floor reflection what to do ?

[DanDan]

In case anyone is wondering. I continue these banters with SAC in order to keep him occupied. Keeps the litter contained.
Let's observe the lack of replies to the following questions.
So SAC, what exactly are you suggesting? A panel to mitigate the extra path?
What type of panel, exactly please. And where would you put it, exactly please, referring to the original poster's diagram?
Also please explain by what mechanism this panel would diminish the problem, again exactly please.
Finally please explain how, with any practical sized panel, say 2x4 feet, why a 110Hz wave front would simply not go around it?

DD

[SAC]

...This is simply an extension of his inability to understand the purpose and use of an ETC response as it provides only gain and time and no derivative frequency readout...

This parody of a discussion continues despite his questions having already been answered several times because he utterly fails to understand the time relationship responsible for the initial null as he is firmly stuck in the frequency domain.

You see, for him there is no other.

He continues to randomly confuse modal behavior with specular reflections as he fails to understand the nature of time based superposition and how it relates to comb filtering.

Thus he is out looking for a 110 Hz reflection!

But no one tell him that his illogic also renders the oft prescribed bass traps ineffectual as they are undersized relative to the wavelength of his modal pressure waves!

This is why it would be irresponsible to turn him loose with a more complete computer based measurement and analytical platform that encompasses acoustic behavior in the time domain, as he would sit despondently worrying about how so many functions of his platform are broken due to the measurements possessing only gain and time axes and lacking a frequency axis. Truly the source of software support nightmares.

Keep on clapping. And you might want to stop imposing on your imaginary friend to move the mirror around, as you utterly miss the point of damping first order reflections with anything short of bass traps.

But don't worry - it can all be blamed on that "high level maths" that unfairly require such advanced "high level" skills as multiplication and division. We can all appreciate your fear! 'Tis the stuff of which comedy is composed...

After all, one can only be expected to know so much!!!

Meanwhile, we'll just keep laughing.

[DanDan]

As I said

Let's observe the lack of replies to the following questions.

There we go. Not a single answer. More bile. Classic SAC, Foxfyr.

Also note the curious inexplicable outbursts of laughter.

What kind of funny is being an insulting bickering banned nuisance?

DD

[SAC]

Amazing. At first one was prone to think it a bit of an act as you became the embodiment of the Black Knight in Monty Python's The Holy Grail.

But what has become painfully apparent is that you really don't have a clue as to the physics of this behavior compounded by the fact that you don't realize it!

[DanDan]

The OP asked for help with his floor bounce problem.
foxfyr SAC blustered in, with his usual fanfare of vitriol and bile, but claiming to have a solution to the floor reflection issue. So I have asked him to share it with us.

So SAC, what exactly are you suggesting? A panel to mitigate the extra path?
What type of panel, exactly please. And where would you put it, exactly please, referring to the original poster's diagram?
Also please explain by what mechanism this panel would diminish the problem, again exactly please.
Finally please explain how, with any practical sized panel, say 2x4 feet, why a 110Hz wave front would simply not go around it?

He will not answer. He has no such solution. He made up this mythical floor bounce solution in his frenzied lust for bickering.

DD

[Dange]

SAC does have a valid point, although it is hidden amongst a lot of sniping with DD.

If the null is due to a reflection from the floor interfering with the direct sound from the speaker, i.e. a comb filtering effect. Then what is needed on the floor is a broadband absorber, not a bass trap capable of stopping 110Hz.
In this presumption all the sound is reflected by the floor not just 110Hz.

However 110Hz null due to reflection and then interference just doesn't seem right, too low in the frequency range. It's either a room mode effect or you're within the nearfield of the monitors

[DanDan]

Dange with respect, may I ask you exactly what you are suggesting?


What type of panel, material, thickness, and size? Where would you put it? By what physical mechanism will it diminish the level of the bounce and thus the comb filtering and the FR dip?

I am simply saying that I can see no possibility that a relatively 'broadband' absorber, on the floor or vertical, or even worn as a bib :-) could possibly stop or even significantly diminish a 110Hz wavefront.
That is just physically impossible.


DD

[Dange]

I can explain what I'm on about, I did have a clear answer but now on reflection (haha!) I'm not so sure...

I think SAC is saying that the broadband absorber will effect the phase at 110Hz, but not necessarily reduce its intensity by much. Changing the phase of the reflected wave will alter the interference seen at the receiver. I maybe completely wrong however

[DanDan]

Dange (rous) ground there

Here's a deal. If SAC is prepared to answer the simple questions, I am prepared to perform an experiment exactly by his specs.

