 |
|
Note: To enlarge an image, left click on the image with your mouse.
Introduction:
|
|
|
Proposal
A friend approached me and asked if I could build some horn loaded cabinets
for four EV (ElectroVoice) PRO/15 woofers and two Altec 291-16 compression
drivers mounted on Altec 311-60 horns (see photos 2 and 3). He said he
would like something similar to the Altec 817 cabinet (see photo 4). He
wanted me to use 1" MDF (Medium Density Fiberboard) with the cabinets painted
black. I knew the cabinets were for his living room, and I could have taken
the easy way out and made some MDF cabinets painted black, but here was
my chance to try building wooden horns and finishing them inside and out
using oak wood. I had always wondered about the difficulty of building
curved wood cabinets and how they would appear when finished.
Photo 2 |
Photo 3 |
Photo 4 |
Planning
I downloaded from http://www.soundpractices.com/altec.html
several
drawings of Altec cabinets, including the 817. I could see right away that
a clone of this cabinet would require a forklift for handling, and looked
for alternatives. The Altec 816 looked like half an 817 (see photo 5),
so I scaled both drawings up to full size, and sure enough, the horn portion
was practically the same. The height of the 817 was twice that of the 816
to accommodate the second woofer. Looking at the Altec factory drawing
of the 816 (see Draw-1), I thought that cloning this cabinet in furniture
grade wood was practical.
Photo 5 |
Draw 1 |
So I contacted my friend and told him about the impracticality of building a clone of the Altec 817, and made him a deal: I offered to build four 816 cabinets with 3/4" MDF, covered with 1/4" red oak plywood on all sides to give him his desired 1" thickness and cover the inside with red oak veneer.
This would offer half the weight and size of the larger cabinets, and he could arrange them in a MTM (Midrange Tweeter Midrange) configuration if desired. I also offered to build them for the cost of materials only if he would promise not to use grill cloth. I figured the visual effect of finished wood would look better than grill cloth. My friend accepted.
|
I redrew the 816 drawing and allowed for the thicker cabinet while keeping the same interior volume (see Draw-2). I also allowed for more bracing of the back panel where an access panel would be needed. I calculated the need for 1/4" oak plywood to allow the grain on all sides (except the back) to run from front to back. I am fortunate to have a neighbor who cuts all my MDF on a professional table saw at his employment. So all I had to do was assemble the box, cover it with veneer or plywood and rout the edges. This neighbor recommended trying 1/8" birch plywood for the curved wood portion as it bends fairly easily. So it was off to the lumberyard to get the MDF, oak and birch plywood, and veneer. |
Formaldehyde
Cutting, routing, and sanding MDF creates a lot of fine dust. After finishing projects involving MDF, I have always suffered from fluid in my lungs and spitting up mucous for a few weeks. The local urban legend says this is caused by formaldehyde used in the glue that binds the wood particles together in MDF. While at the lumberyard I noticed a brochure touting MDX, which has the same weight, look, and size as MDF, but is formaldehyde free. At $14 more per sheet than MDF, I decided to try the MDX. After all, my friend was paying for materials.
Construction
I decided to build one cabinet first so I would not waste the remaining materials in case there were problems. I gave one sheet of MDX to my neighbor with the instructions for cutting the pieces to make one cabinet. He delivered the pieces the next day and I checked to see that the dimensions were correct. Assuming the vent was to be the bottom of the cabinet, I decided to build the cabinet from the top down. I placed what was to be the top on a work bench and drew the top view of my altered drawing, full size on the MDX (see Draw-2). Using a three foot stick with two adjustable points, I determined that the curved part of the drawing was the arc of a circle with a thirty inch radius. Since the 1/8" birch plywood seemed rather flimsy, I decided to use two layers. Using the thirty inch radius minus 1/4" to allow for the thickness of the doubled birch, I scribed and cut six pieces of scrap 3/4" plywood with a curve that had a radius of 29 3/4" and made 6 pieces 15" long. These were to be the "ribs" behind the curved birch plywood.
