Jump to content

What's Going On


derrickdj1

Recommended Posts

On 12/2/2016 at 2:56 PM, derrickdj1 said:

Rob, you bring up a good point in asking about the filters used.  I never go beyond a three filter max when adjusting the subs.  Filters work on both sides of the target and the more used, the more undesirable IMHO.  With this setup, I am only using two filters on SW1 and one on SW 2.  A good raw FR is possible without using a lot of filters.  Also, the dip that normally occurs at the XO can be eliminated for the most part.

I coudn't agree more, although to clarify I think the amount of filtering may be even more important than how many filters.  Make several 2db adjustments and I would expect a great sound versus a couple of big adjustments say, 6 or 8db.  In that second case, the sound might be great at the MLP but would be odd or unpleasant at many other places around the room.  

 

 

Link to comment
Share on other sites

The main point is that many of us are also using auto EQ and when extra DSP and PEQ are used, it can alter the auto EQ.  There are 3 or 4 levels of Auddysee, and a few different version of MCACC.  These programs are doing hundreds to thousands of things.  My goal is not to undo to much of the auto EQ.  I do know MCACC corrects standing waves and to much EQ will mess that balance up.  If you need a lot of filters, the subs may needed to be re-positioned in the room.

 

EQ the subs to the room and other speakers and not to how straight the graph line is.  In one of the above diagrams the line is straight but, the system is out of phase.  I don't worry about how straight the graph is for the most part.  A range of +/- 5 is acceptable.  People shoot for +/- 3 trying to match the outdoor or anecohic measurement from 1 or 2 meters.  I use a tolerance of the +/- 5 since the system is vented and tuned below 20 Hz.

Link to comment
Share on other sites

1 hour ago, derrickdj1 said:

The main point is that many of us are also using auto EQ and when extra DSP and PEQ are used, it can alter the auto EQ.  There are 3 or 4 levels of Auddysee, and a few different version of MCACC.  These programs are doing hundreds to thousands of things.  My goal is not to undo to much of the auto EQ.  I do know MCACC corrects standing waves and to much EQ will mess that balance up.  If you need a lot of filters, the subs may needed to be re-positioned in the room.

 

EQ the subs to the room and other speakers and not to how straight the graph line is.  In one of the above diagrams the line is straight but, the system is out of phase.  I don't worry about how straight the graph is for the most part.  A range of +/- 5 is acceptable.  People shoot for +/- 3 trying to match the outdoor or anecohic measurement from 1 or 2 meters.  I use a tolerance of the +/- 5 since the system is vented and tuned below 20 Hz.

With 8 subs are you saying that you get a phase readout that has no dotted lines showing issues?  That would be impressive.  

Link to comment
Share on other sites

No, I did not say that, lol.  Phase changes with frequency so, nobody will have a perfect graph throughtout the 20 Hz -20 kHz.  Try to get the best phase relationship in the subwoofer passband will reduce ringing and boomy bass.  The bass problems wil occur in the modal region under 2-300 Hz.  Luckily, MCACC Pro has a full band phase correction for all the speakers when playing movies or music.

 

Actually the bad looking graph did not sound bad due to the phase correct function.  It is still bettter to get as many things corrected as possible.  I could use 2 ft. thick bass traps in the back of the room but, that is no going to happen.  The Bass decay is low enough in the room.  Now wall resonance is another thing, lol. 

 

A bass decay graph is also useful when looking for ringing and will get better if phase is made better.  An acceptable bass decay is less than 600 ms but, something under 400 ms is  better.  If you were to do a waterfall graph it would just be a 3D version of the bass decay 2D data.  Here is a current graph of the system with just the front subs on.  The second is with all 8 subs.  Near 20 Hz the decay is much longer and the need for more room Tx's that I'm not will to do at this point. 

12.4.2016 BASS DECAY.png

12.04.2016  all 8 subs.png

Link to comment
Share on other sites

I am coming in late and have read the OP and following ones. I must admit I am confused by actual graphs you are displaying. Some graphs have labels and some do not. Here and there through the thread I see references to the different color line meanings. Also, are all these measurements from a single LP and at the same exact height?

