My turn to build JC's DBB

[Rudy81]

My outside curve shows no dip at 300Hz. It was at 150Hz. I was hoping the new port arrangement might rid of that dip.....hmm

http://forums.klipsch.com/forums/storage/6/959349/dbb%2011.jpg

jc

JC, glad you are looking at this. I am posting some other runs with some interesting results. I'd really appreciate your ideas.....again.

[Rudy81]

Here is what the plots looked like before I separated the plots.

[Rudy81]

The interesting thing with this nearfield measurement is that you need to take a measurement of the woofer and one of the port to get a full picture of the bass bin performance.

This plot is of the port only. Same enclosure as the woofer above.

The green line is of the unmodified enclosure. The blue line is of the port with the pillow in it.

Pretty interesting to see how the port functions by itself.

[Rudy81]

This plot shows the result of the woofer and the port together with the pillow in the cabinet. Really interesting stuff how the port and woofer interact to reach the 40Hz designed by JC. The blue line is the port and the green is the woofer.

So, I need you 'experts' to tell me if you think I should run out tomorrow and get 4 pillows???

Also, I would have thought that adding a pillow would greatly affect performance by reducing the cabinet volume. But, based on the results the difference at the frequencies of interest ( 40-900Hz) there is only an improvement in that area.

At the higher frequencies there is a huge hit in performance, particularly approaching 2kHz. and maybe that's a good thing for a bass bin.

[Rudy81]

JC, I looked at your plots and I was able to reproduce that dip location by moving that 3/4" board around inside the enclosure. I originally tried the board, mimicking what I saw in some of the Klipsch bass bins at the factory. Of course, this was not a scientific trial, but it was easy to move the dip up or down along the frequency spectrum depending on how I placed the board...weird. However, the dip was always there. That is, until I put the pillow in there. I also used a wadded up blanket and had the same positive response.

Ideas on why this is happening? Should I go with the pillow idea?

Overall, the performance of the bass bin is terrific! I'm just tweaking the little things I found in my build. It seems this thing is going down to 35Hz just fine.

I love the sound paired with my Crown K1 amp.

[Rudy81]

I forgot to mention, there is no smoothing or modification to the raw plots in either the nearfield or outside measurements.

[DrWho]

At the higher frequencies there is a huge hit in performance, particularly approaching 2kHz. and maybe that's a good thing for a bass bin.

Why do you say that? Ideally, we only want the port to have output at the tuning frequency. Everything else you see is sound from the back of the woofer leaking out through the front...and then at 300Hz you're seeing the pipe resonance of the port. I'm actually surprised that the pipe resonance is so loud....I would seriously investigate ways of mitigating that.

As far as the pillow...adding more is going to raise the ESR of the rear cabinet capacitance which will widen its Q....which is often referenced as making the low frequency seem more extended. I personally prefer to look at the waterfall plots to see when the cabinet resonances go away. Maybe Mike Hurd could post some of his measurements that demonstrate it very well. The thing is, you really gotta measure outdoors to see it clearly.

Btw, does the dip at 1kHz-ish change if you move the mic up and down?

[Rudy81]

Mike, please translate your comments on the pillow issue for us novices. I gather you don't think this is the solution.

I see what you mean about the pipe resonance. The 'pipe' is a triangular port at the top corners of that enclose...in this case, the top one. It is interesting that the port 'hump' at the 300Hz seems to coincide with the dip in the woofer response.

Taking the speaker outside is out of the question. These bass bins make the Khorn bass bin seem light. This nearfield measurement system seems to do a decent job of measuring response when compared with the outside measurement I took just prior to lugging them upstairs.

So, in layman's terms, do I have a problem with the port? Would adding the pillow help or hurt? How do I proceed from here? Is this a big deal and worth the trouble?

Thanks for your help.

[jwc]

Bentz is far better at this than I. I appreciate you taking the time to do what you are doing. I wasn't aware of you placing "panels" w/in the cabinet and getting different responses. Interesting. However, I can see how stuffing of some kind can change internal reflections.

