6.1.2 Keel Access Cover Flanges

This entry is part 23 of 42 in the series 06 - Fuselage

I made the flange for the large opening on the copilot side of the keel in Florida.  But the forward and aft access covers need to be dealt with.  This is accomplished by taking the pieces which were cut off to make the opening and adding fiberglass layups over them which extend to the keel. The covers are first taped into place then the area around the opening is covered with duct tape.

2008-04-05 1310 IMG_67082008-04-05 1310 IMG_67092008-04-05 1310 IMG_6710

Next BID layups are placed over the cover and adjacent area of the keel.

2008-04-13 1157 IMG_67262008-04-13 1157 IMG_6727

When it cures, I drilled the holes for the screws that would hold it in place. The cover and flange is gently pried away from the keel.  The flange is then trimmed back.

2008-04-13 1202 IMG_67292008-04-13 1203 IMG_6730

All that’s left is to put nutplates on the keel and now the access cover can be screwed into place.

2008-04-13 1614 IMG_6734 2008-04-13 1615 IMG_6735 2008-04-13 1620 IMG_6736

6A.3.1 Rudder Pedal Installation

This entry is part 24 of 42 in the series 06 - Fuselage

I received the rudder pedals from Velocity. In past Velocity models, the rudder pedals behaved differently than most other airplanes. The (old style) pedals were independent.  Meaning that you could depress both pedals at the same time. Once a pedal is pressed past the point that deploys the rudder, pressing the pedal further would activate the brake. There was one advantage to this in that you could press both pedals in flight and by deploying both rudders achieve a kind of speed brake.

But the rudder pedals of nearly all other airplanes are interconnected. Meaning you could only press one at a time.  And brakes were activated by pushing on the top of the pedal. Hence, “toe brakes”.

The new, upgraded rudder pedals are like that of traditional airplanes.  And those are what I bought.

So I mounted the pedals on the torque tube assembly and began trying to figure out how high to mount the assembly.

Here the pedal assembly is just resting on the keel.

2008-04-18 0936 IMG_6755

To determine if the height from the floor was correct, I put some pieces of foam to create a height of about where I thought the seat would be then sat down I tried out rudder pedals at different heights.

2008-04-18 0936 IMG_6756

Once I had to height dialed in, I bolted the whole assembly to the canard bulkhead.

 

6.2 Assemble Seats

This entry is part 26 of 42 in the series 06 - Fuselage

I decided to build up at least one seat that I can use for testing. Basically, I have a seat bottom and a seatback. I have to screw these to hinges. The hinges allow the seatback to fold forward so people can get in a out of the back seat. I’ve got a new style hinge. I’m rapidly coming to the conclusion that new=a lot more work. The old style had a fixed upright position. Which means that you had to decide when you were building the seat. The new style lets you adjust it at anytime, just like a car seat.

2008-04-19 1712 IMG_6782

Here’s the new style hinge.

2008-04-19 1714 IMG_6787

Old style hinge.

But the problem is that I couldn’t get the seatback to fit the hinge. I tried and tried. I tried all sorts of different ways of getting it to fit. It just wouldn’t. Then I started looking at other peoples build sites. That’s when I discovered that nobody else have seat hinges like mine.

So I called the factory. They said I would have to send the seatbacks back down to Florida to be modified for use with the new seat hinges. 🙁

6.6.2 Install Instrument Panel

This entry is part 25 of 42 in the series 06 - Fuselage

The flow chart said to mount the instrument panel. This seems a bit premature, but hopefully the person that wrote the manual knows what they’re doing. (turns out they didn’t)

  2008-04-21 1218 IMG_6791

The instrument panel is supposed to be 21″ aft of the canard bulkhead (That’s the bulkhead that the rudder pedals are mounted to). To make sure I get the right location, I marked a couple 3/4″ x pieces of wood and clamped them to the canard bulkhead. This will give me the right distance.

Then I tried to put the panel in. Tried is the operative word. It’s made oversized so I have to trim it to fit. Which means put it in as far as possible and mark where it binds, pull it out, trim and repeat. And repeat, and repeat.

 Once I got the panel to fit in the correct location, I applied duct tape to the back edges. I then applied a 2 BID layup on the left, right and bottom middle of the panel and let it cure overnight. In the morning, I drilled holes through the panel where the layups were.

