Active DIY HT MTM Speakers

GV#27 wrote: »

Good idea getting that center to center spacing as close as possible.

I forgot to mention that the smallest circle is just the schematic port location (not sure if that is the best location or not - but it would effectively reduce the baffle at the tweeter). I spaced the woofers by a common formula, with an 1800 Hz xover, then squeezed the tweeter in tight. The tweeter would be flush, and the woofers would be surface. I could actually move the tweeter closer to the center, but I'd be eating up more of the tweeter flange under the woofers.

I've seen so many MTM designs where they seem to ignore the woofer spacing crossover rules. Are they going for aesthetics? Build simplicity?

GV#27 wrote: »

I used a similar baffle layout for an MTM center channel(now replaced by an active 3way).The small flanged SEAS that I used (H1397)allowed for close spacing of the 2 6.5" mid/woofers.Center to center spacing is even more important for an MTM positioned horizontally.The crossover was at 2k.
With your tweeters low resonance 1.8k with 24db slopes should be no problem.

I would be tempted to place the port on the rear panel,then any chuffing noises will not be heard.

Aswell as the delay you are adding for the tweeter,you might need to add some baffle step compensation so that the speaker does not sound lean in the lower mids/ upper bass.It will require modding your Rane xover .I can provide more info on this if you would like.

Do you have access to schematics for the Rane?

I do have the schematics. Rane is really good about that sort of thing. How, generally, is bs compensation different from equalization?

You know, I keep reading and reading, and it seems that BSC in a design ends up being kind of a dart thrown with a blindfold on. The speaker's in-room response is just so unpredictable, even the models for predicting based on the baffle itself and tweeter are pretty close to guesswork.

So it seems that the better thing to do is first, limit the baffle step in the first place, by cabinet design, and second, adjust once it is all in place, if even necessary. For a nice two-channel setup, making the baffle around the tweeter really, really tiny (like B&W's top-mounted tweeters) would seem to be key, as well as having nice tapered edges on the baffle. If done well, it seems the need for BSC would be eliminated.

In the case of HT speakers such as those I am planning, where the box is much simpler (gotta build a slew of them, after all, and they need to be mounted simply), minimize the baffle step with a narrow baffle, off-center spacing, filleted edges, etc., and then EQ out the rest. I'll be using Rane's THX eq's anyway, and since we are only talking about a few dB, the magnitude and center frequencies of which won't be known very accurately until the speakers are complete and mounted, it just makes sense to fix it last.

I could be wrong, of course. Feel free to tell me so. Really.

Revised baffle attached. The size of the Precision Port flare for a 2" port requires placement at the bottom (along with the principle of leaving space beside the port equal or greater than the diameter of the port). So a 20" cab with .75" walls just barely contains everything properly. Tuning will be 40-50 Hz, well below the sub xover of 80 Hz. Between that and the flares, port noise should not be an issue at all.

Alternate baffle design attached. I've adjusted the cabinet dimensional ratios to be golden, which only slightly widened it. The alternate design would use a .5" layer of mdf as an additional layer on the baffle underneath, which would offset the drivers so that sound-absorptive material could be placed in the hatched area. Besides the treble that reflected off the woofer cones themselves, that should eliminate the baffle-step problem entirely, right?

GV#27 wrote: »

Your new design may well reduce high frequency reflections and improve the frequency response in the treble range.However baffle step, (or as Joe D'Appolito calls it diffraction loss) is a lower midrange/ upper bass issue.
It will still occur with your new design as it relates to baffle width. (The wider the baffle the lower the frequency where the step will begin).Because of their longer wavelengths, at progressively lower frequencies the sound waves wrap around the baffle instead of being reflected by it.So upper mids will get some reinforcement from baffle reflection's while the lower mids and bass do not causing an imbalance.

My understanding is that all wavelengths less than the distance of the driver to the baffle edge will be reinforced, and will be operating in 2 pi space. The wavelength of 1800 Hz (xover point) is 7.5". The cabinet is 8.625" wide. Half that is 4.3125, which corresponds to a wavelength of 3100 hz. Horizontally, that should be the floor of frequencies reinforced by the baffle.

What I don't know is if the woofers themselves are counted as baffle width to the tweeter, that is, as the sound travels from the tweeter more or less vertically, the distance it crosses over the woofers would include a wider range of frequencies that would be reflected off the woofer cones and reinforced.

GV#27 wrote: »

You would use the full width 8.625" to calculate were the step starts.Aswell the transition from half space to full space is gradual as frequency decreases.


The woofers are definately reflective surfaces for the tweeters output albeit moving ones.This causes doppler or FM (frequency modulation)distortion of variing degrees.

What if I coated the woofer cones in velvet? That would be sweet!