How to make quasi-anechoic speaker measurements/spinoramas with REW and VituixCAD

[mwmkravchenko]

Yes I do. I'm just starting with the off axis measurments and I was almost sure that nearfield measurments will not work here. I was just curious if the ground plane would be better or in the open space

Ground plane measurement is the best in many respects. Remember that you need to have 2 metres groundplane to equal one metre 2 pi simulation. Also having done many many ground plane measurements, the odd smallish structure in the area will not make that much difference. If you orient your loudspeaker properly it will make zero difference. I am interested in how deep a null you get in the response between the front and rear driver. There will be some point where this happens.

Mark

 

[BURNART]

Ground plane measurement is the best in many respects. Remember that you need to have 2 metres groundplane to equal one metre 2 pi simulation. Also having done many many ground plane measurements, the odd smallish structure in the area will not make that much difference. If you orient your loudspeaker properly it will make zero difference. I am interested in how deep a null you get in the response between the front and rear driver. There will be some point where this happens.

Mark

I'm also expecting nulls. There's not any spinorama of this kind of speaker that I know.
The next thing that bothers me now is that woofers should be in some angle that depends of the distance between woofer and the microphone. Then if I'm rotating the woofer to measure off axis should I keep the angle? Because when I will reach 180 degrees I would have woofer that is angled slighly up not down if that makes sense.

 

[mwmkravchenko]

I'm also expecting nulls. There's not any spinorama of this kind of speaker that I know.
The next thing that bothers me now is that woofers should be in some angle that depends of the distance between woofer and the microphone. Then if I'm rotating the woofer to measure off axis should I keep the angle? Because when I will reach 180 degrees I would have woofer that is angled slighly up not down if that makes sense.

The angle of the woofer is tiny in comparison to the wavelength of the operation bandwidth of this loudspeaker. Even if you use it to 250 hertz the wavelength i is a little more than 1.3 metres. Close to the driver air behaves more like a fluid than a gas, as you get farther away the pressure dissipates and is more difficult to influence by the angle of the driver.

If you want to compare, you can test the loudspeaker as it would sit in your room, and on it's side and measure to determine if there is a large difference in the polar pattern from either set of measurements.

When I do polar measurements I use a thin sheet of plywood and lay it on the ground as a smooth surface. I live in the countryside, so no pavement for me!

Mark

 

[BURNART]

The angle of the woofer is tiny in comparison to the wavelength of the operation bandwidth of this loudspeaker. Even if you use it to 250 hertz the wavelength i is a little more than 1.3 metres. Close to the driver air behaves more like a fluid than a gas, as you get farther away the pressure dissipates and is more difficult to influence by the angle of the driver.

If you want to compare, you can test the loudspeaker as it would sit in your room, and on it's side and measure to determine if there is a large difference in the polar pattern from either set of measurements.

When I do polar measurements I use a thin sheet of plywood and lay it on the ground as a smooth surface. I live in the countryside, so no pavement for me!

Mark

So it is okay if it would look like this?

 

[fluid]

So it is okay if it would look like this?

Just leave it flat. You have two woofers at different angles in the cabinet so the advice of tilting the cabinet down won't work anyway. That method is assuming you have a woofer on a flat baffle and are trying to make the measurement valid to a higher frequency.

 

[mwmkravchenko]

So it is okay if it would look like this?
View attachment 356419

When you look at some of the directions for doing a groundplane plane measurement you do have instruction to angle the device under test forward about 10 degrees.

 

[BURNART]

When you look at some of the directions for doing a groundplane plane measurement you do have instruction to angle the device under test forward about 10 degrees.

Sure, that is what I'm aware of, but when you look at my terrible drawing then it will not work for this kind if woofer I think.
Either or I will leave it flat all the time or I should at 0 angle measurement make it 10 degrees forward then at 90 make it flat then at 180 make it 10 degrees backward to make it symmetrical.

 

[bigjacko]

So it is okay if it would look like this?
View attachment 356419

When speaker is on the ground baffle will become twice as long. Raise the speaker up in the air is better if it is convenient for you. But then you will really need to point into mic at that point.

 

[mwmkravchenko]

When speaker is on the ground baffle will become twice as long. Raise the speaker up in the air is better if it is convenient for you. But then you will really need to point into mic at that point.

If you do this you need to incorporate a correction for floor bounce. Your measurements are no longer anechoic.

 

[bigjacko]

If you do this you need to incorporate a correction for floor bounce. Your measurements are no longer anechoic.

That is a defect for doing this way. For low frequency it is not a huge issue. Too much compromise in life....