 

In the OP what are the graphs? All 8 subs only without mains or the first one with set to large and second set to small? What is the red line and what is the blue line? You didn't explain or label anything and my pea sized brain can't put it together.

 

In the 6th post: 1st graph small second large? Red lines are measured with mains and blue are the mains or the subs only?

 

If I had MCACC I would probably know what these graphs and lines mean. Please help me put this together in my head so I am not spending all my brain cells trying to figure out what the actual problem and question is. 

Link to comment
Share on other sites

The first two graphs are after auto EQ with MCACC.  The top graph is calibration with speakers as Large.  Speakers set to Large kills the low end as if no subwoofers are in the system.  The second graph is auto EQ with the mains set to small.  Now, the low end is way up and looks good but the 40-100 Hz area has a dramatic drop.  This is far from what is normally expected.  Setting the speakers to small should have fixed everything.

 

Further down on page 1 of the thread a graph of the subs only and a graph of just the mains was show.  The sub graph looks ok and declines as we move up in frequency or out of it's passband.  The graph of the mains looks ok and is dropping as frequencies go lower.  The mains were XO at 80 Hz.  It looks like if you combined the two graphs, things should result in a near flat FR.

FR Subs only.png

FR Mains.png

 

The  FR that resulted from combining mains set to small, XO at 80 Hz with the subs gave this:

FR run 2 mains plus subs MCACC.png

 

To get this result, there is cancellation and a phase problem since it did not exist in the Omnimic graphs of the subs only or the mains only.

Link to comment
Share on other sites

Since most of us running subs have some type of auto EQ, how do we fix it with what's at hand and that is fairly simple.

 

Four quick and easy things to do after autocalibration:

Do these 4 things one at and time and keep best result

  1. Change Mains from Large to small/ck FR

  2. Change XO up or down by no more than 20 Hz, Ex 60 Hz or 100 Hz for an 80 Hz XO, ck/FR

  3. Change sub/s phase/ck FR(this is almost never need if done correctly earlier)

  4. Change sub distance(never change more than 7 ft. or a 180 degree phase shift occurs)

 

 

A quick FR with just SW1 with the main and a FR with SW2 with the mains tells the the problem was with SW2.  Now, do we shorten or lengthen the distance.  The wide dip at the XO  points in the direction to shorten the distance.

 

Looking at the 4 quick things # 4, said never change the distance more than 7ft or 7 ms since this will result in a 180 phase shift.  The distance was shorten by 4 ft which is the same as 4 ms and that fixed the problem.  The 4 ms is close to a 90 phase problem that caused all the problems in the subwoofer passband.  Below is a last graph after the 4 ms correction. 

 

I know it's may be hard to follow but, the goal of the thread was to go thru my approach of solving this problem. The ideal was to have a diagnosis of the problem to work from.  In this case it was a phase problem.  Knowing this avoid going off into a bunch of directions, testing and measurements that would not have helped solve the problem.

 

10.02.216 FR no smoothing.png

Link to comment
Share on other sites

This is a problem with a system composed of different types of subs and roll-off all at different frequencies.  This is also a benefit since the system can be morphed several different ways for HT and music.  The different  roll-offs result in multiple phase shifts due to group delay.  I've been working with a mixed system for 5 or 6 years and have a good ideal of what's going on. 

 

The large LLT subs in the SW2 group of rear subs have the driver 90 degress located from the port.  The  fix was around 90 degrees, lol.  Having a way to measure FR with REW or Omnimic is all that is needed and a methodical method to solve what is going on in the room.  There was no need to do a room mode calculation, impulse response, waterfall graph, etc. to figure this out.  You don't have to be an REW or Ominimic expert.  One can quickly get lost in all the different measurements which is why some people don't use them.  A lot can be accomplished from a simple approach and a little thought.

Link to comment
Share on other sites

On 12/5/2016 at 9:36 AM, derrickdj1 said:

Since most of us running subs have some type of auto EQ, how do we fix it with what's at hand and that is fairly simple.

