Directiva r1.2 design and build

[Alan J]

My hand is healed enough to resume working but I have some things to do before there will be substantive progress: Remake my router table to be flatter and with a more readily adjusted fence, and assuming I stick with lock miters for the corners build a sled to cut them with more control and safety.

 

[Okankav]

is there any update?

 

[Alan J]

Regrettably, not just yet, due to one distraction after another, plus a great vacation, that's held up the workshop re-do. Over the next few weeks we have a grandchild babysitting stint, a big deal wedding, and my wife's getting a brand new hip. I'm actually hoping that while I'm home and on call for her recovery I can be getting some things done!

 

[Okankav]

Hİ Alan, is there any update?

 

[Alan J]

Soon, I hope! My hand is pretty well healed (I think that the therapist will send me on my way this week), my new and better workbench is complete, and my new router table is in place.

I need to make a cabinet material choice, informed by both acoustic properties and desired finish (and, OK, practicality). The plan is for the baffles to be the natural cherry probably with a simple Danish oil finish, and the boxes a charcoal color.

I spent this morning looking at Richlite (mentioned in @Rick Sykora's new thread on damping materials and have samples in my cart. But right now I'm most likely to go with 18 mm plywood either stained or painted, or MDF.

 

[Okankav]

hi alan, what will be crossover point in the last case? I remembered you worked with lr4 around 2,2 khz. Is there any change in this area?

 

[Joe BKNY]

Soon, I hope! My hand is pretty well healed (I think that the therapist will send me on my way this week), my new and better workbench is complete, and my new router table is in place.

I need to make a cabinet material choice, informed by both acoustic properties and desired finish (and, OK, practicality). The plan is for the baffles to be the natural cherry probably with a simple Danish oil finish, and the boxes a charcoal color.

I spent this morning looking at Richlite (mentioned in @Rick Sykora's new thread on damping materials and have samples in my cart. But right now I'm most likely to go with 18 mm plywood either stained or painted, or MDF.

Glad to hear your hand is doing much better. It wasn’t intentional, but you did the forum a service by reminding us to be careful!

 

[Alan J]

hi alan, what will be crossover point in the last case? I remembered you worked with lr4 around 2,2 khz. Is there any change in this area?

Right. We were thrilled by listening tests with LR4 around 2100-2200 Hz. This was with a DSP crossover, limiting myself to functions that are in theory available with passive components (not that it will be fully practical to replicate).

[EDIT] With that said, I'll be open to changes when I zero in on a passive crossover. I'm wary of going too high, at least not without ensuring that any higher frequency stuff from the Purifi is well managed. But crossovers that look best in simulation have consistently been in the 2000-2400 range and I'm likely to land somewhere in the lower half of that range.

 

[Alan J]

It wasn’t intentional, but you did the forum a service by reminding us to be careful!

In that spirit, a few pics:

New sled for passing vertical panels through the router (the tall piece of MDF there only for illustration):

A test lock miter joint (almost dialed in) done with the sled, along with stock guides for the piece that goes through flat:

In both cases, the setup makes it hard to get one's hand near the bit, and it's conducive to standing alongside the workpiece, not behind it. I also have proper dust collection now.

 

[Alan J]

Plywood for the enclosures is in hand! After a lot of searching and ruminating, and some testing of finishes, I chose Columbia Forest Products' 18 mm Europly plywood. The product is marketed as a substitute for Baltic birch. It has 13 birch plies plus thin face and back veneers of the chosen species. They import the core from Poland, then apply face and back veneers. At $108 for a 2x8' half sheet, it was more expensive than the full sheet price for BB but BB is still coming from Russia and I wanted to avoid it. In addition, test samples of our likely finishes on birch didn't look good. We were happy, in contrast, with finishes on Columbia's less expensive maple-faced plywood, samples of which cut without splintering of the face veneer.

It was easy enough to measure the elastic modulus: Support the sheet at both ends, place a weight at the center supported by a narrow board, and measure the displacement. Elastic modulus in the face grain direction measured at 9.7 GPa, which is a bit higher than published values for BB. Density is 85 lb per 4x8 sheet (722 Kg/m^3), compared with around 675 Kg/m^3 for BB.

 

[Alan J]

I'm occupied by other things over the next 2+ weeks, so a quick update...

Cutting of enclosure panels is finished. Lock miters are tedious to set up and require sub-millimeter precision especially if using a lock miter on the rear panel.

Both the black cherry baffles and the Europly enclosure parts are at 6% moisture content. From there, the plywood parts would not be expected to move much. In conditioned space, the baffles would be expected to grow a mm or so horizontally, or several mm if ever stored in un-conditioned space. Baffle mounting will keep the baffles centered on the enclosure while allowing for a few mm of movement on each side. Accordingly, the enclosures are about 0.5 mm smaller than the baffles all around, and there will be a thin spacer (stay tuned for details) so that the break between enclosure and baffle looks natural.

Coming up next: Purifi mounting to allow for some movement while providing a good seal (and perhaps some isolation in the process), as well as detailing mounting of the baffles to the enclosures.

 

[terryforsythe]

Beautiful wood work!!!

 

[Okankav]

hi Alan,

did you add any mass to sb passive radiators?

 

[Alan J]

hi Alan,

did you add any mass to sb passive radiators?

Yes. The current bass alignment calls for 40 grams. I'll revisit this later, after the enclosures are complete along with any needed bracing and the crossovers are installed.

 

[Alan J]

Beginning work on the Purifi mounting...