 2008-04-23 1847 IMG_68002008-04-23 1848 IMG_6804

New mounting tabs.

2008-04-23 1842 IMG_6799

Panel temporarily held in place with clecos.

Here’s what I learned much later:  The panel didn’t have to get installed now. It could have (and should) have waited until much later.  Had I waited, I would have mounted it farther aft.  As it is now, with the seat in the position where I would be flying from, it’s a decent reach to get to the panel.

6.8 Doghouse Edge Finishing

This entry is part 27 of 42 in the series 06 - Fuselage

Next was reinforcing the doghouse and doghouse opening. The “doghouse”, by the way, was my first clue that things were not going to be easy. If you look at “Task 2” on the task list, you’ll notice it says “Cut out Doghouse”. In the actual task list, it also says that task is in “Chapter 6”. But it doesn’t say WHERE in Chapter 6. There are 31 8 1/2 x 14 pages in chapter 6 with about 41 separate tasks. Some of these tasks do not apply since they were already done during the pre-build done at the factory (these steps are not identified in the manual so you have to read through another list to find out that it doesn’t apply to you). So the only way to find the steps involved with this task is to read through the entire chapter.

I eventually found the section on page 35. Which means the very fist step in building the airplane was in Chapter 6 on Page 35.

Go figure.

Anyway, I cut the doghouse out back in January. Where the cut was made is now raw, exposed foam on the doghouse and the fuselage. The reinforcing is supposed to be accomplished by first digging out some of the foam between the inner and outer skin and filling it with a mixture of epoxy and cab-o-sil. Cab-o-sil (called “cabo”) is an additive that is mixed with epoxy to make it thick.

 What that does is effectively cover the foam.  And while epoxy/cabo is strong in compression, it’s not very strong in shear.  So why not cover it with a layer of fine BID?

Here’s the front of the doghouse opening after I removed some of the foam between the inner and outer skins.

  2008-05-30 0728 IMG_7113

Here’s where I made another “engineering change”. Malcolm at Hangar 18 doesn’t fill the groove with just epoxy and cabo. He lays in a bundle of glass fibers the entire length on the groove. This creates an extremely strong arch. The only problem is that the fiberglass strands that he uses come in very large quantities that I would only need a fraction of. So I took some unidirectional fiberglass cloth (in unidirectional cloth, the glass fibers run in one direction only and are held in place with some small cross fibers) and stripped out the fibers by hand until I had enough to do both sides of the opening and both sides of the doghouse.

Here are the four bundles of glass fibers. Two short ones for the front and two longer ones for the rear.

  2008-05-30 0729 IMG_7116

Then it was time to saturate the glass with epoxy and embed it in the groove.

One of the bundles being saturated in epoxy.

2008-05-30 0904 IMG_7119

Close up of the end.

2008-05-30 0903 IMG_7118

Rear opening with epoxy/cabo mix ready for the “horsey tail”.

  2008-05-30 0729 IMG_7115

Once the fiberglass strands were in place I covered them with a layer of BID. After it cured, I needed to get the space between the canard and fuselage consistent. So I covered the doghouse side with a layer of duct tape and mushed in an epoxy/cabo mix in between. Afterwards, I pulled the tape off and I had a perfectly even gap between the doghouse and the fuselage.

Finished edge.

2008-06-25 1239 IMG_7184

6.8 Canard Reinforcements

This entry is part 28 of 42 in the series 06 - Fuselage

Now that the elevator torque tube has been positioned, I’m going to get the canard mounts installed. There are a couple steps that the manual has being done all at once. But I’m a little concerned about doing multiple steps at the same time of I’m going to do this separate steps. The first task is to make sure the canard fits in the opening, is level (side-to-side), has the correct incidence (level front-to-rear) and is perpendicular to the centerline of the fuselage

The first step is to get the fuselage level left-to-right and front-to-rear. Placing a level on the firewall showed that I was level front-to-rear. Side-to-side level is determined using the center spar (what the wings mount to). The center spar sticks out about 5 feet on either side of the fuselage. But a two to four foot section of the top of the spar is not necessarily representative of the entire span. So first I tried running a string from one end of the spar to the other and using a little bubble level that hangs from the string. But after fussing with that for a while, I kept getting variations. So I gave up on the easy way and used a water level. A water level is a long piece of clear hose filled with water. You put one end of the hose at the end of the center spar and the other end of the hose at the other. When the water level is at the same point of the spar, it’s level. But with one person, there’s a BUNCH of walking back and forth until it’s just right.

