Project completed based on MathCAD sheets

[tda]

Hello guys,
I just finished my project based on MKJ Ported Box MathCAD Sheet.
Unfortunately, I measured the FR and is far from the SPL graph predicted by the FR measured.
I measured the nearfield response of the driver and port.
The main problem is the port response because it’s very weak; the peak it’s 15 DB less than the driver response; besides that, the peak is around 86Hz instead of 55 Hz, how the Sheet predicts it…I don’t understand why. Maybe the enclosure it’s not complete sealed…maybe the volume of the enclosure it’s too much for my small driver (4’’, SD=65cm^2)…I’m in trouble anyway.
If somebody have some clues, please let me know, it will help me a lot, since is my first project based on MKJ sheets. If it’s helpful, I will post also the measured FR graph and the Ported Box sheet data.

[Therac]

You could try to model the enclosure in WinISD and see if it gives the same results...

-Therac

[ashok]

Hi TDA,
Post the internal dimensions of the box , the speaker's TS parameters and vent dimensions . We could simulate it and see if it matches your simulation.
Cheers.

[tda]

T/S
Qts=0.570
Qms=2.490
Qes=0.740
Vas=4,44 l
Fs=82 Hz
Re=3.690
BL=3.67000 N/A
Sd=65 cm^2

Enclosure volume=27 l
Vent diameter=5 cm
Vent length=7 cm

I run the simulation in WinISD and the graph is almost the same as the MathCAD sheet…
What could be the problem?

[BobEllis]

Look at the predicted excursion. trying to get any significant LF output from a driver this small pushes it - you probably are well over the linear range and close to the mechanical limits.

A 5 liter sealed box ought to give you an F3 of around 100 hz. add a small sub and enjoy.

[tda]

---------------------------------------------------------------------------------
Look at the predicted excursion. trying to get any significant LF output from a driver this small pushes it - you probably are well over the linear range and close to the mechanical limits.
-----------------------------------------------------------------------------------
Dear BobEllis,
I don't think I understand what you are saying.
I'm looking at cone excursion in winISD, at 80 Hz I have 2 mm...it’s too much for a 4’’ driver ?
Can you reply a little bit longer and detailed answer please?
Thanks

[MJK]

Hi tda,

Did you measure the T/S parameters or are those manufacturer's numbers? Since the two simulations give approximately the same results, I am suspicious of the T/S parameters that you entered.

[tda]

I mesured myself the T/S parameters.
What exactly parameter looks wrong ?

[MJK]

Hi tda,

I am not sure any of the parameters look wrong, I was just asking if you had measured them. Using manufacturer's specs is one of the biggest sources of error when comparing computer simulations against as-built measured results.

You said that you are measuring near field for the driver and the port, how did you set the reference levels for each when you tried to compare the measurements against the analysis. Also, the MathCad simulations are set up to give far-field results and if you were measuring near field the results will be a little bit different. How close are your near field measurements? How about the locations of peaks and nulls, were they at similar frequencies as the the simulation predicted and just the magnitudes are incorrect? Or are the entire SPL curve shapes incorrect?

[tda]

Dear MJK,
These are the measured results and the Ported Box sheet data that I used.
I see 2 problems:
1.The port response peak is around 90 Hz instead of 55 Hz
2.Port response has a maximum dB at –4 dB, and the driver response is around +6 dB. That a means that the port count very little….The combined response it’s almost the same as the driver response. Both nearfield measurements are taken at the same output and input audio levels.

Thanks