AVS Forum banner
521 - 540 of 2509 Posts

·
Registered
Joined
·
55 Posts
I did loopback measurements from the NU6000DSP Output with several filters activated. The measurements were taken with a

  • Tascam US144 Sound interface, analog Output to
  • Behringer NU6000DSP input , Behringer Input attenuation pot at 50% ,
  • poweramp output over a "FH-103 High to Low Impedance Converter" (device to attenuate the voltage) to the
  • analog Input of the Tascam Sound Interface.
In REW I made a soundcard calibration that includes the impedance converter, which has a significant low frequency roll off.

The neat measurement with no DSP filter has a noticable low frequency roll off, I am not exactly sure if any of this results from the measurement chain but I did my best on the calibration file.

In the Pictures you see which filters were activated in the Behringer DSP and how they measured in REW. The attached zip is a MSO config to simulate those filters on the neat measurement.
So please, can someone with more knowledge check which filter Quality definitions Behringer uses in their Inuke DSPs?

Thanks
rumpeli
 

Attachments

·
Registered
Joined
·
1,864 Posts
Thanks for that.

I just noticed that MSO has no first-order HPF or LPF. I'm going to add those filter types so I can try to match the low-end roll-off of the iNuke amp as closely as possible with a first-order HPF. Once that's done, it should make figuring out the Q a little more accurate. It will take me a couple of days to get everything set up, as there's cool weather here and I'm using that to catch up on yard work.

Did you record what the Q1 and Q2 settings were in the iNuke? I don't need to know beforehand, but will eventually.
 

·
Registered
Joined
·
55 Posts
Did you record what the Q1 and Q2 settings were in the iNuke? .
;)
Q1 : Q = 1
Q2: Q = 2

Have a look at the screenshot of the Behringer DSP contol software. It shows all filters on Channel A. Filter #3 is active (botton highlighted inn orange) all others are bypassed.The graph shows the frequency response of the amp output signal with the activated filters. The bottom line has drop down selection fields for the filter type. One line above you can select the quality of PEQs.
You can see all 6 filter settings that I recorded the frequency response for. individually.

It would be good if you could check and confirm, if the Behringer Q definition indeed is the Q (RBJ) as listed in the MSO filter report.
 

·
Registered
Joined
·
1,034 Posts
I did loopback measurements from the NU6000DSP Output with several filters activated... The neat measurement with no DSP filter has a noticable low frequency roll off, I am not exactly sure if any of this results from the measurement chain but I did my best on the calibration file.


If you are interested a member documented a method to extend the low frequency response on the iNukes by adding several capacitors. I did it on the 6000 that I use for my Crowsons and it cost me $10 in parts and took about 30 mins. You can replace two caps and extend it somewhere around 10Hz or add 6 caps and get it down to 2-3Hz or so. I have photos of my modifications in the last 1-2 pages of my amp build thread, along with a link to the original thread that has a long how-to showing his signal measurements.
 

·
Registered
Joined
·
55 Posts
If you are interested a member documented a method to extend the low frequency response on the iNukes by adding several capacitors. I did it on the 6000 that I use for my Crowsons and it cost me $10 in parts and took about 30 mins. You can replace two caps and extend it somewhere around 10Hz or add 6 caps and get it down to 2-3Hz or so. I have photos of my modifications in the last 1-2 pages of my amp build thread, along with a link to the original thread that has a long how-to showing his signal measurements.
yeah, I found that too as I was searching for options to add pre outs to the amp. However, I see no reason to do this modification as the rolloff can be compensated with the dsp filters.
 

·
Registered
Joined
·
1,864 Posts
It would be good if you could check and confirm, if the Behringer Q definition indeed is the Q (RBJ) as listed in the MSO filter report.
Okay, so I did some tests without a first-order HPF to simulate the roll-off of the iNuke. These were the steps:

  1. Made four sub channels, each with one PEQ
  2. Each PEQ has a 100 Hz center frequency. Two have a 5 dB peak and the other two have a 5 dB cut
  3. Make four plots. Each plot contains the iNuke measurement and the response of the corresponding sub channel
  4. Adjust the trace offset of the PEQ response so its peak has the exact same value as the peak of the iNuke measurement
  5. Adjust the Q of the PEQ using the MSO tuning feature to match the traces as closely as possible
  6. Observe the MSO Q (Qmso) from the Properties window, compare it to the RBJ Q (Qrbj) from the filter report and the Q specified in the iNuke user interface

There are four plots. I'll just show one sample.