A few times I experimented with pillows in a chamber.....I would notice changes on a curve to a subtle degree. The thing I took away from that was how the bass sounded. I found no consistencies but I concluded the idea of using "stuffing" per se to a minumum. The bass wasn't as snappy the more I put in there.

jc

[Rudy81]

JC, your experience makes sense to me from a logic standpoint. Right now things sound very good and I was only messing with the panels and the pillow since I had a little free time. I was surprised to see that the pillow will 'remove' that 300 Hz dip in the woofer response, which I see as a positive. But, like everything else in acoustics, there is never a free lunch so I figure something else will change.

If the pillow will reduce overall output, no big deal, I have plenty of sensitivity left on the Crown amp. However, the pillow sure takes up a lot of internal volume.

I suspect that at certain frequencies, the angled rear walls reflect the soundwave right into the woofer cone.

The port plots are neat in that I can see how the port adds the lower response to the cabinet. What is perplexing is the hump in the port response that is around that same 300Hz.

I'm open to ideas and suggestions on how to proceed from here.

I guess I can try the pillows in all the cabinets and see how the sound is affected.

As far as Dr. Who's comment on the port issue, I have no idea on how to fix that. Ideas?

I'll play around with it again some more as soon as I get some suggestions on how to proceed.

[Rudy81]

One more thought. When I built the bins, I use 2" polyfill to line the cabinet opposing walls. However, I did not have enough material to line all the walls, so I alternated which walls got treatment. The others are just the BB plywood. I am wondering if I should start by adding polyfill to the naked walls and see if that takes care of the problem. I am just thinking the standing wave could be happening by the one back wall that is not treated. Thus, making the pillow thing unnecessary.

[Rudy81]

Spent some time running the tests over again. Results are consistant with what I posted.

- Making the port smaller increases the hump at 300Hz.

- The port with polyfill behind it responds almost exactly like the one without polyfill on the wall behind it, no real difference.

- The pillow most definately smoothes out the 250Hz dip in the woofer response.

- The big port dip after 1kHz does not change no matter what I do.

[Rudy81]

On the miniscule chance that the plots are being affected by my pc's cheap onboard audio card, I decided to set up the Tascam US-144MkII I used for the outside measurements on my HTPC. Ran the plot for the woofer with the pillow in it and got the exact same result without the dip. Here is the waterfall produced from that plot for the woofer.

[Rudy81]

Here is the waterfall for the port with the pillow in the cabinet.

[Rudy81]

Here are both plots with the Tascam sound card and the pillow installed. The port 300Hz hump is a strange one and I have no clue how to treat that issue. The woofer seems pretty nice out to 700Hz, then runs into a big dip. However, this confirms that with the pillow in the cabinet, the dip around 250Hz disappears.

Now I need you experts to tell what, if anything, I should do.

[Rudy81]

I decided to go back and re-read the aforementioned nearfield measurement paper and noticed that the published reading taken by Mr. Keele are actually rather similar to my results. His reading of the port measurement also has a big hump, although his is 100Hz higher. This makes me less concerned about trying to tweak the port at this point, since there is not much I can do to cabinet now. The only option would be to make the port smaller or seal the ports and open new ones along the back. There is no way to tell how the latter would work out. I know making the port smaller makes the hump more pronounced and reduces the low performance a bit.

I have yet to find any plots of other vented enclosures so I can compare, other than the one presented in the paper.

[Rudy81]

Maybe Mike Hurd could post some of his measurements that demonstrate it very well. The thing is, you really gotta measure outdoors to see it clearly.

Btw, does the dip at 1kHz-ish change if you move the mic up and down?

No, moving the mic does not change the 1kHz drop.

