Home Theater Forum and Systems banner

101 - 120 of 187 Posts

·
Registered
Joined
·
502 Posts
OK, so I went a little wild today on making graphs. I used my own B2031P speakers and another's that just came in. Both are claimed to be matched pairs. Well, let's just see.
First "matched pair":
A) average response:

B) average response:


A)polar response:

B)polar response:


That's closer than any 2 I've built.

Next "matched pair"
C) avg response:

D) avg response:


C) polar response:

D) polar response:


Maybe not perfect, but you can tell they are making an earnest effort. Pretty impressive really.

Dan

Edit: looks like I may have screwed up the C and D avg response. That looks like the D avg twice.
 

·
Registered
Joined
·
502 Posts
Corrected avg for C:

and again for D:


Here's what one of these polar responses would look like with 1/3 octave smoothing like you'd see from one axis in a brochure:


Supposedly some say this more closely represents what we hear.:dontknow:

Dan
 

·
Registered
Joined
·
2,625 Posts
Supposedly some say this more closely represents what we hear.:dontknow:

Dan
Well your ears aren't linear devices they are spacial so how you perceive sound is spacial. Because LFE reflects more than HFE you will see an upward curve from the reinforcing reflections. This is why treating reflection points is often counter productive. I must say it's an impressive graph.
 

·
Registered
Joined
·
502 Posts
OK, so I took some shots of the innards today. I'll get some stuff to do modifications and see if it makes a measurable difference. I'm going to try and generate some plots using the new REW so we can get a better look at resonance issues. The crossover looks like a minimalist WGed 2-way with just a cap and resistor in the tweeter with a 2nd order low pass. That's why you can see the woofer's break up creep into the tweeter's range of the response. The active version should do much better in this regard even though there's not a huge problem in the passive version. We see poly caps, an iron core inductor, and a wirewound resistor. The baffle is stout on the woofer half, but thinner on the tweeter. Enclosure damping is certainly less than I've ever used. That may be where we could best improve the design.




Any last minute ideas?

Dan
 

·
Registered
Joined
·
502 Posts
OK, no suggestions so I went out on my own. I bought some of the pink stuff(well actually yellow) and beefed up the cabinet lining. So I'll post the results of response and impulse changes. First will be prior to stuffing, the second will be after the stuffing.







OK, nothing to really get excited about there. My guess is that the slight improvement in the treble has something to do with cabinet vibration or something. I was bummed by the results, but I also wanted to test my diffraction theory with the ports next to the tweeter. So I got some cotton balls from the wife's "stuff" drawer and filled the front of those ports with white cotton balls!:yikes: Well, the results are something to see:




No need for a side by side.:boxer:

Dan
 

·
Registered
Joined
·
502 Posts
OK, let's also look at the cotton stuffed ports graph 1/3 oct smoothed vs. the added stuffing material's graph 1/3 octave smoothed:


Can you tell which is which?

Dan
 

·
Registered
Joined
·
2,625 Posts
You aren't doing the right measurements for box resonance. You will want to make a waterfall graph to see what's really going on. Go -40db. For treatments you will want to use 2 or 4" wrapped rockwool.
 

·
Registered
Joined
·
502 Posts
I know, but I didn't do it in the first place so I had nothing to compare to. The next one's I'll do it all.

Here's what happened on the low end. One line is with no fiberglass stuffing added, one is with additional fiberglass, and then the one that's about a dB higher is with the cotton added. Not that in all actuality it is not louder, it was just so the different lines wouldn't get confusing.


Dan
 

·
Banned
Joined
·
3,807 Posts
Discussion Starter #110
Did you do them both or just one? You could do waterfalls with the untreated one vs. the treated.

Also, how did plugging the ports affect bass response? Got a picture of your cotton ball application?
 

·
Banned
Joined
·
3,807 Posts
Discussion Starter #112
Didn't notice that last graph went down to 20hz. Looks like the real story, from my perspective, is how plugging the ports with cotton seems to smooth and extend bass response.

Got a picture or technique writeup?
 

·
Banned
Joined
·
3,807 Posts
Discussion Starter #113
I know, but I didn't do it in the first place so I had nothing to compare to. The next one's I'll do it all.

Here's what happened on the low end. One line is with no fiberglass stuffing added, one is with additional fiberglass, and then the one that's about a dB higher is with the cotton added. Not that in all actuality it is not louder, it was just so the different lines wouldn't get confusing.


Dan
Noting your measurement times, why is there only 4 minutes between w/fiberglass stuffing, and without? Did it really only take you four minutes to complete the mod and take the measurements?

Also, did you complete the other mods as well, caulking the tweeter and sealing the woofer w/ superglue?
 

·
Registered
Joined
·
502 Posts
I only added the fiberglass and did no other mods until stuffing the ports with cotton. It only took a couple minutes thanks to the cordless drill and precut fiberglass I had since I had already taken the other apart and I made duplicates of every piece. I wish I had done the other mods, but I was in a hurry to get some measurements done before the next refrigerator/freezer cycle and just wanted to see what changes would happen with minimal difference in measuring conditions. I was thinking about using some dynamat or some other fancier vibration damping materials at a later date as well as the rockwool.

Check out the differences in the spectrographs before and after the cotton in the ports. Not that they tell us anything new, but a new look at an old solution. Sorry about the extra noise in the ones with cotton. As you can see from the impulse, the first reflection is a bit sooner.



 

·
Registered
Joined
·
502 Posts
I organized my photo bucket and that low end graph got moved so I'll post it again:


Oh yea, the cottoned ports have the more extended bass. All I did was grab some cotton balls and stuffed them in the face of the ports. There's like 6 or 8 in each and they're partly sticking out for ease of removal. Another 2 minute tweak.

There we go.

Dan
 

·
Banned
Joined
·
3,807 Posts
Discussion Starter #116
I still think the low-end is the more impressive thing here. We're looking at a 10db jump at 30db with no obvious side effects. Can you post pictures or details of what you did? Was it really just stuffing cotton balls in the ports? Anyone with speaker building knowledge that could comment on why the result was so drastic? Could you remeasure at some point in the future, both with and without cotton to make sure the change isn't a fluke/mic placement/speaker placement issue?
 

·
Registered
Joined
·
502 Posts
I was shocked by the lack of obvious side effects with the port stuffing. It really was just as simple as stuffing cotton in the front of the ports. I'f I do more measurements today I'll redo them just to make sure. Strangely to my ear it sounds like less bass, but the graph says more. I'll do them at the same level next time.

The diffraction reduction was the bigger difference to me. These things aren't exactly pumping out the bass no matter what. Of course for their size and all, they ain't bad there either.
 

·
Banned
Joined
·
3,807 Posts
Discussion Starter #119
red is with cotton and pink is without?

Seems to me that plugging the ports gives much better bass response, but I'm thinking that the opposite should be true. Can any speaker designers speak to this? Is it because it's just cotton, which probably isn't an air-tight seal?
 

·
Registered
Joined
·
502 Posts
OK, this is by the port:


But I had to turn the volume up to get the graph to work. My best guess is that the near the cone the port and the speaker are out of phase enough to cause the cancelation in the previous graphs.

Here's a graph of each 2 feet away. The stuffed ports are red, the blue is just straight ported.


Dan
 
101 - 120 of 187 Posts
Top