This is why I wanted to use a first-order HPF, in order to get a good match at low frequencies. Since this is just a "one or the other" determination, I thought I'd try it this way. I matched the data from a little below the PEQ peak frequency to the top end of the frequency band, to avoid the low-frequency errors. Here are the end results:

Q1, 5 dB boost
Qmso = 1.300 (from Properties window)
Qrbj = 0.974862 (from Filter Report window)

Q1, 5 dB cut
Qmso = 1.390
Qrbj = 1.04235

Q2, 5 dB boost
Qmso = 2.620
Qrbj = 1.96472

Q2, 5 dB cut
Qmso = 2.700
Qrbj = 2.02471

So it looks like the iNukes use the RBJ convention for Q, just as you said. I've attached the project below. BTW, the .msow file tells MSO which tabbed windows should be opened up when opening the project.
 

Attachments

·
Registered
Joined
·
1,864 Posts
For LS6 and LS12 filters use the low end corner frequency as shown in the main window or at the start of the filter description in the report. Don’t use LS6 (alt) or LS12 (alt) filters.
I just thought I'd mention a bit about the LF Shelf (Alt) filters. These were inspired by the Harman study on target curves. In that study, participants kept the "half-boost frequency" constant at 105 Hz and varied the LF boost to find the most preferred setting. If you had an LF shelf with a boost of, say, 6 dB, the "half-boost frequency" is the frequency at which the boost is 3 dB. The "Alt" variants of the shelving filters specify a center frequency, which is the same as the half-boost frequency. This allows you to vary the LF boost while keeping the half-boost frequency constant, which is useful for specifying target curves. For the normal LF Shelf filter, you specify the LF corner and boost. But if you keep the LF corner constant and vary the boost, the half-boost frequency varies, making them a PITA for specifying target curves.

You can use the "Alt" variants of the LF shelving filter just fine with the iNuke. Just go into the filter report, and it tells you the LF corner value. This is the number to enter into the iNuke software. To illustrate this, I've created a project having two LF shelving filters. One is an "Alt" LF shelf with a center frequency of 105 Hz and a boost of 6 dB. The other is a conventional LF shelf with a boost of 6 dB and an LF corner of 88.3465 Hz, obtained from the filter report of the "Alt" filter. They are the same filter, specified in two different ways, as can be seen in the graph.
 

Attachments

·
Registered
Joined
·
55 Posts
Q2, 5 dB boost
Qmso = 2.620
Qrbj = 1.96472

So it looks like the iNukes use the RBJ convention for Q, just as you said. I've attached the project below. .
First of all, it's easier to work with my measurements if you add -77.1 db gain to have a zero reference level. I should have done that in REW already.

I tried to mimic the Behringer high pass roll off in MSO by importing a txt file with all zero SPL and Phase as "flat measurement", creating a new sub channel with "flatt" added and applying second order highpassfilters to that one. Taking into account the min freqeuncy limit of 10 Hz I had the best result with a HPF variable Q 12 db/Oct at 11 Hz and Q = 0.73.

I also created a filter channel with the Q = 2, 5 db boost measurement and applied a PEQ at 100 Hz, 5 db cut , Q = 2.67 so that the filtered measurement matched the Behringer measurement with als DSP filters bypassed. The filter report states Q (RBJ) = 2.0022

?Unfortunately, at times I sill have trouble to transfer the filters as calculated by MSO into the Inuke DSP and I don't yet know why. Not always but sometimes, the small SPL graph in the DSP software user interface looks a little different from the filter channel gragh in MSO and more severely the check room response measurement really sucks. :confused: ?
Edit: maybe I found a relly stupid mistake I made, let me check...

cheers
rumpeli
 

·
Registered
Joined
·
1,736 Posts
I believe your fundamental assumptions are completely flawed. The idea of a "source/sink" setup has its origins in the Double Bass Array (DBA) concept. In such a system, there is an array of subs on the front wall, usually four in quantity, but maybe more, arranged in such a way that the wavefront from the front subs approximates a plane wave. This setup is duplicated at the back wall, with the back wall subs having inverted polarity and a delay corresponding to the depth of the room, in such a way as to cancel reflections from the approximate plane wave produced by the front sub array.