 

Four quick and easy things to do after autocalibration:

Do these 4 things one at and time and keep best result

  1. Change Mains from Large to small/ck FR

  2. Change XO up or down by no more than 20 Hz, Ex 60 Hz or 100 Hz for an 80 Hz XO, ck/FR

  3. Change sub/s phase/ck FR(this is almost never need if done correctly earlier)

  4. Change sub distance(never change more than 7 ft. or a 180 degree phase shift occurs)

 

 

A quick FR with just SW1 with the main and a FR with SW2 with the mains tells the the problem was with SW2.  Now, do we shorten or lengthen the distance.  The wide dip at the XO  points in the direction to shorten the distance.

 

Looking at the 4 quick things # 4, said never change the distance more than 7ft or 7 ms since this will result in a 180 phase shift.  The distance was shorten by 4 ft which is the same as 4 ms and that fixed the problem.  The 4 ms is close to a 90 phase problem that caused all the problems in the subwoofer passband.  Below is a last graph after the 4 ms correction. 

 

I know it's may be hard to follow but, the goal of the thread was to go thru my approach of solving this problem. The ideal was to have a diagnosis of the problem to work from.  In this case it was a phase problem.  Knowing this avoid going off into a bunch of directions, testing and measurements that would not have helped solve the problem.

 

10.02.216 FR no smoothing.png

Most of what you have explained to me, in person and in threads, has sunk in over time.  Like most wisdom, it is received, but rarily understood immediately.

 

Recently, through craigslist and an online purchase, I installed bass traps in 3 corners and have sound panels covering most of the theater front wall, and the reflection points including the ceiling.  A really amazing deal on CL helped this out, as I wasn't into spending many hundreds of dollars on these if I didn't have to.

 

Well, I thought about your phase discussion a lot.  I dwelled on it, thought about the fact that I have 4 monster subs in a room in which none were more than about 10 feet away from the listening position.

 

2 interesting things came from this thought process, and thinking about your phase discussion.

 

1) Tactile response, frequency response, phase and the graphs.  So in my quest for a straight line graph, what was missing exactly, and why is it that at least one seat of the three on the couch always had a bad graph?  I corrected with PEQ, I had tried many sub positions.  I had manipulated the delay on the different subs and no matter what I did, at the positions I used, it just wasn't coming into full focus.  So, I thought further and began moving them around again.   Out of the corners (a bit), moved one into the L-shape.   No matter what I did, I still had the one seat by the L out of whack -- or the seat on the other side.  But never all 3.  There was at least 1, if not 2 or 3 big dips in the response.  Even after I found what I was sure were the four best positions in the room - and you know what I'm talking about when I say I tested them all thoroughly....I just couldn't get them to respond with flat graphs.  SO.....

 

I dove into the phase, and considered the front and back subs as 2 groups, as your system allows (even though mine considers them 1 sub).   Since no delay was always optimal no matter what combination I tried, I tried delays as a group.  And putting a 3ms delay on the back subs.....brought all 3 seats into almost a perfect response!  And I checked the waterfall graphs (per your decay discussion) and saw that if anything, they were better this way.  Of course to find this, I tested delays on front and back from about 0 to 6ms.  At 3 in the back -- it is as good as I believe possible.  The amazing bonus is the tactile response.   I got it!  Yea lol!   What #cecaa850 was talking about in my thread...that there was a dip at the tactile response area.  Well the graphs were fixed, and the delays somehow brought it all into focus and BAM! I've got a mini-DDJ theater. :)  

 

 

2) Crossovers, the room, and making the best of it.    Even though I agree with everything you say about crossovers, one thing I have learned in my room is that the fronts simply cannot be crossed below 120hz and get the response graph (or close to it), that you would want.  I've tried every possible position and they are good, but not good enough in that range.  So in making the best of it, I cross at 120hz (and sometimes lower the center for dialog), and I am fine with it.  The only way to make them work better is rearranging the room, which I am not going to do!  

Link to comment
Share on other sites

Check out the latest DDJ and CEC....CEC I even have a mini peak at around 75 :)  

I personally love that they "peak" at around 15hz, in the inaudible range as the more tactile I can get there the better.

Notice with no smoothing that I have no dips or peaks in the bass region.  

All 3 main seats have similar response...but watching Transformers again today put a big :):):) on my face.

I'm finally in the game with the big kids lol.

 

 

3 main seats no smoothing 120916.jpg

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...