Let's be very clear now.

SAC said

Assuming the null is caused by a reflection, you simply need a broadband absorber capable of mitigating a simple reflection delayed by 4.5 msec that travels an additional 5.1 feet further than the direct signal.
An initial ETC would have verified the existence of such a reflection, as well as a simple and quick verification of the precise path (for which a vertically mounted absorber would work just as well as a horizontal absorber laid precariously on the floor!)

I have a studio here, stand mounted ADAM S3A's. Geometry roughly equivalent to the original poster's diagram. I have readily available MicroTraps, Mini Traps and HF MiniTraps, I also have raw panels of 100KG Rockwool. All of these are approx 2 x 4 feet. I have FM2. This doesn't do ETC but generally reflections are totally identifiable in the Impulse Response graph. There is also a Log Squared view. Furthermore I can am am prepared to export the Impulse Response files, so that anyone can join in the analysis. Massenburg Preamps, ProTools HD192, DPA 4060. Ready to go.

So SAC, what panel or panels, mounted exactly where?

DD



W

[Glenn Kuras]

Actually DD check out the following lab numbers.

GIK Acoustic Panels are tested and certified.

The 244 or the Monster do just fine down in that range placed on the floor.

SAC may I recommend a bit less sword fighting in this area of GS? Just a idea.

[Dange]

Using 114cm for the height of the loudspeaker and the height of the receiver, then having 120cm between them, I've calculated this comb filter response. NOTE: this is based only one reflecting surface in a free field, i.e. would be the same as having the loudspeaker and receiver outside in the middle of a field, no reflections from walls, only reflection from the floor. Ignore the numbers on the magnitude scale. The wave is treated as a spherical wave, emitting from a point source

Very basic acoustics but gives a null at 124 Hz

[Dange]

Originally Posted by Mixary

the distance to my speaker is about 140 cm and the real ear level is 114 cm.
in this position the dip is at 135 hz. the dip is very close but very deep ( from 120 to 150 HZ and 15 db max at 135 hz.

140cm between them gives a null at 135hz

However there are too many variables for this to be that simple surely? Arrival times of reflections will answer the question as to whether it is a reflection from the floor

[RKrizman]

Originally Posted by DanDan

I am simply saying that I can see no possibility that a relatively 'broadband' absorber, on the floor or vertical, or even worn as a bib :-) could possibly stop or even significantly diminish a 110Hz wavefront.
That is just physically impossible.


DD

Very funny!

Okay, who is going to be the first one to market the Bib Absorber, Ethan or Glenn?

Btw, your post reeks of common sense in what seems to be a needlessly convoluted and vitriolic thread.

-R

[DanDan]

Thanks Glenn. However those lab tests involve very large areas of trapping.
10 Square Metres is the norm over this side of the pond.

My proposed test is real world, similar enough to the original poster's situation. It would deliver a result and we can make of it what we will. However SAC has claimed a solution to the original poster's 'problem'. I believe the 'solution' is both impractical (floor placement) and illusory (vertical placement).

Floor placement was also ridiculed by SAC

And on the floor? LOL! I suggested that a more prudent alternative would be to position a vertically mounted panel close to the signal source in order to intercept the source vector, just as is commonly done to mitigate work surface reflections with monitors mounted on pedestals behind the work surface.

I am quite prepared to try both floor and vertical locations, to SAC specs. He also suggested

My alternative suggestion was that a simple broadband absorber such as is normally used for any first order reflection is suitable. Maybe the friend you holds your mirror can explain that to you...But I doubt you will understand even then!

MicroTraps are often used for RFZ first order reflection control, however that was also ridiculed. I could quote again. However I am prepared to let that slide and go with thicker MiniTraps to give this test more than a fair chance.

I must also say, I have never seen any studio anywhere with such floor or vertical treatment, but perhaps they said that when the wheel was first introduced......

Irrespective of all this inconvenient reality, I am still prepared to do the test. Let's roll.

DD

[DanDan]

Dange, thanks for going to the trouble of doing those calcs and graphs.
I have pointed out in my Room Analysis Primer V2, specifically that the frequencies on some software graphs are not accurate at all. They are not even intended to be.

What got my attention here was the frequency of the dip changing with listener distance. I was initially wrong in my comment on that. On reflection :-) I reckon that changing frequency suggests that this dip is caused by the floor bounce.
We are of course ignoring the wall and the ceiling.
Could you take a look at the original posters diagram and calculations please.
If 110Hz is simply wrong then perhaps this is a wall or ceiling bounce. In that case a thick panel, with FRK on the back, facing an equal depth airgap, would be a very viable solution. Such a solution would have the double whammy of being part of the RFZ.

