ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [All]   Go Down

Author Topic: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic  (Read 13232 times)

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946

Part 1 -  Prep:

I'm performing the wedding ceremony in about a week on some new RCF TT22A tops, 905AS subs, and a Sabine 4802 processor at my home in Conroe Texas before I ship the system overseas.  I'm planning to use Smaart or REW for the measurements.

I'm guessing I need to do the initial sub crossover analysis outdoors, correct?  Driveway, open garage door, 20 feet down the driveway from the opening,....

How loud is loud enough?  I'll be in a somewhat sleepy neighborhood...  and I think they may be building on the lot next to my house which is going to cause problems....

I'm thinking maybe I can do the initial crossover/alignment for the subs indoors (I have a weird shaped (semi-open plan) house) then take 'em outdoors to fine tune and set up the Sub EQ.  If the TT22As are anywhere close to spec they shouldn't need any EQ.

The only measurement mic I currently have is an uncalibrated ECM8000.  I guess I could spring $399 for a Rational Acoustics calibrated Audix TM1 'Plus' (they seem to be perpetually out of stock on the $299 calibrated TM1) but I'm already about $4,000 over my planned budget so if the ECM8000 is good enough for the sub Xover setup and EQ I might live with it for now..

Logged
Mark McFarlane

Paul Tucci

  • SR Forums
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 371

Part 1 -  Prep:

I'm performing the wedding ceremony in about a week on some new RCF TT22A tops, 905AS subs, and a Sabine 4802 processor at my home in Conroe Texas before I ship the system overseas.  I'm planning to use Smaart or REW for the measurements.

I'm guessing I need to do the initial sub crossover analysis outdoors, correct?  Driveway, open garage door, 20 feet down the driveway from the opening,....

How loud is loud enough?  I'll be in a somewhat sleepy neighborhood...  and I think they may be building on the lot next to my house which is going to cause problems....

I'm thinking maybe I can do the initial crossover/alignment for the subs indoors (I have a weird shaped (semi-open plan) house) then take 'em outdoors to fine tune and set up the Sub EQ.  If the TT22As are anywhere close to spec they shouldn't need any EQ.

The only measurement mic I currently have is an uncalibrated ECM8000.  I guess I could spring $399 for a Rational Acoustics calibrated Audix TM1 'Plus' (they seem to be perpetually out of stock on the $299 calibrated TM1) but I'm already about $4,000 over my planned budget so if the ECM8000 is good enough for the sub Xover setup and EQ I might live with it for now..

Mark,

These adventures are always fun.
A couple comments, some of which may just be reminders, not knowing how far along this path you are by now.

Better to learn or reinforce one's measurement chops with a less than ideal mic than to wait til you have one.
In the end, you're still gonna override the measurement results with an ear. I would.

Your sub/main alignment is dependent on positioning of the boxes. If you have the luxury of repeatability for each gig you're golden.
Assuming your goal is to make the two boxes play nice together in the acoustic crossover area, you want similar behavior in the time and frequency domain so that when they add, they add together constructively. Acoustically, 1 + 1 = more than 2, or -alot.
You can get there inside of not-bothering-the-neighbors-volume. ( Langston may disagree:) )  I will typically tune in the arena while the lighting goofs are carrying on a conversation at FOH. 80 - 85dB is plenty.
What you need to know is the behavior of the low end of your main box, and then make the sub do the same in the crossover area. Most importantly, the sub's phase trace should line up with the main's phase trace in the crossover area. That is the key ingredient in making them play nice together. The vantage point of the measurement is key. One could use the auto delay finder, but realize that will make the top end of your mains the place in time you would be measuring from as evidenced by a relatively flat phase trace in that region. That's how you tell whatyour offset time is aligned to. It's also possible to choose a different vantage point. It's called empathy. If only politicians had some.  Choosing to make the crossover area (80 - 120 Hz ish) readout with a flat phase trace is easy. Manipulate the offset until that region settles as flat. This will destroy the top end of the measurement and make the phase trace wrap heavily but that's OK. We are focused on the behavior of the low end of the mains. FROM THIS VANTAGE POINT, meaning DO NOT CHANGE THE OFFSET, manipulate the arrival time of the subs with your processor to match the phase trace of the crossover area. This may include polarity reversal. That's the hard part.
I think you can do this part indoors. I would use a groundplane measurement on axis (that's where the people are).  Yes, the reflections from your interior will corrupt the measurement but the general phase trace will probably be visible. For the final EQ, definitely take it outside in a reflection free zone.

Yeeha!

PT
« Last Edit: May 06, 2012, 02:53:07 PM by Paul Tucci »
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9534
  • Atlanta GA

ONe thing that would add/or change to Paul comments is a big "it depends" on whether you delay the subs or the tops-to the other.

In some cases you would delay the tops to the subs.  This would be a case in which the sub signal gets to the mic later than the tops.  Could be a horn loaded or bandpass type sub-the low pass filter used or the physical location.

