DO 18" Dayton Build and Vibration Question/Concern

This is my first DIY endeavor, I've built a dual opposed 18 inch setup, 22.5 x 24.75 x 26.75 dimensions (not including outer baffle), basically a Gorilla clone with slightly altered dimensions and bracing:



The subs sound good and measure well (close mic looks like I'd expect from a sealed driver). From 10 Hz up to 80-90 or so the box is perceptually inert, meaning I can't feel any resonance or vibration at all anywhere on the box. A smoke test didn't show any air leaks. I know the drivers are in-phase. I did test the drivers in free air and didn't observe anything that gave me concern. The box is stuffed with around 5.5 lbs of polyfill.





My question has to do with higher frequencies, around 100-110 Hz or so. I feel some vibration on the outer baffle (the driver is countersunk) and on the driver basket frame. As I said before, this isn't audible to me (it's not a "rattle" at all) and doesn't appear to show up as anything weird on the measurements, but it's troubling given that I can play LOUD at any of the lower frequencies and the box is dead as can be. The vibration doesn't extend to the sides or the tops, only the baffle and the basket. If I hold a screwdriver loosely into one of the screws you can then feel and hear the vibration against the screwdriver. The driver was mounted per a method I've seen highly recommended on the forum. I removed the pre-installed gasket, used adhesive to attach some 3/4 wide by 1/4 thick closed-cell foam tape to the mounting surface, dropped in the speaker, drilled holes and installed screws, then tightened them down in an alternating pattern. After they played awhile I checked to see if maybe I didn't get the screws tight as I had seen another member with this issue (though I believe his was described as a "rattle") and tightened them fully (probably another 1/8th of a turn or so on some of them) to the point where I'm afraid to go any tighter.

Has anyone ever encountered this before? As I said before, this is my first build, but my assumption is that this isn't "normal". The most perplexing part is that at low frequencies, even playing LOUD with the drivers at very high excursion I get nothing in the way of vibration but at higher frequencies with the drivers barely moving I do.
Edited by MrSmithers - 2/1/13 at 4:40pm

Thanks sir. I learned a lot going through it and owe basically everything I've learned thus far to this forum and especially to the help of a couple folks. Gorilla was essential with his feedback and advice. I agree that part of me says if it isn't audible and doesn't seem to show up in measurements to let it go, but if it's not supposed to be there then knowing it is there will bother me.
And that's part of why I posted. I get the idea of front to back bracing being of equal importance, but many of the recent builds (Gorilla's and similar) also include no front to back bracing and don't seem to be having or talking about this issue.

Bill, if I was going to try to add bracing, would you try to go all the way from panel to panel, or do you think tying the front panel to each nearest tic-tac-toe brace would be sufficient?

It's 3/4" MDF, yeah. I'll see what I can do about some panel-to-panel bracing, though I'll have to work around the existing bracing structure. I'm concerned with lining them up with the mounting bolts, I haven't had much luck screwing into endgrain, even with pre-drilling a hole.

The inner baffle mounting surface is only 3/4, it's not a double baffle. That's something I considered but perhaps I underestimated the benefit. Again, one of my biggest points of confusion is the fact that it's only happening at high frequencies and the fact that several guys, including Gorilla who has 4 boxes now, aren't reporting any issues. If the answer is a double baffle or simply panel to panel bracing in the other direction, that's fine and I'll go that way going forward, but I'd still like to understand the discrepancy.

Correct, the total baffle is 1.5", the mounting surface is only 3/4". Yeah, I can't speak to the tape, it's just what I'd seen Bossobass and N8DOGG and some others recommend. There was no crossover during this testing, but even crossing at 80Hz (which I imagine is likely) the signal won't be down enough by the problem frequencies to do anything significant.

Thanks sir, I did build it at work. When I asked the boss if I could use our space to build a "speaker box" I'm not sure if he knew what he was getting himself into... I haven't got it into the apartment yet, but the output is definitely for real and I have high hopes about the SQ.
Thank you for all your help. As someone who'd never turned on a router before I learned a ton and you were an amazing amount of help. I think I'll try to get some panel to panel bracing in there as Bill suggests and then I'll call it good. I do really like the tic-tac-toe bracing as it gives you two nice contact lines for both panels and you can make the "center square" large enough for the magnet to just seat into. It's more wood efficient if you use strips or whatever rather than cutting it out of a piece, but I wasn't that concerned about the added couple pounds and the wood usage wasn't a big deal.

Two things that I tried that I found really helpful:
One was to use an H brace to route my circles. I was concerned about the circle sliding or dropping, even with another piece behind it so I made an H out of a couple of scrap pieces. Then I could clamp my stock piece to the legs of the H and run a couple screws into the circle that was being cut out.


Probably 100% unnecessary but it gave me confidence and piece of mind while routing.

Another was, I was having a lot of trouble actually constructing the box and keeping everything square while I got all the pieces aligned and started putting clamps on. I used some of the scrap pieces I had and made a couple of these blocks:

and found them really useful. I just made them by screwing the longer piece into a base piece. From then on I used pieces of MDF to screw the other two pieces in square. Glue would probably be just as easy or better, but I didn't have time to wait for it to dry.

I think it'll fit in the Prius though I certainly will need help going up the steps. I will say that when you buys 3 sheets of MDF at Home Depot and tell them when they're ripping that you're putting it in a Prius they kind of give you a funny look until they see it done. I can go 34 wide by 64 long with the seats folded down though!

Also I had no idea what I was getting into with this CV 5000 amp. This thing is a monster, it must be made of cast iron. Here's a shot of the internals:

It pulls about 70 watts at idle, I saw someone asking that someplace. It's also pretty loud, I'm gonna look into working on that.

I will check the driver in free air again but I don't recall noticing anything out of the ordinary. This testing was done with the crossover at it's highest point (200Hz) as I was getting naked FR sweeps, I was running REW and using its signal generator to do sweeps and test at specific frequencies, all sine waves. I agree that with a 4th order LP the sub will be down a good bit by the "problem range" and mentioned as much, and so it should be even less of a problem than it is currently, which is likely "not much" especially with program content vs sine waves. It's more of a situation where I wanted to garner opinions on the "why" so that I could correct it either now or at least on future builds. The idea of driver design being the issue seems unlikely, as many members are using this driver by now (Gorilla has 8!).

The speaker wire I have cut just long enough to give me enough room to lift it out or wire it outside the box. Then of course I have it ran through polyfill and through the bracing. Is there a more positive way to make sure it doesn't do anything funny when installing the driver?

Appreciate all the feedback guys.