Whats going wrong? ( my vented sub measurement)

[Aucosticraft]

Hi,

I took measurements of my sub woofer but the graph looks odd. As shown in speakerworkshop the port graph measurement has to be lot sharper and higher than the cone, I am not getting it. also I measured the same with "creative" subwoofer. the graph looked very much like said in speakerworkshop. So I feel something is seriously wrong.

So pls tell what that graph tells you ? what must be wrong?

[Baldin]

Are the port and the woofer cone placed right next to each other?
One the Creative, was the port e.g. placed on the backside of the box and the woofer on the front?
Try to almost stick the mic into the port when measureing. And if the port and woofer is placed right next to eachother, then try to place a board or something between.

[wintermute]

Hi Acosticraft, what is box tuning freq supposed to be. You should have a peak at the box resonance frequency. Looking at your graph that would appear to be between 50-70Hz.

Have you done an impedance measurement of the driver in box? The mid way point between the two impedance peaks should be your box tuning frequency.

Also what type of box is this? Bandpass??

I only have my own subs measurements to go off (standard bass reflex), They do show more of a peak than yours, and the peak does correspond to where the tuning frequency should be (around 30Hz). The main thing that looks odd to me with your posted plots is that the rolloff of the port response doesn't seem steep enough. I would expect it to roll off sharply below the resonant frequency of the box (for a bass reflex). I have no experience with bandpass, but the response curve of the woofer looks like those I have seen modeled for bandpass.

In the attachment of my nearfield measurements red trace is the port, green trace is the woofer, and blue is the summed response after adjusting the port level for the port area.

Tony.

[Aucosticraft]

Originally Posted by Baldin

Are the port and the woofer cone placed right next to each other?
One the Creative, was the port e.g. placed on the backside of the box and the woofer on the front?
Try to almost stick the mic into the port when measureing. And if the port and woofer is placed right next to eachother, then try to place a board or something between.

The cone and port placement for both mine and creative are in 90 degs.
during meausrement the mic was places nearly 1 inch away from center od dust cap and precisely in the center of port. also the placement of mic was at where the port ends.

Originally Posted by wintermute

Hi Acosticraft, what is box tuning freq supposed to be. You should have a peak at the box resonance frequency. Looking at your graph that would appear to be between 50-70Hz.

Have you done an impedance measurement of the driver in box? The mid way point between the two impedance peaks should be your box tuning frequency.

Also what type of box is this? Bandpass??

I only have my own subs measurements to go off (standard bass reflex), They do show more of a peak than yours, and the peak does correspond to where the tuning frequency should be (around 30Hz). The main thing that looks odd to me with your posted plots is that the rolloff of the port response doesn't seem steep enough. I would expect it to roll off sharply below the resonant frequency of the box (for a bass reflex). I have no experience with bandpass, but the response curve of the woofer looks like those I have seen modeled for bandpass.

In the attachment of my nearfield measurements red trace is the port, green trace is the woofer, and blue is the summed response after adjusting the port level for the port area.

Tony.

The enclosure was designed using WinIsd. The port is tuned @ 36Hz, and is simple bass reflex.One thing I probably missed is, the output to the sub is from sub out of my amp. which has LPF @ 160Hz. but I think this should not be the problem. had done impedance measurement long before. but has received mic (WM61 from digikey) only two days ago.
I will do impedance measurements again and update here.

one more thing I would like to know. You have simply added port and cone measurements, is that right way of doing it? Speakerworks says you need cone piston area,port diameter to calculate result of the merger of two.

[wintermute]

one more thing I would like to know. You have simply added port and cone measurements, is that right way of doing it? Speakerworks says you need cone piston area,port diameter to calculate result of the merger of two.

I did originally and got an unexpectedly good result Terry set me straight with this post you use the formula with the SD of the driver and of the port, subtract the port result from the woofer result and you have the number of db you need to drop the port response before summing the two measurements

knowing about the 160Hz low pass makes the graph make more sense and no I don't think it would be causing an issue, but you could try measuring with the speaker running full range just to rule it out.

If your box tuning freq is supposed to be 36Hz then I definitely think there is something not right. After summing the responses how does the response look compared to the original sim?

If you play a 36Hz sine wave and watch the driver, it should look basically still, but there should be a lot of air coming out the port.

I'd definitely do the impedance measurement, just so you can check that the box tuning frequency is really what you think it is. If it is then there may be something wrong with the nearfield measurements.

Another thing, you should see a sharp dip in the woofers response at the box resonant frequency, I'm not seeing that on your graph.

Tony.

[Aucosticraft]

Here is WinIsd and actual measurement.

[Aucosticraft]

Originally Posted by wintermute

Another thing, you should see a sharp dip in the woofers response at the box resonant frequency, I'm not seeing that on your graph.

Tony.

exactly the other thing that is bothering me.

In my actual measurement the tunning frequency looks to be @ 39.55hz. but then too the graph should show its typical characteristic.

hope I am expecting right.

[wintermute]

OK it looks like it is pretty close then.. seems to be about 38-39Hz. I'm stumped for now... Maybe try doing the summing and compare it to the winisd predicted response. Maybe someone else can chip in.. My experience is limited to the one bass reflex enclosure that I have made, and other measurements I have seen posted. (here is an example from Claudio's site) http://www.claudionegro.com/sw/swaco...nearfield.html

Tony.

[Aucosticraft]

No problem wintermute, Thanks for the efforts.

[pinkmouse]

Just a thought, is the box stuffed too full for the ports to work effectively?