VAF Forums

VAF Forums (
-   RV Building Tips/Techniques (
-   -   The Unhappy Cracky Club (

Scott Chastain 01-11-2018 12:26 AM

The Unhappy Cracky Club
On the night of December 1, 2017, I flew my RV-8, N898W, Descending Dove, to Tehachapi, CA (TSP) and spent the night in the pilotsí lounge. I found after-hours access to a room with plenty of space on the floor to inflate my mattress and unroll a sleeping bag. I wanted to explore the town of Tehachapi the next day.

I awoke at first light. When I did, I quickly packed up my belongings and went out to the Dove to load up my sleeping gear. Pulling off the canopy cover, I discovered a ten-inch crack on the left side of the canopy directly over the passenger seat. Somewhat sickened by the sight, I covered it back up and spent the rest of the day exploring Tehachapi and mulling over my options with a pit in my stomach.

Later that afternoon, I found a local pilot by the name of Joe who helped me stop-drill the crack. I flew home to Merced (MCE) and watched the crack carefully over my left shoulder. It did not appear to be migrating further from the hole.

But over time, it did. I applied Weld-On #3 acrylic adhesive as some VAF members suggested. Then, two days before Christmas, I took a friend of mine to Half Moon Bay (HAF) for lunch. Following the return flight, I saw that the crack had splintered upward another 4 inches. I pulled the canopy and loaded it onto my truck to attempt a repair at home.

The RV-8 canopy was purchased from Todd Silver back in 2006. It was installed using the Sikaflex method without rivets or screws. The canopy skirts were also mounted to the outside of the plexiglass using Sikaflex, then riveted to the frame below the trim line. Thus, a full replacement of the canopy would entail having to fully strip the plexiglass from the skirts and frame, repainting the frame, and sacrificing the custom-made skirts in the process. Additionally, since it was a Toddís canopy, more likely than not, the windscreen would also need to be replaced since it bore a unique profile and tint to those sold by Vanís.

I foresaw in the replacement option a project that would involve a lot more time and effort and money than I was prepared to sacrifice at the time, so repair of the canopy and living with a battle scar was the option I chose. I decided that when it came time to overhaul the engine, upgrade the panel, and renew the interior, the canopy would be replaced, too. But for now, it was off to the garage for some homebuilt surgery.
There, a simple jig was fabricated to mount the canopy frame in such a way that it could be banked sharply in any direction without any possibility of falling to the floor and shattering. The forward portion of the frame was jigged in the roller receptacles using 5/8-inch hardwood dowels.

Scott Chastain 01-11-2018 12:29 AM

Opening Up the Crack
There were well over 1400 airframe hours on the canopy, and many thousands of hours beyond that were spent baking in the sun during airshow performances and flightline displays. About 100 Hobbs hours prior to becoming a part of the Unhappy Cracky Club, the canopy began sliding on the rail with difficulty. At first, I attributed the rough slide to a worn-out C-661 slider block; following the flightline fissure in Tehachapi, however, it was obvious that a side load on the block was causing the problem. The plexiglass had shrunk and pulled the frame to such an extent that the sliding process was being impeded. After the crack occurred, the canopy suddenly glided freely again on the rails.

The repair process began by heating the shop to over 70-degrees and positioning the canopy so that the crack lay horizontally. The surgical area was thoroughly washed with soap and water, then cleaned with PPG DX330 Acryli-Clean.

A Dremel #125 V-groove bit was used to begin the process of opening up the crack channel and relieving some of the built-in stress on the acrylic.

The Dremel bit was patiently and gradually worked to follow the crack deeper and deeper into the plexiglass as the V-groove began to form.

The crackline was clearly visible as the V-groove deepened.

Scott Chastain 01-11-2018 12:32 AM

Preparing the V-Groove
A piece of wet-dry sandpaper was inserted at the base of the crack as the V-groove channel widened.

For the final 5 or 6 inches near the apex, the crackline suddenly began to flatten out and carry the fissure into the interior of the bubble at a 45-degree angle. For that reason, the V-groove was opened much wider at the top portion than at the base.

After the channel was fully opened, all sharp edges were sanded using 120-grit sandpaper. The end result was a V-groove that measured a quarter of an inch at the top and gradually narrowed toward the skirts to the initiation of the crack itself. There, the previous application of Weld-On #3 appeared to remain intact.

The surgical area was once again cleaned with solvent then prepared from inside the canopy using aluminum tape. The tape served a double function of stabilizing the plexiglass, freeing up any distortion while preventing the bonding material from dripping through.