A good trick is to start by adding 10ms (or more depending-------) to both the tops and the subs.  Then "locate" one or the other (the one you "think" is getting to you first).  Now you can actually add or subtract time from either cabinet to get the alignment needed (phase wise).

When you get a good alignment-note the DIFFERENCE between the two delays.  Change the lowest delay time to 0, and enter the difference in the other cabinet (be it the sub or the main) delay time.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946

...
A good trick is to start by adding 10ms (or more depending-------) to both the tops and the subs.  Then "locate" one or the other (the one you "think" is getting to you first).  Now you can actually add or subtract time from either cabinet to get the alignment needed (phase wise).

When you get a good alignment-note the DIFFERENCE between the two delays.  Change the lowest delay time to 0, and enter the difference in the other cabinet (be it the sub or the main) delay time.

Nice trick Ivan to start both at 10ms, thanks
« Last Edit: May 06, 2012, 01:36:25 PM by Mark McFarlane »
Logged
Mark McFarlane

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946

...A couple comments, some of which may just be reminders, not knowing how far along this path you are by now.
...
...Your sub/main alignment is dependent on positioning of the boxes. If you have the luxury of repeatability for each gig you're golden.

I've read a bit and played a bit with an RTA, and I've worked as a geophysicist for a few decades and have written acoustic wave equation modeling and imaging software so I understand all the acoustic theory... I'm getting McCarthy's book,...

My current plan (at least for now)  is to forego clustered subs for a while (as I have used for the past 7 years) and either use a ground stack or tops over subs on adjustable poles.  To keep the relative sub-top front-side position constant I thought I'd make some 4" poles and use those short invisible poles when I ground stack. Seems like a simple way to keep the positing constant.  This is the setup I plan to use for testing, top sitting on sub with 4" pole just to align them.

My assumption is that the ideal starting delay, what I program in the speaker processor, will be for the top directly on the sub: from an on-axis 60 foot back audience perspective I don't think raising the top a few feet on a pole should make much of a difference in the time alignment.  A 6' pole would make a difference in sub/top arrival time in the front row (e.g. tops way over the users heads) compared to the back row, but that's probably better than blowing out the front row.  Whad ya gonna do??? 

I guess I should run the math on this, but I'm thinking the height of the sub from ground, and the top-to-sub distance, and the relative ears of the audience are fine tuning things done at the venue, or maybe they are negligible in the context of room reflections, power alley,...

I rarely have time at my normal venues to haul and plug everything in so I want a default setting that is 'as good as I can make it'.

Thanks for responding Paul.

« Last Edit: May 06, 2012, 01:38:21 PM by Mark McFarlane »
Logged
Mark McFarlane

Mac Kerr

  • Old enough to know better
  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 7552
  • Audio Plumber

My assumption is that the ideal starting delay, what I program in the speaker processor, will be for the top directly on the sub: from an on-axis 60 foot back audience perspective I don't think raising the top a few feet on a pole should make much of a difference in the time alignment.  A 6' pole would make a difference in sub/top arrival time in the front row (e.g. tops way over the users heads) compared to the back row, but that's probably better than blowing out the front row.  Whad ya gonna do???

Yes, a 6' pole (9' high speaker?) would make a difference between the very front and the back, but less so for a standing audience, and a seated audience would be so far out of the pattern of the mains that the possible loss of power at some point of the crossover region would be a negligible problem.

Assuming a 100Hz nominal crossover, the wavelength is about 10'. To get out of the summation zone you need a path length difference of more than 2.5'. If you are crossing over at a lower frequency (longer wavelength) the lengths involved increase.

You are on the right track.

Mac
Logged

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946

Step 2: Selecting the Crossover Point

I'm thinking about crossing over at 80Hz.  These speakers are plenty loud for my gigs and I'd like to push the crossover phase issues as low as possible,...  Then again,... is this too low?

Here's the specs on the TT22A, apparently flat to 70Hz.

A high-res version of the spec sheet is available here: http://www.rcf.it/fr_FR/c/document_library/get_file?p_l_id=249233&folderId=22490&name=DLFE-3110.pdf

« Last Edit: May 08, 2012, 01:46:07 PM by Mark McFarlane »
Logged
Mark McFarlane

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946

Here's the specs on the sub http://www.rcf.it/c/document_library/get_file?p_l_id=251487&folderId=22493&name=DLFE-4407.pdf. It looks like there is a high-cut filter applied to sub for this graph, I'm fairly sure the sub goes over 70 Hz, that's what the dealer told me...
Logged
Mark McFarlane

Paul Tucci

  • SR Forums
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 371

Here's the specs on the sub http://www.rcf.it/c/document_library/get_file?p_l_id=251487&folderId=22493&name=DLFE-4407.pdf. It looks like there is a high-cut filter applied to sub for this graph, I'm fairly sure the sub goes over 70 Hz, that's what the dealer told me...

Looks like a polarity switch and the user-selectable xover point of 80 or 120 is on the self powered box itself, nice.
Adding another xover in your processor for "safety" is the wrong thing to do. It will only add more phase shift.