That would be a great result, a collaborative and helpful one, normal, untainted. The best side of GS activity.

DD

[LaLaFaV]

We had a similar issue, except our trouble frequency was around 300Hz. After putting broadband panels on the floor, under the console at the floor reflection point, it had no positive effect on our null. In fact, it seemed to make it worse. We did many measurements with the REW program. We tried panels in many different positions in the path of that reflection and nothing was a remedy to what we believe is actually a room mode issue.

We wanted to test to make sure, though. I'm glad we don't have to put anything down to cover our nice flooring or make some other ugly compromise. Even a rug made negative changes to the mix position response.

[Dange]

Originally Posted by DanDan

Could you take a look at the original posters diagram and calculations please.

120cm between receiver and loudspeaker. Receiver and loudspeaker both 124cm high

Null at 110Hz

I take back what I said earlier about it not being a reflection, looks like there's a chance it could be

[DanDan]

Dange, thanks again. So that, with the changing frequency, pretty much confirms that it is a floor reflection which is causing this dip, do you agree?
Sure an Impulse Response or ETC graph would confirm this, but it is hard to doubt a tape measure and calculator.
DD

[Glenn Kuras]

Thanks Glenn. However those lab tests involve very large areas of trapping.
10 Square Metres is the norm over this side of the pond.

????????????
The numbers that are listed are in sabins, so the unit will do very well to that frequency range. I do agee the more you put in the more sabins increase (absorption)

I must also say, I have never seen any studio anywhere with such floor or vertical treatment, but perhaps they said that when the wheel was first introduced......

A floor like that would last about a day in most studios. lol lol

[DanDan]

Marc are you still around? Can you confirm that the issue is really at 135Hz, not 110Hz. Did you try the Wall Bounce Calculators?
DD

[Glenn Kuras]

Originally Posted by RKrizman

Very funny!

Okay, who is going to be the first one to market the Bib Absorber, Ethan or Glenn?



-R

I'll let Ethan take that one.

[Glenn Kuras]

Originally Posted by DanDan

Marc are you still around? Can you confirm that the issue is really at 135Hz, not 110Hz. Did you try the Wall Bounce Calculators?
DD

Something tells me he ran for cover when the bombs started dropping.

[DanDan]

Glenn, perhaps in a Sixties style studio, you know, beanbags, flesh, hair..........
Sabines of course, understood. I am a bit uncomfortable with the relative sizes here though. A panel 2 x4 feet, thickness anything greater than say 2 inch, will surely kill HF and even MF bounce. However I kinda imagine a bass wavefront (apart from sailing right through it, minus the 0.1 or so absorption quoted from OC) will spill all around it. i.e. I am not fully convinced that Ray analogies are that appropriate here. Still rays do seem to deliver the dip centre frequency, which changes with distance. Speaking of which, doesn't SBIR change with distance? Marc come back, it's safe now. How far are your speakers from the front and side walls and the ceiling. Earlier numbers are unclear to me.
Best, DD

[Dange]

Originally Posted by DanDan

Dange, thanks again. So that, with the changing frequency, pretty much confirms that it is a floor reflection which is causing this dip, do you agree?
Sure an Impulse Response or ETC graph would confirm this, but it is hard to doubt a tape measure and calculator.
DD

There is another possiblity. Is it known if the test was carried out with one loudspeaker or both in the stereo pair? If it was both loudspeakers a similar combing effect would be observed, not necessarily at the same frequencies, but it's possible that the path lengths are similar.
So if it is seen with just one loudspeaker, then it seems to point more to a floor reflection.

The impulse response and the frequency response are basically two ends of the same thing. This problem neatly shows how. In the time domain you have signals arriving at different points in time, due to wave interference effects this results in filtering in the frequency domain. Simples

When looking at the impulse response, the accuracy of arrival times measured by software needs to be taken into account, as we are talking milli seconds

[DanDan]

Dange, I suspect that if both speakers were driven, the mic was most likely at the mid point. Perhaps Marc will return now that the discussion is civilised.
The Impulse Response is a given. Most of software we use takes an IR and derives the various graphs from that. Aside- IR reverbs are simply wonderful.
I have noted the fascination with ETC and have quizzed several of it's proponents. No-one has yet demonstrated how it shows reflections clearer, simpler, or better in any way, than the Impulse Response Graph itself. If I come across as ETC bashing, and I probably do, that would be an entirely wrong assumption. I take in general, a devil's advocate view of Measurement Software. Expectations are way overblown IMHO. People expect wonders from such tools. There is an expectation of answers to 'How is my room' 'Identify it's problems', and 'prescribe an exact course of treatment'
If you were to read my Room Analysis Primer V2, that subtext should be apparent. Obviously it's primary function is to get the reader up and going.