The expectation that having two subs in the front of the room and two in the back can be made to behave as a DBA is unrealistic. Reversing the polarity of the back subs is a recipe for bad time domain performance, even if the frequency domain can be made to work. If you want a DBA, make a DBA. It cannot be wished into existence.
Hi Andy, curious as to what you mean by "bad time domain performance"?

The source / sink method can be achieved with as few as one sub front and one sub rear. I've done it many times. Of course may not be as good as having more subs available but in my experience is one of the best methods of achieving good bass in a small room if you only have two subs available.

You can scale from there to two subs front, two rear (at floor level), the next step up would be two subs front, two rear at mid-height, then four front, four rear.
 

·
Registered
Joined
·
1,736 Posts
Non-minimum-phase behavior, for which flat magnitude response does not guarantee that time-domain performance will also be optimized.
Audible consequences of non-minimum phase behavior? Minimum phase would normally be associated with a narrow and sharp dip in the response which we generally hear through. Narrow dips are surely more preferable to room mode resonances and a non-flat response? The simple fact of the matter is that the source/sink method, even if not involving four (or eight) subs, generally gives you extremely consistent seat-to-seat response in a multi-row, multi-seat home theater, and does not require any aggressive, high Q EQ (which generates it's own phase shifts) in the region in which the source/sink array is working except for a shelving or low pass filter.

Have you experimented much with the source / sink approach?
 

·
Registered
Joined
·
1,864 Posts
Audible consequences of non-minimum phase behavior?
The original question was about measured data, not audible consequences.

Minimum phase would normally be associated with a narrow and sharp dip in the response which we generally hear through.
I think you meant "non-minumum phase" above. Mathematically speaking, non-minimum phase in an analog system means the transfer function has zeros in the right-half plane (e.g. an all-pass filter). This concept was originally introduced by Bode in his 1945 book Network Analysis and Feedback Amplifier Design.

The simple fact of the matter is that the source/sink method, even if not involving four (or eight) subs, generally gives you extremely consistent seat-to-seat response in a multi-row, multi-seat home theater, and does not require any aggressive, high Q EQ (which generates it's own phase shifts) in the region in which the source/sink array is working except for a shelving or low pass filter.
I normally like to distinguish fact from claims of fact.
 

·
Registered
Joined
·
1,736 Posts
The original question was about measured data, not audible consequences.







I think you meant "non-minumum phase" above. Mathematically speaking, non-minimum phase in an analog system means the transfer function has zeros in the right-half plane (e.g. an all-pass filter). This concept was originally introduced by Bode in his 1945 book Network Analysis and Feedback Amplifier Design.







I normally like to distinguish fact from claims of fact.


Sorry yes I meant non-minimum phase.

I'm not sure what you mean "claims of fact", I have plenty of data on source / sink from BEM simulations plus actual in room measurements. For sure some of those rooms have areas of the response with sharp dips and some non-minimum phase behavior (but critically only for some, not all seats), but source / sink (or it's more developed forms CABS etc) gives the best response for the room in terms of flatness of response and seat-to-seat consistency. Better than any other documented approach to multi sub optimization in a rectangular space with consistency of wall/ceiling construction.

I think you dismiss source / sink too easily. You should try it out even in its two sub form. It actually works very well, but it's quite hard to dial in the delay and level of the rear sub without multi-position measurements, so isn't for everyone.


Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
1,736 Posts
The original question was about measured data, not audible consequences.







I think you meant "non-minumum phase" above. Mathematically speaking, non-minimum phase in an analog system means the transfer function has zeros in the right-half plane (e.g. an all-pass filter). This concept was originally introduced by Bode in his 1945 book Network Analysis and Feedback Amplifier Design.







I normally like to distinguish fact from claims of fact.


Have you read the various published papers on source / sink incl those outlining two sub approaches?


Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
1,864 Posts
I'm not sure what you mean "claims of fact"
I define this as someone saying "X is true" / "X is a fact" without providing supporting evidence.

I have plenty of data on source / sink from BEM simulations plus actual in room measurements.
I'd be interested in seeing the data. (Really, I'm not trying to be a wiseguy here.)

I think you dismiss source / sink too easily. You should try it out even in its two sub form. It actually works very well, but it's quite hard to dial in the delay and level of the rear sub without multi-position measurements, so isn't for everyone.
Maybe so. I'll give it a try when I get the chance (which will be in a few weeks).
 

