building wave guides from zaph's site

[peterS]

(sorry snag it or photo-bucket dumped the rez)


(bracing)

*note that the port is not to scale

these are the speakers remodeled to spec

now instinctively the bracing is skimpy

worth while to double the wall thicknesses?
definitely think i need more elaborate bracing
thoughts?

[JonMarsh]

There are many schools of thought...

I generally prefer thick front panels, 1" side and rear walls, and reasonable number of braces, but sometimes I'm thought a little over the top. Construction like the PE prebuilt cabinets works pretty nice, but they have 1" MDF fronts, and a brace running the length top to bottom; it splits the sides up into thin strips. See the Modula MTM thread for those cabinet details.

In this case, the use of addtional side to side braces is a good way to go without putting so much work or weight into the speaker. Will probably work pretty well as is- how much weight are you willing to put up with? What's the construction material you're going to use? 1/4" HDF can be added to BB ply or MDF, and is a nice working surface for veneering or painting or truck bed liner.

~Jon

[jkrutke]

Jon's got the right idea about weight - I'm usually a little weight conscious when I'm thinking about what I'll do for bracing. Those waveguide TMM's are already freakin heavy. It probably wouldn't hurt for you to add more bracing if you don't mind the weight. That's probably a better idea than double wall thickness. I'm not a proponent of double walls after some testing and deflection calculations I've done.

10 years ago I used to test enclosures using the good ol' knuckle rap. These days I use an accelerometer with impulses and sweeps. I haven't done that in a while either because after a while you pretty much know what the results will be.

The funny thing about using the knuckle rap on a box that uses simple small crossmembers is that it fails. knocking on one side, you transfer vibration directly to the other side. But using the loading of a real in-box woofer, the walls are symmetrically and inversely loaded - and the simple chunk of wood going across does an amazing job minimizing vibration.

[augerpro]

JK what did you find out about doubling walls that you didn't like?

[Chris7]

I'm building a pair of the Waveguide TMMs right now. (The cabinets and baffles are finished, but I won't have time to do any more work until the holidays.)

Having built the cabinets, I think they're pretty solid. The specified bracing seems more than adequate to me. They appear to pass the rap test, though I am not an expert. John is not exaggerating that they're very heavy, even without the additional 20 lbs of drivers and coils.

If you did want to increase the thickness anywhere, I might suggest the baffle. Once you cut out and rabbet the holes for the drivers, there is very little material left on the baffle. What remains is pretty light, and I can imagine it being susceptible to vibration. That's where I would consider doubling the thickness.

[jkrutke]

Agreed on the baffle thickness. I like to use removeable baffles, and sometimes I like to add a 3/8" layer inside, but inset so I have a little rabbet to help brace and position the baffle. And of course, your screws get a little more meat to dig into. Hmmm, the only picture I have of a baffle back like that is my sub. (3/4" 2nd layer on the back, airflow scallops and hurricane nuts) The only issue with thick baffles is insuring your driver can breath.

augerpro: Using double wall thickness will never hurt an enclosure, as it does add *some* stiffness and damping, but it's biggest effect is lowering the panel resonance which is not necessarily a good thing. A more efficient use of material would be better/more internal bracing, IMHO. I've got some calcs, measurements and findings that I may post in an article someday.

[Chris7]

Makes sense I definitely like the removable baffle.

One thing I think is worthwhile mentioning because I didn't realize this at first... in the perfectionist crossover:



I think R5 includes the DCR of the inductor L4. So rather than buying a 7 ohm resistor for R5, you'll probably want to buy a 6 ohm resistor (because your L4 coil will have a DCR close to 1 ohm).

You'll notice in the two more recent projects ("ZD5" and "Bargain Aluminum MTM"), Zaph changed the way his crossover diagrams are drawn to be a bit clearer about this. e.g. Here



notice the green resistor. I think in the Waveguide project, R5 would be green if drawn using that legend.

[peterS]

must there be two chambers or can they be consolidated to allow for a larger port diameter

[JonMarsh]

Larger port diameter would be required with two chambers, but then it would also require a longer port, which might be a problem. Unibox is your friend... it's free.

[capslock]

Depends on what frequency range you're looking at. Down low, no amount of damping is going to be sufficient, so better go for stiffness, i.e. braces, braces and more braces. By the way, OSB is cheap and pretty stiff. It just makes finishing or veneering a bear...