Next, I routed the circular hole in the driver mounting board, then glued 1/4" oak plywood on the board and routed the edges, including the driver circle. I then installed 1" oak tape on the edge of the cut out circle. Placing the oak side down, I centered the woofer over the circular hole and marked where the eight mounting holes would be drilled. Since I planned to use 1/4-20 button head Allen bolts, with metal inserts (see explanation of inserts later), I drilled the holes to a depth of 1/2" using a 9mm (23/64") drill, and installed the inserts with an Allen wrench
|
I have had troubles in the past trying to use T-nuts in MDF, as the points of the T-nut would bend instead of digging into the MDF. This resulted in a poor joint and difficult removal. A trip to my friendly hardware dealer turned up an ideal solution. An insert that screws into the wood and has internal threads to accept a bolt (see photo 6), and requires no access behind the insert. I used E-Z LOK # 9030 inserts that require a hole of 9 mm (23/64") drilled to a depth of 13mm (1/2"), and accepts a 1/4-20 bolt. For an EZ-LOK catalog check: http://www.ezlok.com Is it OK to mix metric with American sizes on vintage speaker cabinets ? |
Using these also meant that the driver mounting bolts would not show on the finished side of the cabinet. If you err on any of the insert holes, it is easy to drill out the hole, glue a dowel the same size as the hole you just drilled out, cut off flush, and redrill the hole to the required size.
Next I temporarily attached with 1 1/4" dry wall screws, the woofer mounting board, 14" from the front edge and centered on the MDX panel that would be the top. I marked this board 4" on either side of the center line. The curved portion module would have to go from this mark to a point 1" from the side on the top panel's front edge (see Draw-2).
Making the curved module
First I scribed another curved line 1/4" inside the existing curved line (to allow for the thickness of a double layer of 1/8" birch plywood). Then I attached with 1 1/4" dry wall screws, a 2 X 6 and a 2 X 4, each 16 1/8" long at either end of the inside curved line just drawn, to the top panel. I carefully marked the 2 X 6 and 2 X 4 where they were to be cut to follow the shape of the curve. Then cut the ends of the previously made 3/4" plywood "ribs" to fit exactly between the 2 X 6 and 2 X 4.
Photo 7 |
I gave my neighbor two 2 X 6s and two 2 X4s marked for cutting (shaping) and six shaped ribs, along with the 1/8" birch plywood. He cut the 2 X 6s and 2 X 4s as marked, installed the "ribs" using glue blocks, glued and stapled two layers of the birch plywood to this frame, leaving the birch 1/4" long at each end to be trimmed at final assembly of the module. When he returned the assembled module to me, I installed the MDX sides on the cabinet top and fitted the curved module inside by trimming the ends of the birch to fit exactly at each end. I countersunk the staples holding the birch to the frame and filled the resulting "dents" with automotive body filler (Bondo). I then sanded the birch smooth to accept the oak veneer. I use the filler instead of wood putty as it dries very fast and is extremely hard. It also allows for sanding in a few minutes (see photo 7). More information on Bondo can be obtained from their site: http://www.bondomarhyde.com/bondo.htm |
Photo 8 |
After rapping my knuckles on the curved birch, I wasn't impressed, nor was my friend. He decided to add five pounds of modeling clay to the back of the birch on each module. This was a new one for me, but I'm willing to try new things. I made a tray from scrap wood to roll out the clay to the proper size to fit in the space between the "ribs" in the rear of the module and cut some wood strips from scrap 1/4" plywood. The installation of the clay went smoothly. I covered the rear of the birch with Liquid Nails, laid the clay in the space between the ribs, and using more Liquid Nails, covered the clay with the plywood strips (see photo 8). |
Assembling the cabinet
I loosened all the dry wall screws holding the module in the cabinet,
clamped the mounting board to the two shaped 2 X 6s on the rear of the
modules, and while aligning the modules, mounting board, and the cabinet,
drilled eight holes through the mounting board and 1" deep into the 2 X
6s. Next I installed the inserts in the 2 X 6s, and bolted the modules
to the mounting board, and tightened the dry wall screws while keeping
the correct alignment.(see photo 9) Note the driver mounting inserts installed
previously. I checked all seams where the module, mounting board, and top
and sides of the cabinet met. Then I marked all areas that needed to be
planed or sanded to make a good fit. Next I disassembled, sanded, and planed
the marked edges to afford a good fit, and reassembled the pieces and checked
the fit again. I repeated the process until satisfied with the fit (see
photo 10).