Looking from the right side, you can see where I have one end of the water level clamped to the center spar.

 2008-08-03 0850 IMG_7284

Here’s a closeup of the right side. Notice the level of the water is aligned with the top of the spar.

  2008-08-03 0851 IMG_7285

Next, I check the side-to-side level of the canard. First, I used my four foot construction level to get it close. Then I opened the garage door and walked about 20 feet from the back of the plane. Using the top of the right end of the center spar, I raised and lowered my head until I could just see the top of the right end of the canard. Then I looked to the left to check that the canard was the same as the center spar. It wasn’t. So I sanded the bed that the canard sits on one side and shimmed a bit on the other. Then it’s back outside to eyeball it again. After a dozen or so iterations, it was perfect.

Looking from the back, you can just barely see the canard above the spar. It looks like a little white line. In fact, you can see a bit more white (the canard) on the right side than  you can on the left.

  2008-08-03 0830 IMG_7278

Lowering the camera just a little and now you can’t see any canard… Well, there is just a tiny bit on the far right.

  2008-08-03 0830 IMG_7279

Then I had to check the incidence. There’s a gauge that sits on top the canard and you put a level on that. When the gauge is level, the incidence is correct. So I had to do some tweaking to get the incidence right. But screwed up the level (side-to-side). So I had to fix that which screwed up the incidence. After a while, I had both correctly set.

Looking from the fuselage out along the canard. On the top is the “incidence gauge”. When the top of the gauge is level, the canard will be at the correct angle of attack. Resting on top of the gauge is my combination square/level.

  2008-08-03 0845 IMG_7280

A close up of the level. The front is a bit low (or the back is too high). A little shimming and…

  2008-08-03 0845 IMG_7282

Perfect!

  2008-08-03 0847 IMG_7283

Then I had to check that the canard was square to the fuselage centerline. This is done by measuring from the left end of the spar to the left end of the canard. Then you do the same on the right side. Since the spar is square, the two sides will be the same if the canard is perpendicular (it wasn’t). So I had to make some adjustments. Which messed up the two previous adjustments. So I got the side-to-side level, got the incidence right. Then I had to get the side-to-side level fixed (again) which messed up the incidence. Once I got those to right, I had to tweak the perpendicular setting which messed up the other two.

Spar/Canard Measurement

I probably spent about two hours getting this right. And keep in mind that this was just to make sure that I could get all three axis correct. Once I was sure I could do this without any sanding or grinding of anything, it was time for the first major step. The canard rests on the fuselage at the bottom of the opening. I removed some foam from between the inner and outer skins.

Left side of the fuselage where the canard rests.

  2008-06-25 1239 IMG_7182

I filled the space between the inner and outer skin with a mixture of epoxy, milled fiber (for strength) and cab-o-sil (for thickness). Then I put a layer of thin aluminum tape of the bottom of the canard to prevent the canard from being permanently attached to the fuselage.

Now it was time for the hard part. I lowered the canard in place, the epoxy squished out of the space between the canard and fuselage. But now I’ve got the get the canard set correctly. So I spent the next hour or so adjusting, readjusting, readjusting, readjusting, readjusting, etc., etc., etc. It’s a good thing the epoxy takes hours to set up!

The following day I was ready to install the canard tab mounting bushings. On the bottom of the canard are two mounting tabs that are the primary means of attaching the canard to the fuselage.

These tabs fit up against the “canard bulkhead” which you can see in the previous picture. It’s the part that goes across the fuselage under where the canard sits.

One of the canard mounting tabs

  2008-08-04 1233 IMG_7294

First I removed and then reinstalled the canard and then checked all the measurements again just to make sure the canard would in the exact correct position… And it was. Then I drilled a 1/4″ hole through the hole in the tab through the canard bulkhead. Then I removed the canard. I opened up the hole in the canard bulkhead to 1-1/4″ so the bushing would fit. Then I enlarged the hole in the tab to 5/16″ so I could fit the bolt through.

Then I applied Vaseline to the bolts, nuts, washers and canard tabs. Finally, I reinstalled the canard, coated the bushing with structural adhesive, put the bolts in and tightened the nuts. Then checked the alignments of the canard (again).

Right side canard tab.

