So I recently built an 8" Coax based on the eminence beta 8CX and compression tweeter and its in a birch ply cabinet. Cabinet feels very strong and its about 22" tall, 9" deep and 12" wide. Its got only one pair of braces about halfway down like a cross which touches the side walls and the front / back walls. Knuckle knock test reveals a very dead solid sound. Woofer is mounted just above that cross which is about half way so its closer to the top portion of the cabinet. Its vented and the tuning is about 45Hz. Bottom firing. It will sit on stands which will not be covering the vent.
For sound absorption I've got a layer of 1" fiberglass lining all wood areas and 2" thick right behind the woofer area.
Sound is very good IMO but I can detect a slight bit of extra energy around 500Hz which sounds boxy. I'm thinking its cabinet sound because I didn't hear it when I had the woofer in an open baffle with the same crossover.
What are some things to try to reduce this?
Many thanks!
Tino.
For sound absorption I've got a layer of 1" fiberglass lining all wood areas and 2" thick right behind the woofer area.
Sound is very good IMO but I can detect a slight bit of extra energy around 500Hz which sounds boxy. I'm thinking its cabinet sound because I didn't hear it when I had the woofer in an open baffle with the same crossover.
What are some things to try to reduce this?
Many thanks!
Tino.
Try lightly stuffing the port with an old sock or WHY. Adjust the amount of material in the port to suit your taste.
Hi,
3" lining of one end of the long dimension is a good idea.
Lightly BAF stuffing half the cabinet away from the port also works.
rgds, sreten.
3" lining of one end of the long dimension is a good idea.
Lightly BAF stuffing half the cabinet away from the port also works.
rgds, sreten.
What's the size of the vent? Likely its resonating at higher frequencies.
It seems you have taken enough precautions in damping the enclosure internal volume. But one can only be sure by making some measurements. The enclosure walls can and will behave unexpectedly leaking too at higher frequencies (coincidence effect).
It seems you have taken enough precautions in damping the enclosure internal volume. But one can only be sure by making some measurements. The enclosure walls can and will behave unexpectedly leaking too at higher frequencies (coincidence effect).
Tino,
A mechanics stethescope will help you find where the cabinet is vibrating (i suspect that it is, sounds like you don't have sufficient bracing)
An externally hosted image should be here but it was not working when we last tested it.
A regular stethescope will also work but not quite as well.
While playing a variety of music, put the tip of the stethescope on various places on the cabinet. If the sound level drops as you remove the tip you have a cabinet vibration there.
dave
Thanks everyone! Interesting thing about the stethoscope! I see them on ebay / amazon for quite cheap so I might invest in one 🙂
Hmmm are you sure that's the case? the energy is down fairly low and the crossover was approx. 2Khz second order. Compared to the open baffle, the major difference in the curve is an extra couple db centered around 400-600ish so I suspect its something centered around 500...
"Boxy" sound is often cabinet resonance. Others have used phono carts placed in contact with the cabinet to identify cabinet panel resonances as well as the stethoscope. Using something with an electronic output can help you quantify the problem more than simple listening tests.
Bracing may or may not be the answer. Bracing will raise the resonant frequency, but also will raise the "Q" of that resonance. Applying damping material to the inside of the cabinet can help reduce the frequency and "Q".
Adding isolation to your co-ax to reduce the energy being transmitted to the cabinet may help, too.
Bracing may or may not be the answer. Bracing will raise the resonant frequency, but also will raise the "Q" of that resonance. Applying damping material to the inside of the cabinet can help reduce the frequency and "Q".
Adding isolation to your co-ax to reduce the energy being transmitted to the cabinet may help, too.
Hmmm good point jplesset. So basically a way of mounting it which would decouple the unit from the cabinet? I can think of something and maybe I can also try a little bit of dynamat just to see if it does anything (added to the walls). I don't have any but I can purchase a small amount to see if it helps a little. If it does, I can add a little more.
Not always a bad thing. Higher Q means less bandwidth to excite it, and higher frequency means less energy to excite it.
I have built many successful boxes where the goalis tomake any resonance potential high Q and high frequency so that they are unlikely to ever get excited.
We have a recent build where adding a layer of thick veneer pushed down the resonance frequency and decreased its Q and brought with it a serious issue with excited panel resonance.
dave
I've just been reading a little about thin wall cabinet design such as what Harbeth is doing. Sounds kind of interesting.
Foam weatherstripping from the hardware store is pretty inexpensive. Rubber grommets for the screws. Not too tight. That's how KEF did it back in the '70s. Worked well.
Dave, I'm not at all sure higher "Q" is a good thing. Yes, it may be excited less, but if excited, it will store more energy, and release that energy for longer. I'd rather the energy gets dissipated as quickly as possible, and that means low "Q", not high "Q".
jay
jay
When we got the KEFs from the early days of that, we found that we preferred them without the grommets.
dave
I have been building boxes now for sometime now with the goal of getting panel resonances up and hopefully their Q as well. With the energy content of music such as it is, it is very unlikely to have the situation where a cabinet wall is subjected to a long duration limited bandwidth signal of sufficient energy to charge the resonance. Also as frequencies rise, the relative thivkness of the sheet material increases and you get more inherent damping making it even harder to excite that resonance. Low Q resonances (Toole) are more audible as well.
dave
Dave, that's not exactly what Linkwitz found when he measured cabinet resonances. See his old article where he noted that at resonant frequencies the cabinet was as loud as the driver.
SB1980-3way
I certainly preferred the sound of the "R" series as designed.
SB1980-3way
I certainly preferred the sound of the "R" series as designed.
Page 2 on RHS, item #3 of this paper
http://www.artalabs.hr/papers/im-aaaa2007.pdf
states the following.
"3. The threshold level is proportional to Q-factor,
which means that resonances with a low Q are
more audible than with a high Q. Doubling the
resonance Q-factor raises the threshold of
resonances audibility for 3dB. "
Peter
http://www.artalabs.hr/papers/im-aaaa2007.pdf
states the following.
"3. The threshold level is proportional to Q-factor,
which means that resonances with a low Q are
more audible than with a high Q. Doubling the
resonance Q-factor raises the threshold of
resonances audibility for 3dB. "
Peter
If 22" and 12" dimensions are external, the internal dimensions translates to about 660 Hz full wave and 640 Hz 1/2 wave resonance in the box, respectively. Both are close to 500 Hz, are you sure the problem is in the vicinity of 500 Hz and not a little higher up?
Try to cover left and right sides and top and bottom with 2" or 3" fiberglass.
Do not stuff the port!
Second possibility for the 500 Hz "sting" is a strong higher-mode resonance from the port. Measure the output of the port - if you have REW or ARTA (any mic will be fine for that), or try some loudspeaker modelling software to see what is going on, or simply block the port and listen if the 500 Hz resonance is still there.
Wow I bet that's probably it. I didn't even think of any of that as an issue. I'll definitely stuff the tops and bottoms more and try again. I've just learned something! Thanks 🙂 I did stuff the ports and it was still there just as bad so I bet its that resonance you just quickly calculated!
- Status
- Not open for further replies.