Photo 9 |
Photo 10 |
After being satisfied with the fit, I disassembled it again and installed
the veneer to all the visible interior portions of the MDX, (while careful
to watch the grain orientation) and the birch portion of the module leaving
1/4" extra on the front edge to be trimmed to fit later. My practice in
applying veneer is to thin the contact cement using lacquer thinner and
roll it on both surfaces. I used the "Practical Pro" model RC-9T, 1/4"
Nap roller purchased from my local Formica dealer. I have never had any
trouble with bubbles or "hard" spots in the finish using this method.
Photo 11 |
When I was ready to assemble the cabinet for the last time, I gathered as many clamps as I could to glue and screw the cabinet together, taking care to ensure that all screws were countersunk. No need to putty the screw heads as 1/4" oak plywood will cover the entire outside of the cabinet (see photos 11 & 12). |
Photo 12 |
Before I attached the back panel, I made a 14" X 17" cutout and framed
the resulting hole with 2 X 2s. The 2 X 2s were glued and screwed to the
back panel leaving a 12" X 15" access hole for the drivers with a 1" border
to secure an access panel later. Then the the back panel was glued and
screwed to the cabinet.
Photo 13 |
Using yellow carpenter's glue, I installed the 1/4" oak plywood on all exterior sides, watching the grain orientation again, and leaving about 1/4" excess on all edges. I've found yellow glue to be much better and easier to use than contact cement to secure plywood. After the glue dried, I routed all the edges flush with each other. Then using a 3/8" router bit set to a depth of 3/8", I routed a "V" groove in all the edges except the front. (see photo 13). In these grooves, I glued 3/8" X 3/8" quarter round oak moldings (all front to back edges and all back edges). This style of finishing the edges is a lot easier than trying to miter the edges with a minimum of tools, all it requires is a router. Rounded edges done this way are less likely to be damaged in handling than sharp edges, and when used on "box" cabinets, minimize "edge diffraction". |
Photo 14 |
|
Photo 15 |
Photo 16 |
Photo 17 |
I made the 14" X 17" driver access panels from MDX, covered them with
oak plywood, routed the
edges, and cut a 4" X 5" hole in the panel to make the terminal plate
opening. I like to use the 5 way binding
posts (same as Adcom uses on their amps) available from Allied Electronics
as part # 645-0030 black and
645-0032 red, on everything I build, as it makes it easy to switch
components, switch in test equipment, etc.,
when all connectors are compatible. Allied has a catalog online at
http://www.alliedelec.com
Photo 18 |
|
Then I drilled the holes in the access panel and 2 X 2s, and installed
the metal inserts in the 2 X 2s
(same as the driver mounting inserts). I used 1/4-20 button head
Allen bolts with dress washers to
attach the access panel to the rear panel.
Finishing the woofer cabinets
The cabinet is now complete and ready for sanding sealer (see photo
19). In the past, I had good
results with oil based sanding sealers (especially "Enrich 015"), but
it's difficult to find anything
anymore except water based sealers. All the ones I've tried tend to
"ball up" when sanding, even after
drying for 48 hours in Phoenix (that means hot with low humidity).
So I applied the sealer and fought
the sanding till I was satisfied. After sanding the sealer with #220
grit paper, a coat of clear satin
polyurethane was applied. Then sanded again with #320 paper, another
coat of polyurethane, and
final sanded with #400 paper. After a final coat of polyurethane, it
looked very nice. Sections of 3 1/2"
Fiberglass were then installed on the interior sides, top, and rear
panel, leaving the area behind the port
bare. Now, since there were no major problems or disasters, all I had
to do was build three copies.
See Photos 20 and 21 for front and rear views of the finished Altec
816 clone.
Photo 19 |
Photo 20 |
Photo 21 |
Mounting the woofers
Of the four EV woofers, one had no magnet cover, and protruding from the rear center of the magnet was a 1/4-20 coupler (double female threads). I thought how easy it would be to provide additional woofer support from the rear as well as the normal mounting at the basket rim. To prepare the other three woofers, I removed the decal at the center of the magnet cover using a sharp knife. This exposed a Phillips screw and removing this screw allowed the removal of the magnet cover. These three woofers had a 1/4-20 coupler at the magnet center as well.
