred at 250Hz
Try to fix the bump centred at 500Hz
Try to fix the dip from 2.5Khz to about 12Khz
Not get to excited about the dip form 12Khz onwards.

This doesn't look to bad, we're going to pull down a couple of bumps about 2.5dB and raise a couple of dips about 2.5dB.
Have we got enough filters left?

Press PARAM--> until NR is highlighted.
Turn large dial until 3 is shown.
Press PARAM--> until FREQ is highlighted.

(handy hint: at this stage you can turn your large dial down to check how wide your previous 2 settings ie NR1 and NR2 are in terms of Q and how wide your filters are. They're not to bad as is, but you could go back to either and adjust, returning to NR3 to check)

Set your first HF filter.
Turn the large dial to about 200Hz
using the same steps as before:
Gain to 2.5
Q to about 8.5

Using the previous steps make NR4
Select frequency of 225Hz, can't be done, try 228Hz
Gain (minus) - 2.5
Q = 10

NR5
Select frequency at 5Khz, can't be done, try 4.98Khz
Gain increase to about 2
Leave Q at one.

I went to NR6 to check my NR 5 settings. Woohoo, there's more than 5 filters.
Turns out I had to modify NR5.
When I changed frequency to 5.31Khz and lowered Q to 0.8, it turns out my filter was from
2.35Khz to 12.8Khz, pretty close to the band I was aiming to adjust.

While there is at least one filter remaining, let's see what we can do about the fall off in frequency that starts to occur from 12K. Lot's off people can't hear above 15KHz, it's part of growing up, working with unprotected hearing blah blah blah.
If you're playing to young crowds though...

Select NR6 set frequency to say 16Khz, no good, 16.1 it is.
Increase gain 2.5dB
Q to 7.1

Go to NR7 woohoo.

You should have noticed that a new small bump has been introduced by using NR6.
As well as increasing from 13.5Khz to 18Khz, it has also caused an increase from 4.57Khz to 7Khz above and beyond the settings you made when you created NR5.

Right, that's a ballpark EQ.
If you're running stereo, go through NR 1,2,3,4,5,6. Write all the settings down.
Then press Channel B, go through the entire process of creating EQ in that channel.
Once finished Press Channel B again.
There should be no input/green lights on the input channels.

Many people say flat EQ sounds sterile. Hook up your amps/cabs, and have a good listen.
Play around with your filters until you get EQ that sounds pleasing to your ears

Don't forget, cut rather than boost, and, when you're done, SAVE it.

Once again, this is a guide only. It isn't the only way to do it. you could draw a pencil line in a different spot that meant you only cut and hardly even bother about boosting. It's all up to you.

I hope this thread helps somebody...

More about EQ

There are other ways to address dips. Especially in the subs.
You can set a narrower bandwidth from, say, 40Hz to to 75 Hz, using a LR 24 or 48 HP and a 12dB Butterworth LP at 75Hz, and take advantge of the rising reponse in the subs to cover up to your tops/mains crossover point.

You can then increase the gain on your subs output channel by say 6dB, BUT DOING THIS MEANS YOU NEED TO VERIFY THE SUBS/MAINS PHASE response at the crossover point by performing a frequency sweep through the crossover point.
You should do this everywhere anyway, but you can change phase if required, but youhave to know so that it can be done.

Other ideas:

Last year I was asked to do a gig where I had to play music both indoors and out, at low volume inside, but louder outside, for background music.
But I still needed to be able to have speeches at a higher level when required. That was just a matter of channel faders on my mixer.

No worries said I.

I used my 2 x DR250's and 1 x T39 indoors. And my two x POS fullrange 15" cabs outside.
Obviously, different response, no crossover required for the POS cabs etc etc. The POS cabs ran on a seperate amp.

What I did was;
Opened my MONO 45Hz setup
Pressed the Output 1 button and pressed PARAM --> until I got to highlight "X-Over Adjust Mode". Rotated large dial until I set the mode to "free". This unlinks outputs 1, 3 and 5.

Once done, I then set up Channel 3 output, with a 50Hz HP and a 20Khz low pass.
Then upped the output gain by 3dB to make it louder outside.

This also means you can also use the DCX as a multizone processor too. Technically, you could run three different subs/mains systems in mono (and taylor the HP LP etc) in three different rooms by using only one DCX.
Since the DCX also has output EQ cability, you could alter each system's EQ as well.
But if you do that, please check PHASE.

Grant Bunter wrote:This also means you can also use the DCX as a multizone processor too. Technically, you could run three different subs/mains systems in mono (and taylor the HP LP etc) in three different rooms by using only one DCX.
Since the DCX also has output EQ cability, you could alter each system's EQ as well.
But if you do that, please check PHASE.

Another idea is to use input C for monitoring - either DJ monitoring or live music. With all those outputs you can have several different monitor mixes. The reason why I use input C for monitor mix when DJing (rather than A+B sum in the DCX) is so that I can adjust the monitor volume from my mixer rather than having to reach down to the amp. Reaching for the amps is ok for me, but when someone else is on the system, they won't know how it's done. You could use an aux out from a mixing desk for live music. That 3rd input is a winner IMHO.

Following discussion in this thread:
http://billfitzmaurice.info/forum/viewt ... =4&t=21330

I need to qualify something.

This thread is based on a DCX with firmware version 1.17
Date code of the unit is 1007 (October 2007)

To check firmware version:

Powerup.
Press SETUP button.
Press <-- PAGE (page button on the left) to get to page 6
V x.xx will be in the top right corner.

Firmware grades are available, how high you can upgrade to depends on date of unit I believe.
For instructions:
http://www.icycolors.com/nu9n/deq2496.html

Note: Your computer needs to be midi able. And you need a USB to serial port cable (with it's driver(s) ) if you have a modern puter, or serial to serial port cable.

I believed there was a difference between my two units. ie one muted outputs on powerup, the other didn't.
Turns out, both have the same firmware version.
Because of checking:

Also on page 6 of SETUP:

Press PARAM --> until Mute Outputs on Powerup is highlighted.
Turn large dial to NO if you want that option.

I must have done this on my first unit already, but forgotten...

Checked the behringer site as well:
http://www.behringer.com/EN/Products/DCX2496.aspx

Ok, if your Firmware version is 1.16 or earlier (1.14 or earlier, 1.15) then the highest upgrade for those models is V1.16.

"Newer" models start with V1.17. At present, that is the latest firmware according to the site.