[guest60106]

This post is an update to my previous post “New to OB and Design Thoughts”. I thought I would start a new thread here and maybe spark some new interest.

After much research and sifting through many OB threads and many white papers I feel a bit better about moving forward with my first OB build. Hopefully I have come up with a fairly straight forward “final design”.

I would like to thank everyone (some I have discussed topics with and many I have not) on this sight for all the help. As people look this design over I hope it will spark some questions, input and discussion. From this point forward, I have committed to this basic concept for more than one reason but for the most part it is time to get something built. I have all of the drivers, materials and tools in my possession so I won’t be buying much.

Design.

The design is a purely open back design with wings on the base section at 90 degrees to the rear. The mid/high panel is ¼” thick 5052 aluminum mounted on McMaster-Carr rubber isolation mounts. I did this for two reasons. The first is obviously to minimize the coupling of the bass section. The other is to make driver/panel design changes easier. Just cut different size or shape panels, add the new drivers and install the new panel. The bass section is manufactured from ¾” MDF.

There is a second Focal TI 24TD tweeter mounted on stand-offs and an aluminum backing plate on the rear of the mid/high panel. I plan to wire this out of phase (nothing new here) with the front tweeter. I tried to keep as much of the bracing on the back as I could minimized to reduce the back wave coloration. The reason for the wide frame that holds the mid/high panel is an attempt to follow the same rule. Not sure how well it will work but I am ready to find out.

The grill frame is a separate assembly that is removable and installs from the top. It is held in place with grill retainers and dowel pins. The outside wall fascia of the bass section is ¾” hardwood and the speaker sits on rubber feet.

The entire speaker assembly is 25.5” wide, 41” tall and 8” deep. Try as I did, I could not get it any smaller, working under the constraints I set for the project. It is still larger than I would like, but I think I can work with a pair of speakers that size.

The speaker will be quad amped, the amps driven by a 4-way active crossover. Two sections of the 4-way will drive the tweeters (separating the front and rear) so I can adjust them independently. The other two sections will drive the bass and mid sections.

Hopefully I will be able to start construction in the next couple of weeks.

As always feel free to critique and offer input.

Matthew

[scorpion]

Matthew,

The design is not final. Have a look at Stig-Erik Tangen's rubber decoucling technique here: http://www.diyaudio.com/forums/multi-way/142015-my-open-baffle-dipole-beyma-tpl-150-a-83.html .

Just free your mid and tweeter from the bassinfluence by using something like his decoupling from the the bass baffle. It does have importance.
You don't have to go nude, the speaker baffles are narrow enough to preserve constant directivity high up, but some kind of decoupling would be benificial to your design. Best would be total physical decoupling.

/Erling

[JohnR]

Hi, if you're going to be experimenting with different mid/tweeter panels, just wondering if you might want to make the top support taller?

[guest60106]

Matthew,

The design is not final. Have a look at Stig-Erik Tangen's rubber decoucling technique here: http://www.diyaudio.com/forums/multi-way/142015-my-open-baffle-dipole-beyma-tpl-150-a-83.html .

Just free your mid and tweeter from the bassinfluence by using something like his decoupling from the the bass baffle. It does have importance.
You don't have to go nude, the speaker baffles are narrow enough to preserve constant directivity high up, but some kind of decoupling would be benificial to your design. Best would be total physical decoupling.

/Erling

Erling,

Thank you for the reply. I have followed many of your posts and find your work quite fascinating. The mid/high panel is mounted on rubber isolator mounts with no through fastener. They completely release the 2 panels from a hard mechanical connection. I take from your response that you would like to see more decoupling. I have considered this as well and believe your statement to be true. In fact, this is the 12th iteration of this OB design. 3 of the models that came before this one had completely separated mid, tweet and bass baffles. I decided on this model because if I cannot get enough decoupling with this mounting method, I can mount the separate baffles with a more forgiving method in this design after I do some testing.

My major concern at this point is weather or not the main bass baffle is thick enough. The 4 10s have a combined cone area close to equal that of a 20” woofer. But, they will have much more moving mass than a single 20" (am I wrong about this?). The combined power output capability of the 4 10s will be in excess of 1200 watts. I have considered 3 options.

The easiest would be to wire 2 of them out of phase although I have never seen that done on this type of design with all drivers facing forward. Some of the papers I have read by SL advocate reversing 2 of the drivers and wiring them out of phase for more than one reason not the least of which is to minimize vibration. This is tempting. I can’t think of any reason it would no work. Am I missing something?

