Kit 2641 port length confusion.

kingjamez · « **on:** 24 Oct 2005, 02:22 am »

I just ran the numbers on http://www.psp-inc.com/ (port length calculator). From the drawings the Extremi have about 1.3 cu. ft. after you take drivers, port, and bracing into account. With a 3'' flared port you'd need a 15'' long port. However, the maximum port length can be only 12.5'' before it hits the back wall... and that leaves no room to breath for the port. Given the rule of one port diameter from the wall, the longest port you can get in there is about 9.5'' long, which results in a 37hz tuning frequency, well above the advertised 30 hz.

It's even worse for me as I built my 2641's 0.5'' wider as ok'd on this board. Well, I took off the extra volume from the depth and now can only get an 8.75'' port before I get closer than 3'' to the back wall. That puts me at about 38 hz tuning.

So what is the 2641 really tuned at? Are my calculations wrong?

-Jim

kfr01 · « **Reply #1 on:** 24 Oct 2005, 04:46 am »

You shouldn't have taken extra volume off the depth. The port is 10.5" long. Stuffing adds ~10%-15% effective volume. My guess is that actual tuning frequency is around 32-34hz. The practical difference? Maybe 2db less output at 30hz and below.

Is this a big deal? Probably not. Average room gain will look something like this: http://www.speakerbuilding.com/content/1020/rge.gif (from the book "High Performance Loudspeakers"). Even if the speaker was -6db down at 30hz in an anechoic chamber, in an average room it would be relatively flat (+/- 3db) until somewhere in the 20s. On most commercial CDs, even movies, there is little information below 30hz. I wouldn't worry much.

Kyle R. · « **Reply #2 on:** 24 Oct 2005, 02:10 pm »

Correct, the poly makes the drivers think they are in a slightly larger enclosure than they really are so that will affect apparent volume.

Kevin Haskins · « **Reply #3 on:** 24 Oct 2005, 03:45 pm »

As designed they are tuned at 30Hz.

The PSP calculations are a rough simple formula. As you note... the rear of the port exits close to the rear wall. What effect does this have? Simple... it extends the effective length of the port. How does it effect the functioning of the port? It acts as a longer port and the effective mouth of the port is LARGER than the effective mouth of the port alone. Port distortion is actually very close to that of the non-close to rear wall-opening method but we get a lower effective tuning frequency. The reason people tell you not to do it is because it cannot be accounted for easily in all the calculation models for determining port length. You have to do it empirically and that is what we have done. If you build and measure the impedance data from the 2641 you will find the middle of the impedance dip is right at 30Hz. If you run a high SPL sine wave to determine where port distortion starts to cause a problem you will find that it performs the same as a port with a standoff of 3" or more from the rear wall. How do I know? I measured it before I released the design.

In terms of the port length we cut them here for you so you don't have to do it yourself. Also, I think port tuning is critical. You will find a large difference between a 38Hz tuning and a 30Hz tuning. That is why we cut them here to minimize errors and builder error. If you build the cabinet per the plans with the right amount of stuffing you will get the right tuning frequency. If you don't build them per plans.... well you are on your own.

kfr01 · « **Reply #4 on:** 24 Oct 2005, 05:15 pm »

Well, that's about the best answer I could have hoped for. Sounds like a very smart design.

Kingjamez: don't despair, even if it means making new cuts - the end result is well worth it. The 2641 really does sound excellent. I am finding it more capable of reproducing live music than any other speaker I've demoed.

kingjamez · « **Reply #5 on:** 24 Oct 2005, 06:01 pm »

Thanks for the explanation, I'm not sure I follow that completely, but I'll think about it and try to figure out the port/wall thing. I agree that your graphs do well show a 30hz tuning though, so there must be something to that.

Making new cuts isn't an option for me... see pic below. Besides saying your SOL any suggestions for getting the correct tuning when the distance from the front of the baffle to the rear wall is 11.75''?

I see two options:

1. Make a port 0.75'' shorter than the one in the kit and live with the higher tuning.

2. Bending the Flared port by cutting a wedge out of, and then using a heat gun to bend the tubing to 45 degrees or so. Do you think if I went to a full 15'' port that way and got the mouth of the port away from the wall I'd still get good results?

I suppose worst case is I could try it both ways, the ports are pretty easy to change.

-Jim

Kevin Haskins · « **Reply #6 on:** 24 Oct 2005, 06:59 pm »

Yikes... you don't have enough depth to tune them right with a straight port. I've worked with a 90-degree angle and while it's possible it is a huge headache. You will just destroy the tube if you try and heat it.

I'd try using an overall length of 10.5" for the 3" flared port. That is how we cut them and send them with the kit anyway. That will put you closer to the rear wall than I'd normally like but the result will be a lower tuning frequency. I'd rather have it tuned at 27Hz than pushing it higher. The disadvantage of doing this is that you will have more potential for chuffing or port noise on the rear exit. It would only be audible at very high SPL levels so that is what I'd recommend. At high SPLs the port noise is masked to some degree by the content and other things in the room rattling anyway. I think you will be fine.

kingjamez · « **Reply #7 on:** 24 Oct 2005, 08:53 pm »

Great, that's what I'll do. Thanks for the help!! Lesson learned.... build to the plans!