 

[Ageve]

Thank you for the excellent guide. I know it's less accurate when the speaker is placed close to walls or other speakers, but these results are still quite good. My Revel C426Be stays within +/- 2dB up to 16 kHz. Sorry for the funny looking rugs. I used them to dampen the floor bounce.

I'm using REW V5.30.5 (mac) and a calibrated UMIK-2.

Receiver: Denon AVR-X5200W (multi ch stereo with front/surround power amps off).
Amplifier: Audiophonics MPA-M400ET (Purifi)
Source: AppleTV 4K (sweep generated in REW, 0-20000Hz 256k 5.5s mono).

No EQ or room correction used when measuring.

Compared to measurement by Audioholics:

edit: Another measurement, at 30cm (+/- 1.8 dB from 200 Hz to 16 kHz). Not much difference, but the small dip at 2.6 kHz is gone (caused by the room).

Revel PerformaBe F226Be Tower & C426Be Center Loudspeakers Review

We reviewed the Revel Performa F226Be, a three-way floor-standing speaker, and the C426Be, a monster three-way center speaker using four 6.5" bass drivers to see if they live up to the brand reputation.

www.audioholics.com

 

[Vilhjalmur]

Great write up! Wondering why a Tukey 0.01 window was chosen instead of e.g. Blackman-Harris 7 that would give greater frequency resolution?

 

[Dave Bullet]

Get rid of the soft furnishings when trying to do gated measurements. Soft furnishings can mask that first reflection, allowing errors to creep in. The harder the reflective surfaces, the better.

 

[Wickers]

I would like to ask what the correct measurement procedure is if two woofers share a box and there are also two reflex tubes.
The two woofers are connected in parallel. (Heco Victa 700 Ann.) I've been looking for a solution for weeks, but I just can't find it.

 

[MAB]

I would like to ask what the correct measurement procedure is if two woofers share a box and there are also two reflex tubes.
The two woofers are connected in parallel. (Heco Victa 700 Ann.) I've been looking for a solution for weeks, but I just can't find it.

Vituixcad documents this.
Measurement:

Locate woofer cone close to floor or wall to make half space conditions at LF. Measure near field response of one woofer cone at 5 mm from the center of dust cap. Measure at 5 mm from cone close to phase plug if the driver has phase plug. If two woofers have shared box, feed signal to both woofers and isolate (not brake) the other (which is not under test) with pillow to prevent midrange frequencies going to mic too much.
Measure near field response of reflex port(s) or passive radiator(s). Mic in the center of vent at baffle surface if vent is not rounded. If vent has rounding, penetrate few millimeters inside where tube with constant diameter begins.

Merging:
If you make individual driver and port measurmeents, load them individually and then merge them. If you assume each woofer, port, and radiator has the same response then include the count (2 identical woofers, 2 identical ports, etc...).

There are other ways to merge the data, I use Vituixcad so this is easiest for me.

 

[Wickers]

Merging:
If you make individual driver and port measurmeents, load them individually and then merge them. If you assume each woofer, port, and radiator has the same response then include the count (2 identical woofers, 2 identical ports, etc...).
There are other ways to merge the data, I use Vituixcad so this is easiest for me.

Thank you very much for your help.
I am also trying to use Vituixcad.
I think I pretty much understand that.
But one more question: What is the Delay on the right side of your picture for?

 

[MAB]

Thank you very much for your help.
I am also trying to use Vituixcad.
I think I pretty much understand that.
But one more question: What is the Delay on the right side of your picture for?

Great question. The picture is from Vituixcad's documentation!
The details are in the section on the Merger tool in the documentation.

VituixCAD help 2.0

Enter travel difference to active cone on the front panel at typical listening distance (2-4 m) on-axis to Delay [mm] field. For instance, delay is ca. panel width / 2 + box depth if the port/passive radiator locates in the middle of rear panel. Delay should normally be zero for active cones on the front panel.

It looks like the example in the documentation uses a 370mm deep cabinet.

 

[Entropy]

Any advice for taking measurements of a few different drivers? I'm not really concerned about the frequency response I get out of them, only about inter-measurement consistency (between multiple drivers being tested) and being able to implement the standard suite of tests in a proper fashion.

My biggest issue is mounting the drivers to the test rig. My first thought was to just build a basic plywood or MDF baffle or box and mount the drivers, but seeing that I may have 3-5 different drivers I'm concerned that I could damage or alter the baffle or mounting points and introduce unwanted variables to my measurements. Any takes on this?

 

[No. 5]

A DIY tetrahedral test chamber with swappable driver inserts, perhaps?

 

[HighFutility]

Hey, I have not read all the pages and posts, but I was wondering, with the weather turning for the worse, I wanted to do some measurements of my subwoofers indoors. Can I use the same method outlined here for nearfield bass measurements?