Advice from Purifi (see the Purifi thread over at diyaudio) is to use simple, secure mounting directly to the baffle, directing your creative energies to things that have a greater effect on performance. But I need to accommodate some baffle movement.

Elongated holes allow expansion across the grain. The allowance is for 3 mm expansion or 0.5 mm contraction. In practice (as long as the speakers are kept in indoor living spaces), likely expansion is less than 1 mm. Any contraction is unlikely considering that we're presently at around 6% water content. The round holes at the top and bottom keep the driver centered.

This backing ring holds captive cap nuts. It's not really necessary; I could use smaller individual pieces.

Foam on the driver and backing ring, plus O-rings behind the screw heads and cap nuts will provide the seal. At least that's the plan right now!

 

[Alan J]

A quick update... I've been working on the system for mounting the baffles to the boxes and providing both a reliable seal and accommodating movement of the solid cherry with humidity changes. This has included a lot of experiments with 3D printing of mounting blocks and O-ring seals. Stay tuned on those.

Meanwhile, this week I put finish on the baffles (2 coats of Osmo Poly-X), and today my wife and I glued the enclosures. To reduce anxiety, we used Titebond II Extend which has a 15 minute open time. The two of us were able to apply it to all of the joints in less than 5 minutes, but it really helped to know that we didn't have to rush. Trial runs with scrap showed that, despite the 20-25 minute total assembly time specified by Titebond, the tight, high surface area lock miters would not slide at all once the two parts were brought tightly together (though it was possible to pull them apart and re-position). So the process was (1) dry fit, (2) place alignment marks, and (3) bring the pieces together so that they're properly aligned when they make first firm contact.

 

[Paweł L]

A quick update... I've been working on the system for mounting the baffles to the boxes and providing both a reliable seal and accommodating movement of the solid cherry with humidity changes. This has included a lot of experiments with 3D printing of mounting blocks and O-ring seals. Stay tuned on those.

Meanwhile, this week I put finish on the baffles (2 coats of Osmo Poly-X), and today my wife and I glued the enclosures. To reduce anxiety, we used Titebond II Extend which has a 15 minute open time. The two of us were able to apply it to all of the joints in less than 5 minutes, but it really helped to know that we didn't have to rush. Trial runs with scrap showed that, despite the 20-25 minute total assembly time specified by Titebond, the tight, high surface area lock miters would not slide at all once the two parts were brought tightly together (though it was possible to pull them apart and re-position). So the process was (1) dry fit, (2) place alignment marks, and (3) bring the pieces together so that they're properly aligned when they make first firm contact.

View attachment 449960

Beautiful front baffle. One thing, because it's a thick baffle to reduce the reflections that will show up in midwoofers passband I would bevel the inner side of of the midwoofers cut out hole. 45 degree, 2/3 of the baffle thickness should provide air flow free from turbulence that will usually show up with such thick front baffle.

 

[terryforsythe]

Beautiful front baffle. One thing, because it's a thick baffle to reduce the reflections that will show up in midwoofers passband I would bevel the inner side of of the midwoofers cut out hole. 45 degree, 2/3 of the baffle thickness should provide air flow free from turbulence that will usually show up with such thick front baffle.

I was thinking the same thing. But, it looks like one of them already is together. If there ends up being ripple around 1.15 kHz, some filing may help alleviate that.

 

[Alan J]

Not together yet (blue tape for photo purposes)!

Thanks for asking about this, as it's a chance to sanity check my decisions.

Here it is in cross-section, using Purifi's solid model of the driver. Their model has the space between the cone and the basket as a solid, so you need to ignore that portion of the violet crosshatch. The baffle (apricot crosshatch) is cut back to a thickness of 19 mm. Behind that (also apricot crosshatch) is a mounting ring that's 5-6 mm thick. The mounting ring mainly holds the nuts (not shown) for the mounting screws. Secondarily, It distributes loads and provides a secondary seal. But I think it would be fine to use six short segments.

Qualitatively, due to the large foam piece, the Purifi has a lot of space between the cone and the basket. So I felt comfortable with just the inset on the back. At the left and right edges, the inset comes right up to the inside surface of the enclosure.

Two questions for folks with more experience:

1. Does this seem like enough space? And should I prioritize holding the nuts with six short pieces rather than the continuous ring?
2. @terryforsythe (or others) - Where does the 1.15 KHz from? (Had you not said that, I would have been asking what I should be looking for!)

 

[RickS]

Not together yet (blue tape for photo purposes)!

Thanks for asking about this, as it's a chance to sanity check my decisions.

Here it is in cross-section, using Purifi's solid model of the driver. Their model has the space between the cone and the basket as a solid, so you need to ignore that portion of the violet crosshatch. The baffle (apricot crosshatch) is cut back to a thickness of 19 mm. Behind that (also apricot crosshatch) is a mounting ring that's 5-6 mm thick. The mounting ring mainly holds the nuts (not shown) for the mounting screws. Secondarily, It distributes loads and provides a secondary seal. But I think it would be fine to use six short segments.

Qualitatively, due to the large foam piece, the Purifi has a lot of space between the cone and the basket. So I felt comfortable with just the inset on the back. At the left and right edges, the inset comes right up to the inside surface of the enclosure.

Two questions for folks with more experience:

1. Does this seem like enough space? And should I prioritize holding the nuts with six short pieces rather than the continuous ring.

Six short pieces and the back of baffle, should be beveled like this...

...and more deeply as your baffle is much thicker.