How would a cavity resonance manifest itself with the sound or freq response measurement? Would 7" driver in a 6" opening cause an anomaly at 13560/6= 2250hz or so? Or is it higher in frequency because the driver itself is in the middle of the cavity with the magnet/cone reflecting to the wall 2" away? I would imagine its higher up in frequency than the driver operates in a 2 or 3 way. Full range would likely have more issues.
I was just repeating the mantra of the designer.
I can visualize a conical opening on the back of a woofer projecting energy to the back wall in a different way than a mitered <to the back> baffle.
P
Here's the baffles after a bit more work. I've done the front and back panels separately, as I can't get any larger bits than 10mm chamfer and 12mm roundover for my el-cheapo 1/4" router. The back panel doesn't look as good as I want it to. It's currently rounded, while the front panels is chamfered.
Should I redo the back panel (could go for 12mm plywood if I did), to make the chamfer properly align with the chamfer on the front panels? Or should I simply do some manual chamfering work, maybe adding a bit of putty to smooth things out?
I recessed the screw points a bit at the back, as I expect I need to cover the back of the screws with something to smooth out (eg. bitumen, or just putty?). Should I have left the full depth for the screws to bite in?
I haven't rounded over the front panel top corners, as I expect I won't round over the top side edges of the boxes. Is rounding over recommended here as well?
Lastly, I didn't cut fully through the holes for the tweeters, as they don't need the full depth of the baffle to fit. My thinking was that this insulates the tweeters a bit from the vibrations of the main chamber. But the thickness of the MDF left (about 8mm) might be too little. Would I be better off just sawing through the speaker holes completely?
Thank you for any advice!
Attachments
Yikes! The whole point of a router is that the same thing happens ALL around the hole.
Normally, a "tang" makes sure the router bit traces a hole.
Don't "bear" on the tang means drag the router gently after making sure the depth setting has your cut in mind.
P
Which of the pictures are you referring to? I left the parts where the screws for the woofer sit un-routed / un-chamfered to allow them to bite properly in the MDF.
Last edited:
I assume you are thinking about single driver systems ( here in the multi-way thread), but I would argue I have never seen/heard one done well without some amount of filtering, be it passive in the speaker or active (preferred) in the amp chain. I have never seen a driver of any kind that does not need any help.
There are several old sayings that have wisdom:
Don't over think the small stuff
Fix the biggest problem first
If the OP wants to go further than just knocking the sharp edge off with a rasp, then the only correct way to proceed is to make a series of baffles with different methods and measure to the differences. Round-over, bevel, cove, flair, and so on. Which is more important, lack of edge of total airflow?
but it does have to be smooth.
Or should I say would it be smooth with a rasp?
On the baffle I'm using at present, I have used a 45degree chamfer still need to smooth the edges.
Next one I have a 28.6 dia 9.5 radius bit.
Last edited:
Found this relevant link;
Baffle Cutout Comparisons - Straight vs. Roundover vs. Chamfer Using Wideband Driver
It appears that chamfer is more optimal than rounding. I might just redo the back baffle panel in 9mm plywood and get it chamfered properly. Am suspecting that mdf/ply is a better sandwich than mdf/mdf in any case.
Baffle Cutout Comparisons - Straight vs. Roundover vs. Chamfer Using Wideband Driver
It appears that chamfer is more optimal than rounding. I might just redo the back baffle panel in 9mm plywood and get it chamfered properly. Am suspecting that mdf/ply is a better sandwich than mdf/mdf in any case.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.