PT
Logged

Paul Tucci

  • SR Forums
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 371
Logged

Paul Tucci

  • SR Forums
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 371
Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #10 on: May 06, 2012, 03:24:33 PM »

I guess I should run the math on this, but I'm thinking the height of the sub from ground, and the top-to-sub distance, and the relative ears of the audience are fine tuning things done at the venue, or maybe they are negligible in the context of room reflections, power alley,...


Out of curiousity I did the math.  If A squared + B squared still = C squared in a right triangle, the difference between a groundplane measurement at 60' away and a groundplane measurement at 60' away but with the main speaker raised 6'  would be neglible.

0 height                   squared       0
60 away on ground     squared       3600

Square root of 3600 = 60' path length



6' height                   squared        36
60 away on ground     squared      3600
                                                  3636

Square root of 3636                         60.3'

The difference in path length is .3' = 4".


At 100 Hz, one full wavelength is 1130/100 = 11.3' So we're talking a comparatively small difference between the two measurements.
How many of these differences are there in a total wavelength?    Approx 35.  4" * 35 = 140"     140" / 12" = 11.6' Close enough for this.
If you were predicting earthquakes I trust you were more scientifically exact than this.
Knowing there are 360 degrees of phase in a full rotation of a waveform we can see that 1/35th of a full rotation is a little more than 10 degrees of phase change.
There's enough slop factor to say your measurement with the top on the sub or raised up 6' vertically will be quite similar.

PT
« Last Edit: May 07, 2012, 06:34:29 AM by Paul Tucci »
Logged

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946
Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #11 on: May 06, 2012, 04:00:51 PM »

...There's enough slop factor to say your measurement with the top on the sub or raised up 6' vertically will be quite similar.

That was my suspicion for 60' in the crowd. 20' out from the stack I get slightly less than a 1 foot difference in path length, say around 30 degrees out of phase at 100Hz (napkin calc, actually in-my-head calc).  Not great but manageable.  Inside 20' is problematic but if the tops are really 6' above the front row's ears we have other problems in this hypothetical situation.

I'm not going to worry about it.  Thanks for doing the math.

Logged
Mark McFarlane

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946
Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #12 on: May 06, 2012, 04:27:54 PM »

Looks like a polarity switch and the user-selectable xover point of 80 or 120 is on the self powered box itself, nice.
Adding another xover in your processor for "safety" is the wrong thing to do. It will only add more phase shift.

PT

The 80-120 Xover on the sub appears to only control the signal sent to the OUTPUT jack on the sub, not the sub signal.  At least that is how I read the docs, but it doesn't really make much sense.  It appears the intended use is to allow you to send full range to the sub and daisy chain a high passed signal to the tops, without using an external crossover.

Page 4 of the full sub manual describes the switch function http://www.rcf.it/c/document_library/get_file?p_l_id=251487&folderId=22221&name=DLFE-921.pdf

The TT22A tops have an optional 24db/oct 90Hz high-pass filter which I did not plan to use. http://www.rcf.it/c/document_library/get_file?p_l_id=249233&folderId=22218&name=DLFE-914.pdf

Actually, this is making we now question the frequency response published for the sub.  The sub spec sheet says 40-120Hz response (without specifying if this is -10db or -3...) but the amplitude response on the spec sheet shows 120Hz as more than 40db down. The graph makes one think there is a high-cut filter applied to the sub signal (looks like -6db at 90Hz), but the documentation indicates the switch sets a high-pass only applied to the output. I guess when the speakers are delivered next week I can listen.

Hopefully these won't turn out to be 1-note subs.


Logged
Mark McFarlane

Doug Fowler

  • Member since May 1995, 2nd poster on original LAB, moderator on and off since 1997, now running TurboMOD v1.826
  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2331
  • Saint Louis, MO USA
Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #13 on: May 07, 2012, 01:33:33 PM »


You and your negative time.

I had an assistant on a tuning gig who made EQ changes and entered delay times.  Nothing was working as expected (for a good while, valuable time is being wasted).

I look over his shoulder and he is attempting to enter negative values in the "delay" field.

Reasoning: we need delay, so he entered negative time. 

Yeah, that will definitely affect your alignment job. 
Logged
Brawndo, the Thirst Mutilator. 
It's got electrolytes. 
It's got what plants crave.

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1946
Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #14 on: May 08, 2012, 01:50:28 PM »

Step 2: Selecting the Crossover Point

I'm thinking about crossing over at 80Hz.  These speakers are plenty loud for my gigs and I'd like to push the crossover phase issues as low as possible,...  Then again,... is this too low?

So what do people think about setting the crossover at 80Hz for these TT22A's?  Too low?

FWIW, I ordered a TM1 mic today, and I'm heading to the airport in 5 minutes to catch a plane to the States to pick up all my new toys.  Life is good.  My thanks to Mike Pyle for loading up a truckload for me.
Logged
Mark McFarlane

ProSoundWeb Community

Re: Case Study: setting up RCF TT22As & RCF 905AS with Sabine 4802, mic
« Reply #14 on: May 08, 2012, 01:50:28 PM »


Pages: 1 2 [All]   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.038 seconds with 24 queries.