In the midrange, you could also opt for a stiff and heavy enclosure to minimize transmission to the outside. However, you'd have to rely on heavy stuffing to kill the energy inside the enclosure.

I prefer a sandwich construction of 1/2 or 2/4 in. MDF, a 3 - 4 mm layer of roofing tar, and an inside layer of 5 or 6.5 (1/4 in.) HDF or ceramic tiles. This is very dead in the knuckle test. However, bracing kind of defeats the purpose.

[peterS]

Larger port diameter would be required with two chambers, but then it would also require a longer port, which might be a problem. Unibox is your friend... it's free.

im aware of this but my question is in regard to the two drivers playing different frequency ranges- im assuming it doesnt matter but wanted to ask anyways
because im concerned that the port is too small to perform linearly w/ higher output


wont group delay be an issue with the current tunning?

[Chris7]

im aware of this but my question is in regard to the two drivers playing different frequency ranges- im assuming it doesnt matter but wanted to ask anyways

wont group delay be an issue with the current tunning?

John Krutke actually has a page devoted to your first question, specifically in the context of 2.5 way TMMs:

http://zaphaudio.com/commonenclosuretest.html

Basically, it doesn't matter.

As for your second question, the tuning is up to you. If group delay is very important to you, John does mention the possibility of running them sealed if you cut the box depth and cross over to a sub, though this is a low Qts driver so F3 is rather high.

[peterS]

Updated bracing anyone see anything worth noting before I start building?

[peterS]

also,
i plan on using a dynomat like material with accoustistuff adhered to it on all flat surfaces

[Chris7]

Updated bracing anyone see anything worth noting before I start building?

Looks good. One last thing to check is that the height is right for you. Zaph's listening axis for this design is halfway between the center of the tweeter and the center of the top woofer; adjust the height of the cabinet so that point is at ear level when sitting. Remember to take into account the height of the base and any spikes/feet you'll be using. (I built my cabinets without the base and they're not wonderfully stable, but I don't have kids so it's not a big issue for me. If you do have people around who might tip over the speakers, I suggest building the base Zaph recommends.)

[peterS]

thanks
i got lazy so i didnt add a second vertical brace in the chambers where the ports are but dont think they are of such importance- right?

as for the base it will be added along with spikes and probably the lower cavaty filled steel shot- i dont have it added in the pic because i may have to change the depth to arive at the correct volume

anyone know where to get steel shot and is it worth it?
im almost afraid to ask but would sand work as well?

[Ray Collins]

Any sporting goods or gun shop that sells reloading supplies will have it; lead shot is preferable.

Ray

[Dennis H]

Lead is a health hazard, especially if you have small kids around who like to put things in their mouths. Steel isn't as heavy but it's safer. Sand is lighter yet but it's still heavy enough to ballast a speaker and it's dirt cheap (bad pun ). Make sure the sand is dry before you put it in the box. Any lumber yard will have it in sacks for mixing concrete. I keep a sack around for when the sidewalk gets icy.

[augerpro]

I thought all lead shot was now tungsten/steel? Maybe the sporting goods store has some high tungsten content shot. I like the sand idea though.

[peterS]

now im still in the process of building these but since the front baffle will be removable that gives me the opportunity to change things

i dont really like what i see on zaphs readings for the seas tweet and am intrigued by the neo3 (non pdr) i think the "poor" of axis response is an asset in my small room

what would i need to do in terms of crossover modifications and more importantly anyone see any problems using the neo3?

[JonMarsh]

Neo3 with the waveguide is very unlikely. Won't mate up properly with the throat. A 2 kHz 2nd order crossover may be a stretch for the Neo3.

[Chris7]

i dont really like what i see on zaphs readings for the seas tweet and am intrigued by the neo3 (non pdr) i think the "poor" of axis response is an asset in my small room

If you're willing to spend a little extra, you can likely swap in the Seas 27TBFCG with no crossover mods. You'd have to buy both the TDFC and the TBFCG, then swap the faceplates of the two in order to mate properly with the waveguide. The TBFCG generally has lower distortion if that's what's bothering you, and it also has a bit of a dip in the very top octave. Note that I haven't tried this, but it's an option I'll try if I'm dissatisfied with the TDFC. Though I have my pair on hand and have given them a preliminary listen and haven't been disappointed so far.