Here is a nice little discussion by Mike Hurd regarding pillow stuffing. This refers to his subwoofers, but I think the principles might remain valid for the DBB application. http://www.hometheatershack.com/forums/ported-subwoofer-build-projects/18349-tc-1000-sonosubs.html

I have decided that only trying things out will show results. I'm off to Walmart after dinner to pick up my "test" pillows. []

[DrWho]

Hey Rudy, it's a very busy week for me at work so I'm not ignoring you, but could you send me your measurement files so I can poke around your waterfalls a bit?

Also, what is the total length of the port in the box?

I actually have no clue what is causing the 300Hz content in the port. My gut feeling is that it is related to the shape/dimensions of the cabinet...kinda like the modal response of a room, but inside the speaker enclosure instead. The wavelength of 300Hz is 45", so one place to look is for dimensions that are even multiples of that dimension....so like 1/4 wavelength, 1/2 wavelength, 1 wavelength, 2 wavelengths, etc...

To put my comments about ESR of the rear volume capacitance into English....basically the rear volume and port have a damped resonance that requires energy to stimulate, and that's how the low frequency response is extended. Adding stuffing in the rear chamber turns some of that energy into heat, so the overall LF output will eventually start to decrease, but the rolloff will get shallower at the same time. PWK would always argue that no stuffing was the most efficient, but I feel the resulting standing waves/resonances offset that advantage. So like JC, I think just enough should be used and no more since that maximizes efficiency while still fixing the resonating problem.

And ya, that link to Hurd's subwoofer is exactly what I'm talking about....that's as close to textbook as you're gonna get.

[Rudy81]

Mike: No worries, and thank you for taking the time to answer. I'll send you the files tomorrow. I just finished adding one small pillow to each woofer bin, purely based on my test results. Here is a synopsis of what I did.

- One small pillow duplicated what I found yesterday. Same 300Hz dip appears with no pillow and it disappears with one pillow. The overall effect of the pillow was to get rid of the 300Hz dip and I noticed a slight boost in the port frequencies below 60Hz with a slightly wider extension to almost 35Hz. I also noticed that the rolloff above 60Hz was steeper than with no pillow.

- I also decided to try two pillows since I was in test mode. Two pillows also got rid of the 300Hz dip as well, but the overall response of the port took a noticeable hit, both above and below 60Hz.

I ran these tests several times in order to be confident of the results. I could very easily duplicate the 300Hz dip with no pillow and could get rid of it by adding one small pillow. The pillow effectively covers the parallel rear wall and the angled side walls. The strange thing is that the pillow boosts the response of the port below 60Hz and steepens the rolloff just a bit above that.

Regarding the big dip above 1kHz and the huge rolloff above 2kHz. I found that if I moved the mic farther away from the "nearfield" measurement position, the big dip around 1kHz filled in nicely and was 'flat' with the response below that. I also got much improved response above the 2kHz region, reaching toward 5kHz. So, I am supposing the Behringer Mic cannot be used for higher frequencies in that 'nearfield' position.

After placing one pillow in each bin, I spent a little time listening to some of my most familiar songs. I honestly cannot say that I could hear the difference, this is likely more of an academic exercise. However, I did note the bass output needed a little boost on my amp. It still sounds has good, quick, controlled bass. No boominess or objectionable sound.

I guess you could say I'm just tweaking the installation. I am very glad to have found a solution to that 300Hz dip.

[pzannucci]

Regarding the big dip above 1kHz and the huge rolloff above 2kHz. I found that if I moved the mic farther away from the "nearfield" measurement position, the big dip around 1kHz filled in nicely and was 'flat' with the response below that. I also got much improved response above the 2kHz region, reaching toward 5kHz. So, I am supposing the Behringer Mic cannot be used for higher frequencies in that 'nearfield' position.

How about the possibility of cancellation between the two woofers causing the dip. This is one of the reasons companies build 2.5 way systems.

Depending on mike positioning, there will be an overlap of the output of the woofers causing a dip depending on the distance from each of the woofers. This will likely get lower as distance gets farther away.