  2008-07-06 0717 IMG_7210

Left side canard tab. The yellow tape under the bolts is my way of indicating which bolts are temporary. In most locations, locking nuts are used. They are basically single use nuts so if you remove them, you should throw them away. In many locations, things will be coming back apart so I use non-locking hardware. This is how I’ll remember.

  2008-07-06 0718 IMG_7213

The next morning I removed the bolts, pulled out the canard and then reinstalled it again… And checked the alignment again.

Canard bushing in place.

  2008-07-05 0602 IMG_7191

Now that the canard is mounted, I needed to get the doghouse to fit.

Not very aerodynamic.

  2008-07-06 0935 IMG_7215

I used a compass to determine the approximate line to cut on but as usual, I chickened out. I was afraid that I would cut off too much. So I cut a bit shy of the line. Then put it back on, marked it again, took it off, cut a little closer, put it back on, marked it again, etc., etc., etc. And a few hours later…

Doghouse in place.

  2008-07-06 1019 IMG_7216

Now I had to build the canard torsion tabs. These are additional layups that prevent the canard from twisting. First, I had to locate on the canard where the inside of the doghouse will be.

Here I’ve drawn a line to indicate where the doghouse is. I then sanded down the white filler on the inside. Under where the doghouse would touch the canard, I put down a layer of tape. Then Bondo is used to TEMPORARILY hold the doghouse to the canard. That’s right, Bondo. As in auto body filler. I works great as a temporary means on holding parts together.

  2008-07-13 1017 IMG_7219

Then the canard and doghouse are removed as one.

Next, I had to build a form for the canard torsion tabs.

Canard/doghouse on the workbench. You can see two dabs of bondo where the doghouse meets the canard. The wood on the fore and aft outboard parts of the doghouse are the forms that create the torsion tabs.

  2008-07-14 1429 IMG_7220

Duct tape prevents the layups from permanently joining the canard to the doghouse.

  2008-07-14 1429 IMG_7221

Creating a radius with epoxy/cab-o-sil. Triax doen NOT make sharp turns well.

  2008-07-14 1429 IMG_7223

Then as many as six layers of triax are used to create the tabs. Once it cures, remove the forms, sand off the bondo and…

Finished canard torsion tabs.

  2008-07-27 1608 IMG_7253

Once the canard torsion tabs were done, they are covered with duct tape. The canard is reinstalled back in the plane and wood is used to create a form for the forward and aft torsion tabs.

Left rear torsion tab.

  2008-08-05 1932 IMG_7298

Left front torsion tab

  2008-08-05 1932 IMG_7299

Elevator torque tube cutout.

From the right side looking at the canard installed without the elevator. You can see three of the elevator hinges. But there’s no opening on the fuselage for the elevator torque tube.

  2008-07-19 1358 IMG_7230

The first thing is to locate where that hole is supposed to be. Some people eyeball it. Some have used string through the hinge holes. But I got a better idea… LASER!

I setup my laser out beyond the end of the canard and adjusted it some that it was shooting a vertical line through the hinge holes.

View of the outboard hinges with the laser on.

  2008-07-19 1359 IMG_7233

Right side of the fuselage. The center mark is what passed through all the hinge holes. With a pencil, I put a vertical mark where the laser hit.

 2008-07-19 1359 IMG_7234

Then I rotated the laser and repeated with a horizontal line. I put a pencil mark there as well.

2008-07-19 1403 IMG_7235

“X”, or rather “+” marks the spot.

2008-07-19 1403 IMG_7237

Then just to make sure, I changed the laser to a point to re-check.

2008-07-19 1403 IMG_7236

Drill a small hole to get started

2008-07-19 1416 IMG_7239

Then make it bigger

2008-07-19 1420 IMG_7241

Then use the 1″ hole saw for the final opening. You can see the hole on the other side as well.

 

Since the torque tube is attached to the elevator, I have to cut away the area above the hole I just drilled.

Marking the cut lines.

2008-07-19 1436 IMG_7244

After cutting.

2008-07-31 0833 IMG_7277

6.8.2 Doghouse Attach Points

This entry is part 29 of 42 in the series 06 - Fuselage

Now it’s time to fit the bottom of the doghouse to the canard. First I mount the canard on the fuselage then covered the top of the canard where the doghouse rests with duct tape. Then fill the bottom of the doghouse with epoxy/cabo and lower the doghouse onto the canard.