Everyone I had talked with or read about, who had supported the woofer
from the rear in addition to front mounting, had claimed improved bass
performance. Since it looked so easy to do, and I figured it couldn't hurt
anything, I made a 2" X 2" X 18" brace from two pieces of scrap 1" MDX
glued together. Holding this brace against the 2 X 2s on the inside of
the back panel, I drilled a 3/8" hole in line with the coupler at the magnet
center. Then I placed a 4" 1/4-20 bolt (a threaded rod would work as well)
through the hole and installed a large flat washer and two nuts on the
bolt. Next I screwed the bolt in the coupler and tightened the nut back
toward the brace until it was snug. I then re-tightened the basket mounting
bolts and checked the nut again. Next I tightened the second nut to the
first one as a locknut (see photos 22 and 23). Note: the nut and locknut
are not visible in the photos as they are behind the brace. CAUTION: I
did not over tighten this nut, even a 1/4-20 bolt can generate a lot of
torque. It would be possible to damage the driver, or push out the back
panel.
Photo 22 |
Photo 23 |
This completes the woofer cabinets, next was to build the cabinet for the HF (High Frequency) horns.
Acoustic Phasing
Before I could build the cabinet for the HF horn, I needed to establish
the correct position of the HF horn in relation to the woofers to ensure
the two were in acoustic phase. I made a "sled" from a scrap of MDX and
a 2 X 2, (see photo 24), and bolted the HF assembly to it. Then I made
a temporary spacer from scrap 2 X 4s to separate the two woofer cabinets
enough to allow the HF horn to sit between them and allow back and forth
movement. I also stapled carpet scraps to the spacer to protect the woofer
cabinets from damage (see photo 25). Note: We were trying different colors
on the interior of the 311 horn to determine a color to complement the
red oak and go with my friend's decor. This is the reason for the whitish
tinge on the horn. Fortunately, my friend doesn't have a WAF (Wife Acceptance
Factor) problem.
Photo 24 |
Photo 25 |
Quoting from an Altec service bulletin (slightly edited):
"Align the voice coils of the high and low frequency drivers in the vertical plane. Mechanical alignment is only the first step in the phasing of loudspeaker systems at crossover; there are certain to be phase shifts in both high frequency and low frequency drivers that must be compensated for. Once the voice coils are aligned, deliberately wire the high frequency driver with reversed polarity. Set up a Sound Pressure Level Meter (SPL Meter) at the listening position where you can see it while sliding the high frequency horn back and forth. In this situation, an assistant may be able to move the high frequency horn while you read the meter. Playing pink noise (several test CDs have pink noise tracks) through the system (either wide range or limited to a band in the crossover region), slowly move the high frequency horn until the crossover notch appears in the SPL meter. Watch carefully: there is an area only a couple of inches deep in which the notch appears. Position the high frequency horn so that you show the deepest notch you can get. The depth of this notch is a figure of merit for phase matching at crossover; 6 or 8 dB is excellent. You will notice that moving the horn only one-half inch either way will reduce the notch by several decibels. Now, reverse the polarity of the high frequency driver and the system will be phased correctly at crossover."
I tried for a few hours to follow the above directions, to no avail. I was temporarily using an Altec N-500-G crossover for testing to protect the 291 from large excursions, the pink noise track from the Stereophile Test CD 2, and a vintage Radio Shack SPL meter (model #33-1028). I tried connecting the drivers electrically in and out of phase, one and two woofers connected, and positioning the HF horn plus and minus up to two feet from center. At all the above configurations the pink noise output remained at the same level on the SPL meter! As a last resort, I disconnected the N-500-G crossover and tried again. This time I got a definite null of 8 dB with the HF horn connected electrically out of phase, and the front center of the 311 horn protruding three inches in from the front edge of the woofer cabinet. Any position other than the three inch protrusion caused the null to be less than 8 dB. Success at last! Did Altec provide an "automatic" phase correction circuit in this crossover? It seems the reasons for this should be investigated in the future.
Photos 26 & 27 show the four completed Altec 816 clone woofer cabinets.
Photo "28" shows
the set up of one side of a stereo pair in a "MTM" arrangement, to
be used in the design and testing
of a crossover. The HF cabinet is not built yet, but stay tuned !.
By the way, haven't had a sniffle using the MDX and I'm sure I was
inhaling lots of dust.
Photo 26 |
Photo 27 |
Photo 28 |
 |
 |
Left Speaker system
Right Speaker system
Left click on image to enlarge. |
Mid/tweeter horn driver is a Radian PB235 on Sierra-Brooks Mahogany Tractrix Horn. Lights in the cabinet separator columns
|