correct way to measure SPL and Low end extension of PORTED sub

[audfrknaveen]

Hey guys,

What's the correct way to measure a ported sub?
Please see the attached image.

Black graph is at the driver.
Red graph is at the port.
(in both cases the mic is placed with in 1" distance from driver or port)

So we got 2 graphs. What next ?
What's the right way to combine both and get the overall SPL and Freq graph? Are there any correction factors to be added?

Is the method different when the port and driver are on the same side of the box Vs Different sides of the Box?

Any suggestions are greatly appreciated.

Thanks & Regards,
AudfrkNaveen

 

[bentoronto]

Is there some reason you haven't just put the mic where your head normally is? (Or three locations a foot apart and average them - REW does it.) If you are planning more study and EQ, a standing mic stand is a cheap investment.

B.

 

[Brian Steele]

One way I've heard about involves adjusting the port measurement until its roll-off at low frequencies (below Fb) matches that of the driver, then add the two together.

I typically use a "transfer function" method which is a bit complex to describe but it basically involves close-mike measuring the speaker with the vent sealed, measuring it again at 1M, using that to derive a transfer function, measuring the speaker with the vent unsealed at 1M, and then using the transfer function to remove the impact of the room from the measurement.

 

[MrKlinky]

Simply adding the close-mic'd port and cone spl at any given frequency will give a fair approximation of total acoustic output and will be similar to an anechoic response.
eg:
30Hz: 81dB + 103dB ~ 103dB
40Hz: 95dB + 103dB ~ 104dB
53Hz: 104dB + 104dB ~ 107dB

Measuring LF response at the listening point in the room will introduce huge anomalies due to the room response.

 

[stv]

Simply adding the close-mic'd port and cone spl at any given frequency will give a fair approximation of total acoustic output and will be similar to an anechoic response.

when adding nearfield responses i suppose the radiating surface should also be considered, or am I wrong?

 

[MrKlinky]

when adding nearfield responses i suppose the radiating surface should also be considered, or am I wrong?

If the port has has been well-designed this should not be necessary. If it is a tiny port of an odd shape, or overly long, then yes; even with an identical port area you could lose 10dB output over that of a properly designed one.

 

[mbrennwa]

Have you looked at the Mic in Box (MIB) technique?
Measuring Loudspeaker Low-Frequency Response | audioXpress

This is a very simple and reliable method for LF SPL measurements, and you don't need to worry about the correct way of adding near field measurements from the cone(s) and the port(s).

MATAA has a ready-made tool for the MIB technique.
GitHub - mbrennwa/mataa

 

[stv]

If the port has has been well-designed this should not be necessary. If it is a tiny port of an odd shape, or overly long, then yes; even with an identical port area you could lose 10dB output over that of a properly designed one.

not sure if I understand what you mean: for the measurement it should be irrelevant whether the port is well-designed or not, you just want to measure it correctly.

as "nearfield" I consider just some millimeters distant from port and loudspeaker. this merging method was (is) used with "speaker workshop" (see claudio negro's online manual, for example).

 

[mbrennwa]

when adding nearfield responses i suppose the radiating surface should also be considered

That's correct. Take a look at the audioxpress link in my previous post.

 

[bentoronto]

Simply adding the close-mic'd port and cone spl at any given frequency will give a fair approximation of total acoustic output and will be similar to an anechoic response.
eg:
30Hz: 81dB + 103dB ~ 103dB
40Hz: 95dB + 103dB ~ 104dB
53Hz: 104dB + 104dB ~ 107dB

Measuring LF response at the listening point in the room will introduce huge anomalies due to the room response.

I can't figure out what OP is looking to learn. Is this an academic examination of BRs or does OP want to know the results at their chair?

MrKlinky is in essence saying, "Just look at the FR plots and add them together." You might think this is simple to do visually using MrKlinky's table as a rough guide to dB-addition. That's assuming the plots for port and driver are not mutually contaminated.

Unfortunately, as I think Brian Steele was hinting at, you have wholesale variations in phases which - in the particularly degenerate case of BRs - is quite wild. Below the tuning, port and driver are in opposition and in mutual support above. In this age of DSP, BRs have outlived their usefulness with benefits running far behind issues, at least for audiophiles.

B.