Epoxy/Cabo mix squishing out.

  2008-08-12 1553 IMG_7324

Once it cures, a quick cleanup of the excess epoxy and I’ve got a perfect fit.

Now for another departure from the plans. The manual has the doghouse being attached to the canard torsion tabs with four screws. There’s nothing wrong with this approach, but there’s another way that doesn’t use any externally visible fasteners.

First I cut a length of piano hinge a little less than the length of the opening. Then I drill four holes through the hinge and tab.

Here the hinge is temporarily held in place with cleco’s.

2008-08-13 1535 IMG_7326

Then I countersink the screw holes.

2008-08-13 1634 IMG_7327

Apply some structural adhesive and screw it in place. I’ve got duct tape on the hinge to keep the two sides from becoming permanently attached to each other.

2008-08-13 1658 IMG_7328

Inside view.

2008-08-13 1658 IMG_7329

I drilled four holes through the top part of the hinge and the doghouse. I then drilled out the four holes to about 1/2″ and filled the area with epoxy/milled fiber/cabo. Once it cured, the drilled it out again.

2008-08-15 0609 IMG_7331

I covered the top half of the hinge with structural adhesive, lowed the doghouse in place and inserted bolts though the holes from the outside into the hinge. Put the nuts on and tightened. I also filled the holes on the outside of the doghouse.

When it was done, this is what the top of the inside of the doghouse looks like.

2008-08-16 1225 IMG_7337

When I put the doghouse on, this is what it looks like (on the inside) just before it’s all the on.

2008-08-16 1227 IMG_7343

Here it is all the way in position.

2008-08-16 1227 IMG_7344

Then I just slide the hingepin in place and the doghouse is held in place with no externally visible fasteners.

2008-08-16 1227 IMG_7345

From the outside.

2008-08-16 1227 IMG_7346

6.5.4 Install Nylaflow Tubing for Rudder Cables

This entry is part 30 of 42 in the series 06 - Fuselage

The rudders (one of each winglet) are activated by cables that go from the rudder pedals to the back of the fuselage and then out the wings to the winglets on the end of the wings. On my Cessna those cables get from the pedals to the rudder by means of numerous pulleys. On the velocity, the cables go through a nylon tube (Nylaflow) down each side of the fuselage near the floor. The tubing is glassed in place. But at the rear of the fuselage, it has to transition from the floor to penetrate the gear bulkhead about 12″ above the floor. During operation, this unsupported tubing can flex and make for slop in the feel of the rudders. Another Hangar 18 solution is to support the nylafow tubing between the floor and the gear bulkhead by putting inside a short length of aluminum tubing.

Here’s the tubing on the left side. I haven’t pulled the nylaflo yet but it will go inside the aluminum tubing.
2009-05-27 1647 IMG_8641
 Right side
2009-05-27 1647 IMG_8643

And here’s the finished product. nylaflo tubing installed and covered with a layer of BID.
2009-05-31 1021 IMG_8654

Closeup of the aluminum tube where the nylaflo goes in.
2009-05-31 1021 IMG_8655

6.3.7 Install Aft Keel Section

This entry is part 31 of 42 in the series 06 - Fuselage

This is going to require some explanation. The center keel runs down the center of the fuselage on the floor. It’s about 18 inches high at the front and drops to about 6 inches tall where it stops around 6 inches forward of the gear bulkhead. At the rear, it extends left and right to the sides of the fuselage. Inside of the keel is the aileron torque tube. Moving the stick left and right rotate this tube which causes the airplane to bank left and right. The only way to install the keel in a fastbuild kit is to cut it about 75% of the way back. Here’s a picture from February 2008 where Rick was preparing to make the cut. You can barely see the red “cut” line.

IMG_5729

Front part of the keel after the cut.

2008-01-16 1009 IMG_5735

This part of the keel got installed when I was down in Florida. The smaller rear part is called the “Whale Tail” (you’ll see why later). So now I need to install this part of the keel. When I fitted it in position, it didn’t line up with the front part. Two possible reasons; 1) the mold used to make the keel has… warped and doesn’t conform with the floor of the fuselage anymore. Or 2), The mold to make the floor of the fuselage has warped with the same result. Or it could be a combination of the two. In order to get everything aligned, I had to remove some of the material along the bottom rear of the whale tail. The farther back, the more material I had to remove. As I moved forward, less had to be removed.