·
Registered
Joined
·
1,864 Posts
Have you read the various published papers on source / sink incl those outlining two sub approaches?
I have read the DBA papers, but it's been at least a year, maybe more.

It's worth revisiting the classic case of a plane wave tube in acoustics. But the discussion gets long, so I'll wait until tomorrow evening to discuss it further.
 

·
Registered
Joined
·
1,736 Posts
I define this as someone saying "X is true" / "X is a fact" without providing supporting evidence.
Funny you say that, because that's what it came across as to me when you basically told the poster that source / sink was terrible and he should avoid like the plague (or that's how it came across to me anyway). Hence my interjection in your thread. :)

PM me your email addy and I can send you over / put on dropbox some stuff to look at. Looking forward to your thoughts; always good to have a knowledgeable eye take a look at something.
 

·
Registered
Joined
·
15,015 Posts
PM me your email addy and I can send you over / put on dropbox some stuff to look at. Looking forward to your thoughts; always good to have a knowledgeable eye take a look at something.
Nyal, please post your data here so we all can benefit. Maybe we could even run some tests what solution MSO would have found in comparison.
 

·
Registered
Joined
·
1,736 Posts
Markus and Andy, you should read this paper if you haven't already: Fazenda et al., "Subjective Preference of Modal Control Methods in Listening Rooms", J. of the Audio Engineering Society 2012, S.338 http://www.aes.org/e-lib/browse.cfm?elib=16324. They did listening tests to evaluate the preference of different subwoofer placement and integration strategies, and source/sink (with a two sub setup) came out top.

Unfortunately I can say now that it's unlikely I'll have enough time to give you all the information I'm sure you will ask for. I have been rarely posting on the forums or writing my blog due to work and life being so busy.

The key benefit of source / sink over other approaches in rectangular rooms of consistent construction is independence from length mode related frequency response variations. Essentially with source / sink you don't have any length modes in the frequency range the sub array is operating. In a multi-row home theater (the main thing I am interested in, since these comprise the majority of the projects we do) this is extremely valuable, as all the other methods for integrating the rear sub(s) will result in some length mode variation. For example with zero delay on the rear sub(s) per Welti you will cancel odd order modes (1,3) but accentuating even order (2,4) modes. If you add some time delay to the rear sub(s) such as 2, 4 or 8ms or some other number then you can shift where the peaks and nips are (nodes and antinodes), and hopefully arrange them in such a way that there is more consistency in the response across each row. In larger rooms, where you have flexibility with row placement, then these latter two approaches can work well. However I've rarely found in practice that there is any flexibility with where the rows go. They go where they go because of other design elements: row-to-row distance requirements for reclining theater seats; distance from back row to wall behind (to get distance from ears to rear speakers); distance from front row to screen; depth of screen/baffle wall.


What you do have with a two sub setup (one front / one rear) is height and width mode variation, as you would expect. So you have height modes and you have width modes. If you have only one listener at one height then the two sub setup can actually work very well, as the Fazenda paper shows. I have data too I can post. Of course you would expect the response at seats which are along the width axis of the room to have different response, particularly if they the location of those seats starts to approach the null points for the 2nd width mode, because there is no width mode control in a two sub front / rear setup with the subs at midwall. If the room is wide enough, and the other seats on the row do not get near the null points for the 2nd width, then you can actually have very good consistency across the row. With a two sub setup (one front / one rear) you can of course move the front and/or rear sub away from the centerline of the room. This will result in some width mode cancellation but will make the response variation between seats on the same row worse (as you would expect for any non-symmetrical sub setup).

With four subs (two front / two back) at floor level the width mode variation is reduced by cancellation of the odd order modes. You can either place them in the corners or at the 25/75% of room width locations, with the expected typical differences between those two setups.

If you then take those four subs and move them to half height in the room you reduce interaction with problematic 1st axial height mode (unless your room is very tall higher order modes are not generally an issue). This approach is normally called CABS in the literature, though of course there seems to be some differences in use of the terminology, with one of the key researchers calling an eight sub setup as per the below CABS as well.

If you have eight subs, four front / four rear, then you have a DBA, and you get height mode cancellation as well.
 
521 - 540 of 2509 Posts
Top