[peterS]

Neo3 with the waveguide is very unlikely. Won't mate up properly with the throat. A 2 kHz 2nd order crossover may be a stretch for the Neo3.

no i wasnt suggesting using the wave guide-im willing to construct different front baffles

[Chris7]

no i wasnt suggesting using the wave guide-im willing to construct different front baffles

At that point you're looking at a completely different speaker. Assuming you've already built the box (and hence the baffle width is already fixed), you could build the Modula MTM in the same box, since the baffle width is the same. Completely different type of speaker, of course. You trade off the controlled directivity, high sensitivity, 2nd order crossover, and flat impedance of the Waveguide TMM.

[peterS]

im not sure im being clear
i will use the design zaph has made
i just like to upgrade (especialy considering there may be better tweets out there)
what id like to do is swap tweets with the wave guide on - possibly the Vifa XT25
and also swap the baffle to accomidate the neo3

so my question is what are the potential problems with this, what changes will have to be made, and what measurment equipment will i need

[Chris7]

im not sure im being clear
i will use the design zaph has made
i just like to upgrade (especialy considering there may be better tweets out there)
what id like to do is swap tweets with the wave guide on - possibly the Vifa XT25

It's not that easy. Swapping in another driver, apart from a modified 27TBFCG (which uses the same motor and general design as the 27TDFC), is not possible without basically redesigning the speaker.

From Zaph's FAQ (currently not posted because he's busy and not taking questions right now):
Q. I like one of your designs, but I'd like to use a different driver in place of one that you used. Can I do that?
A. No. My designs require specific drivers to work correctly. If you use a different driver, they may not integrate properly and the result is a system that sounds bad. Occasionally, I may know of a tweeter that might work as an alternate and I will list it. But unless I specifically list it, assume the answer is no.

and also swap the baffle to accomidate the neo3

Again, essentially a completely different speaker. You'd need to redesign the crossover from scratch. Nothing about Zaph's design would be preserved. It's a lot of work.

[Chris7]

Just for fun, I thought I'd upload some in-progress pictures of my pair of Waveguide TMMs. Sorry about the poor picture quality, I took these with my laptop webcam. There's still a lot of work to do (the cabinets need veneering, etc.) but they're coming along. The third and fourth pictures are of the tweeter mounting assembly, and the last is one of the milled waveguides.

[peterS]

i guess that answers my question - a whole new speaker- still i may play with that down the road i may give the xt25 w/ wave guide a try

I would greatly appreciate if you could explain how you milled the wave guide as i dont really follow zaphs instructions

[Chris7]

I would greatly appreciate if you could explain how you milled the wave guide as i dont really follow zaphs instructions

I built a jig for my router (I think this is what Zaph did too). First, use a hacksaw to cut off the neck of the waveguide just below the brass threads. You can do this by eyeball, there's lots of room. Then take a scrap piece of MDF, cut a 5.5 inch hole in it so that the waveguide can mount to it. Drill screw holes matching the waveguide's four mounting holes. Install machine screws to hold the waveguide in place so it doesn't move around. Turn the jig over with the waveguide mounted to it. Glue a couple of scrap bits of MDF to the back so the total height of the glued pieces is roughly about the height of the remaining neck of the waveguide. These are just height risers. Take another scrap piece of MDF. Drill a 3 inch hole in the middle. Glue this on top of the risers, with the hole centered around the waveguide neck. Basically now you've got a table with a 3 inch hole in the middle that you can set your router on. As you slide your router around on this table, it will move perpendicularly to the base of the waveguide and cut neatly and perfectly. Just make several passes, adjusting your router a little lower each time.

That explanation is a little wordy. It's easier than it sounds. I would buy one extra waveguide just as scrap in case you make a mistake. I bought two extras and only ruined one, and even that one could be used if I wasn't obsessive about getting the milled neck diameter exactly perfect on both waveguides. (The one I scrapped is about half a millimeter too large.)

There are other ways you could mill the waveguide down. For instance, if you have a drill press with a height stop, you might be able to put a straight router bit in the drill press then just slide the waveguide around on the drill press table. This is abusing the tool, of course, but the waveguide plastic is very soft and cuts easily.

[JonMarsh]

Can also be done with a router table, probably with a little more precision and repeatability- plus not needing to guide the router over a relatively small fixture. Just my 0.02- inexpensive router tables don't cost much and are well worth the money- a wide variety of lift tables are available at different price points. I used a benchtop Freud, but one of the Bosch tables might be easier to find locally (Lowes, for example).