Just for the heck of it I decided to make a pair of speakers for the home using nothing but materials handy around the house. Box materials were salvaged from a previous projct (MDF). Drivers were a Boston Acoustics neo 4t tweeter and 5.X Pro series woofer (5 1/4"), both left over from a previous car audio application. I also used the original passive Boston x-over network to split signals between woof and tweet.
WinISD said flat response down to 55hz with a .329 ft3 box using one 1.56"dia x 3.02" port. What I got is flat response down to 85hz followed by an amazing sinkhole centered around 70hz (65-75), then natural roll-off from low 60's on down. This was without the aid of a subwoofer.
My guess is that there is a cancellation attributable to the room and my listening position. I tried moving the speakers around without much improvement. Just for grins I did a sweep with just the sub playing x'd over at 150hz and the same 70hz hole still existed.
So would it be silly of me to reconfigure the box so that it is tuned to be +4-5db at 70hz to compensate or would the cancellation overide even that attempt?
WinISD said flat response down to 55hz with a .329 ft3 box using one 1.56"dia x 3.02" port. What I got is flat response down to 85hz followed by an amazing sinkhole centered around 70hz (65-75), then natural roll-off from low 60's on down. This was without the aid of a subwoofer.
My guess is that there is a cancellation attributable to the room and my listening position. I tried moving the speakers around without much improvement. Just for grins I did a sweep with just the sub playing x'd over at 150hz and the same 70hz hole still existed.
So would it be silly of me to reconfigure the box so that it is tuned to be +4-5db at 70hz to compensate or would the cancellation overide even that attempt?
What I'd do is take a near-field measurement of the woofer and the port, do a scaled complex sum and see if the notch is a matter of your tuning. Ditto the impedance curve; getting an experimental impedance magnitude curve and comparing it to the predicted one will give you a clue as to whether your realization of the model is on-target.
I had initially thought that reproducing the hole at 70hz with the sub only, indicated that it was less a function of the driver/cabinet/port and more a function of the room/listening position, although I wasn't able to lose the hole by moving the speakers around.
I need to read up a little to learn about near field measuring, scaled complex sums, and impedance curves. These are foreign to me. Any suggestions for learning materials?
I need to read up a little to learn about near field measuring, scaled complex sums, and impedance curves. These are foreign to me. Any suggestions for learning materials?
Broken record here (I seem to be saying this a lot): no one building a loudspeaker should be without Dickason's Loudspeaker Design Cookbook. And it would be a very good idea to also have d'Appolito's Measuring Loudspeakers.
Both are available from a number of sources, including AudioXpress.
Both are available from a number of sources, including AudioXpress.
Thank for the heads-up. Time to invest in some resources before too much trial and error makes this a frustrating venture.
For grins I am going to try altering the configuration so that it's tuned to +5.25 db @70hz. It will only take 1/2 hour and I'm really curious. It's also easy to put back the way it was if I don't like it.
Thanks again.
For grins I am going to try altering the configuration so that it's tuned to +5.25 db @70hz. It will only take 1/2 hour and I'm really curious. It's also easy to put back the way it was if I don't like it.
Thanks again.
- Status
- Not open for further replies.