Here’s the whale tail in position (see why it’s called a whale tail?). In the bottom left corner of the picture you can just see the seam with the forward part of the keel. Where the keel meets the floor is a flange that lays on the floor. But in the back you can see where I had to remove the flange so it would sit correctly.
2009-05-31 1021 IMG_8656

The seam where the rear keel section attaches to the front.
2009-05-31 1021 IMG_8658

Where there’s a flange that contacts the floor, structural adhesive is used to bond the keel to the floor. Where I removed the flange, BID layups will be used to strength.

Here’s the left side of the whale tail where it contacts the wiring duct. You can see a bit of the gray structural adhesive used.

2009-05-31 1021 IMG_8654

Next I created a fillet of epoxy/micro and then covered the flangeless areas with 2 layers of BID front and back. Here’s the same location with the BID layups applied.

2009-06-29 0806 IMG_88522009-06-29 0807 IMG_8853

Chicken strips
On the rear portion of the keel where I didn’t have to remove the flange and the entire front portion of the keel it’s attached to the floor with only structural adhesive.

Here’s the right side of the keel about half-way back.

2009-06-03 0737 IMG_8662

Close-up of the same area.

2009-06-03 0737 IMG_8664

Normally, in this type of joint you would reinforce it with two layers of BID. The manual doesn’t call for this, but I’m going to do it anyway. So I beveled the edge a bit, mixed up some epoxy/micro and then layed down two layers of BID. This will make it much stronger.

2009-06-03 0936 IMG_8665

6.9 Overhead Plenum Lights

This entry is part 32 of 42 in the series 06 - Fuselage

Soon, I’ll be installing the overhead fresh air plenum. I’ll need the have the lights fitted before I get to that point. I looked at Oshkosh for some good LED map/courtesy/flood lights. I found a bunch but as is the case with many things aviation, just because it goes in an airplane, you can usually add a zero to the price. What should cost $12 ends up costing $120.

$110 lights
$130 lights

$145 lights

See what I mean? But just like the door linkage…  Something I know about.

Here’s what I did. I found a light for boats. A Perko 12v light.

2009-08-17 1821 IMG_9148

I bought 4 of these at a marine supply store. Now this fixture has a 12volt incandescent light inside (I’m going to have a 28v electrical system). But that didn’t matter because I just wanted it for the plastic.

2009-08-17 1822 IMG_9150

I removed the bulb, mounting hardware and wires.

I purchased a dozen high intensity white and red LEDs at (of all places) superbrightleds.com.

Now I was venturing into uncharted territory. To have circuit boards manufactured would have cost me about $100. I discovered a way to make them myself. I created the layout using a free PCB (Printed Circuit Board) design program I found on the internet.

2009-08-21 1839 IMG_9151

Now here’s a neat trick: If you print the layout on slick, shiny paper, it transfers to the circuit board better. Finally a use for all those clothing catalogs my wife gets in the mail.

2009-08-21 1840 IMG_9152

I picked up a blank copper clad circuit board and Radio Shack.

2009-08-21 1840 IMG_9153

Then I put the paper on the circuit board and using a regular clothes iron, transferred the ink to the card. The paper comes off by soaking the whole thing in water for a couple minutes and then it goes into an etching solution (purchased at Radio Shack) for 20 minutes.

I cut out the individual, circular PCB and epoxied it to the back of the light fixture. Then I drilled the holes for the LEDs, resistors and leads and started soldering.

This first one was a “proof of concept” prototype. I just wanted to make sure the design worked. So I pretty much slapped it together and installed some test points instead of wires.

2009-08-22 1421 IMG_9157

I SAID it was a prototype. It’s not SUPPOSED to look pretty.

2009-08-22 1421 IMG_9158

Red

2009-08-22 1420 IMG_9155

White

2009-08-22 1420 IMG_9156

WooHoo!

One thing I didn’t like was when I soldered the components in the solder would run all the was along the trace. On real PCBs there’s a “solder mask” that contains the solder. Then I discovered that you can create a poor-mans solder mask using glass paint from the hobby store. This is paint that you can use on glass to create a stained glass effect. After painting it on, you bake it in the oven to cure it. After that, you got a solder mask.

Here’s a “production” model.

2009-09-14 1903 IMG_9233

2009-09-14 1904 IMG_9236

Total cost for the for lights: About $50.