The second would be to mount a ¼’ inch thick aluminum damper on the front or rear of the bass panel. I could do this by making the aluminum panel identical to the bass panel and bonding the two panels together. I have done this with some success in the past but never with an OB design and it is a bit labor intensive.

The third option would be to just make the bass baffle thicker.

Let me know what your input is on this.

[guest60106]

Hi, if you're going to be experimenting with different mid/tweeter panels, just wondering if you might want to make the top support taller?

John,

Thank you for your response. I have enough room in the top of the design to get an 8” mid driver with the large tweeter that I am using. If I go to a smaller tweeter, I could possibly get a 10” driver in the correct position. I don’t think I will be using a bigger driver than that for the midrange. The drivers now are an Arum-Cantus AC-180F1D 7” and a Focal TI 24TD. The Focal face plate is close to 5" in diameter.

[scorpion]

Hi Matthew,

I see, you have done a lot of thinking to arrive at this design. Go ahead, rubber decoupling should be good enough and I should perhaps also have spotted it in the drawings, and above all read more carefully. But do try woodpanels beside the Alu to see what you like best soundwise.

Regarding woofer placement and wiring. I think you could test both options and see what you like best. It will depend on the quality of the woofers which you prefer I think. Mounting two in opposite direction and wire in opposite phase will prevent vibration and cancel 2nd order distortion. There is however an argument that this will make 3rd order distortion more audible, which generally would be considered worse.

All front mounted I think they should be wired in phase. In any case, dealing with as many as four potent bassunits, I would suggest making a baffle of two parts of the same material glued together with damping glue such as 'Green Glue' to make a 'dead' resonant free baffle.

/Erling

[JohnR]

Some of the papers I have read by SL advocate reversing 2 of the drivers and wiring them out of phase for more than one reason not the least of which is to minimize vibration.

To add to Erling's reply - the vibration minimization applies only when the drivers are mounted facing each other as in a Linkwitz W frame. Doesn't apply when the drivers are in the same plane.

With regard to concerns about baffle vibration - I'd suggest building the baffle cheaply first and measuring (i.e. a prototype) as you don't want to find an issue with baffle size/shape etc after you have built a nice baffle. (Mine are still half-inch chipboard!... I'll get a nice one done eventually...)

[D OB G]

Hi Harvylogan.

You’ve obviously given your design a great deal of thought.

One suggestion you might like to consider is an improvement in the resistance to vibration of the woofer panel.

If the panel was one driver high I don’t think there would be an issue, or with a deeper H or U frame, but with it being two drivers high I would worry about the leverage applied to the panel.

It wouldn’t look aesthetic, but maybe some kind of triangulating brace from the top of the panel, back to an extension of the base of the panel might offer a worthwhile improvement
 (a la Quad ESL).

I agree with scorpion that isolation of the mid/tweeter panel from the bass panel is crucial, and may require many trials to get right.

Regards,

David

[guest60106]

Ok,

For now, the reverse polarity of 2 of the 10s will be scraped. Since it will be easy to test with an RTA by just swapping some wires, I will put it on the back burner.

I am also inclined to agree with Erling, JohnR and David on your points about the bass baffle Integrity. The build plan up until this point has not included a preliminary test baffle for the bass units. Not that I didn’t consider it, I did. I just felt fairly confident in the driver layout based on my research. It may be a bit short sighted. So, building a test baffle has been added to the plan.

As for the final design, I played around with the model today and came up with a viable solution for the bass panel integrity. I don’t want to add any more bracing to the rear for fear of coloration of the back wave. This fear may or may not warranted below 100-500 hz but if the bracing isn’t there then I wont have to worry about it. Revision 1, is making the baffle itself thicker by doubling the MDF. This turned out to be a fairly easy fix (see the pics). I have often thought about how much baffle thickness colors the back wave in the respect that the thicker the baffle is the more the back of the driver is sitting in a tube. What the practical limits of this stiffening technique are I don’t know. There are some slight correction methods that I have seen employed such as stepping the whole diameter larger to the rear for each thickness of material applied. This is why I have decided to include a test baffle into my design plan. This very reason is also why I decided to manufacture the Mid/High panel out of aluminum. Doing this allowed me to make the panel ¼” thick and still maintain panel integrity. The midrange driver flange itself is thicker than the panel. The entire back of the driver frame and most of the cone (except the edge roll) protrude the rear of the panel. Will this be audible? I am not sure but it cant hurt. I will water jet those panels this week.

Thank you for your input.

Keep it coming

Matthew.