As ever I think many, many, of the extended features in both REW and FM could be usefully omitted. The auto room Eq in REW, the speaker design stuff in FM. Plenty to edit out IMHO. Both these wonderful products' real strength is in providing affordable and in FM's case easily useable Acoustic Analysis.
Point and shoot is just fine for comparative measurement. The rest is for pros. Sure we can ignore the extras, but initially it can all look equally useful, which leads to much confusion. I would strongly encourage both products' authors to consider Lite and Pro versions rather than continually adding complexity to the current products
DD

[Nordenstam]

Originally Posted by DanDan

I have noted the fascination with ETC and have quizzed several of it's proponents. No-one has yet demonstrated how it shows reflections clearer, simpler, or better in any way, than the Impulse Response Graph itself.

Here ya go!

ETC:


IR:



The difference is more pronounced in ARTA:
IR


ETC:


(Was unable to set the same scale in the pics and I still haven't found the magic button to set the zero time point in ARTA.)


If you start hunting for individual reflections, you'll probably find the ETC to be lots easier to use!

If you look at each reflection for reflection, you'll also see that it's not exactly the same in each picture. The level varies a bit between IR and ETC.

The ETC is more complete than the impulse response. The ETC's are the resultant sum of both magnitude and phase information. Impulse responses only shows magnitude. This may not change the results much in some situation, it may change it more in other situations. It depends on the circumstances. Why is this so important? To put it this way: without phase, equalizers don't do anything at all. As we all know, EQ's can change the level of sounds despite the fact that the magnitude of the sound is the same at both input and output. The only thing changing is the phase. The result of combining phase and magnitude is a change in overall experienced level.

Imagine an engine pulling a trolley along a straight railroad track. If the engine is pulling in the same direction as the tracks, all the pulling power is used to move the trolley forward. The pulling force is in phase. If the engine gets off the tracks and pulls at 45' angle, half the power is lost trying to pull the trolley to the side(where it can't go), so the trolley will move at half speed. If the pull is at 90' angle, nothing happens(given that the trolley don't fall off the tracks), despite the fact that the magnitude of the pull is constant for all these scenarios! If there are two engines pulling at +90 and -90 degrees, the magnitude of the force is twice as large, yet the trolley is not moving at all.

EDIT: don't actually know if it's correct to assume that the "i" part of the complex numbers are representing phase like described above. Sorry, haven't gotten that far yet! Think so, but I'm not sure about it. Hope SAC can clear this out. In math speak, ETC's gives the full set of the complex number information, while the impulse response only give the magnitude component (the value at the real number line).

[Dange]

What does ETC stand for?

[Mixary]

hey thank you for your replies,

i am sorry the last days i was very busy. i checked the calculator and it is very interesting, but on the other hand also tricky, because the side monitor distance is varying by the roof inclination.

when i type the shortest distance, the frequence plot shows a similar curve.
but this is not the hole truth, because the max. dip is 2 db in the calculator.
(not my 15 dbs )

i think my 15 db dip is a result of multipel causes.

i put 40 cm acustic glass wool between the monitors and the listener on the floor and all i can say, you should check it . the mids are much more clearer. floor refections are on topic .

on the other hand, why should they be off topic? a floor is hard reflective wall like the other ones.

tomorrow we will measure a second time and i will upload the plots,diagrams and pictures on wednesday. i promise you


do you want spezial measurements, diagrams ?

cheers

Marc

[Nordenstam]

Marc, it would be great if you can get the ETC measurement for the first 20 millisec or so.

Originally Posted by Dange

What does ETC stand for?

Energy time curve: AES E-Library: Using Basic Energy Time Curve (ETC) Measurements


I really need to purchase that Don Davis and Eugene Patronis sound system engineering book! Found a google book result that explains it a lot better: Sound system engineering - Google Books

In case the link doesn't work, here's a nice picture from that link:


Time domain is horisontal, frequency domain is vertical/left and the ETC is vertical/right.

[Dange]

Originally Posted by Lupo

I really need to purchase that Don Davis and Eugene Patronis sound system engineering book! Found a google book result that explains it a lot better: Sound system engineering - Google Books

Thank you for the links and info. That book does look good....

EDIT:
In fact I just bought it, Amazon make it all too easy