The PlastiFix kit was prepared for use and experimented with a few times prior to application to the canopy. A very light plastic powder was dispensed into a small cup, and liquid was then dispensed into the cup, one drop at a time, using a syringe-tipped dropper bottle. The resulting globule was then delicately stabbed with the syringe and lifted out of the cup, whereupon further dispensing of the liquid melted the powder from the needle as the mixture settled into its bonding bed. The trick was to work quickly before the mixture began to harden, and more than one or two drops into the bucket resulted in globules that could not be lifted back out. Instead, they broke apart and hardened at the bottom of the cup as unusable orphans. One drop from the syringe consistently worked well.

After a little practice on the workbench, it was time to use PlastiFix on the split canopy itself.

Scott Chastain 01-11-2018 12:34 AM

Applying PlastiFix
It was important for the borders of the repair to be masked off to prevent the liquid from needlessly tarnishing undamaged plexiglass. It was also important for the repair surface to be relatively level to allow the liquid to settle properly during cure.

With the syringe apparatus, both ends of the V-groove channel were first treated.

At that point, a more aggressive yet perfectly acceptable technique was used whereby the powder was carefully layered onto the bed of the V-groove and catalyzed with the liquid where it lay. With the syringe removed, the drops were much larger.

There was a short cure time of about ten minutes between layers. The process was repeated several times until the cured bonding agent uniformly filled the V-groove.

It was difficult to prevent a few bubbles from developing in the cured product. Fortunately, most bubbles rose to the surface before the mixture hardened and were popped with the syringe and filled on the subsequent layer. Others did not.

Following 4 iterations, the V-groove joint was filled just beyond capacity. A very slight 1/32Ē layer provided leeway to level the bonding agent to the plexiglass with a sanding block, the beginning of the final step in the repair process.

Scott Chastain 01-11-2018 12:37 AM

Leveling & Polishing
With the bonding agent masking removed, the process of leveling and polishing the damaged area of the canopy began.

The immediate area around the scar was again masked off and block sanded wet using 120-grit sandpaper. There were several areas where catalyzed material bled through the tape and onto the plexiglass. Those areas were also block sanded with 120-grit paper.

The canopy scar was then re-masked to begin the polishing process. Tape was applied in such a way that, for each progressively finer grit of paper used, about 1/4-inch of new perimeter was exposed to allow the polishing area to move outward in succession.

Grit progression for the first stages of polishing went from 120, 220, 400, 600, 800, 1,000, 1,500, to 2,000. Each stage was applied in a 90-degree crosshatch pattern relative to each previous grit to maximize polishing potential and to help prevent grooving.

Acrylic material being removed during the polishing process was evident on the floor of the shop.

Following the 2,000-grit stage, the canopy orientation was reversed so that block sanding the interior of the bubble could be accomplished.

Scott Chastain 01-11-2018 12:40 AM

Polishing Continues
The interior scar was masked off for sanding, but unlike the exterior portion, progressively finer grits were applied in 1/8-inch increments to help minimize optical distortion.

The scar was measured after each grit stage, and sandpaper was cut to width as block sanding continued. This was necessary to allow the sanding block and paper to clear the masking tape and make contact with the scar as it was sanded.

A shop rag nested inside the canopy absorbed water and was used to pick up abraded acrylic throughout the process.

As the polishing progressed, small occlusions became visible along the upper edges of the V-groove.

The cause of the occlusions was unknown. Great care was taken to clean the plastic prior to bonding. Nevertheless, whether or not the imperfections would develop into full-blown cracks was yet to be seen. The worst-case scenario would be a repeat of the repair process a second or even a third time, and if those attempts failed, a full canopy replacement would likely be in order. Thus, it was decided to allow the occlusions to remain for in-flight testing.

After block sanding both sides of the scar up to 2,000-grit, the final stage of polishing awaited a Headlight Magic kit delivery from UPS. So far, the repair appeared somewhat promising, and with the exception of the tiny occlusions noted earlier, the canopy was coming out better than expected.

Meanwhile, a replacement C-661 slider block was measured, drilled, and prepared for installation.

Scott Chastain 01-11-2018 12:44 AM

Final Repair Results
The final stage of polishing progressed from 2,000-grit to 4,000-grit to 12,000-grit, and then finally to an abrasive micro-grit polish supplied in the Headlight Magic kit from Polyvance.

The interior side of the canopy was completely polished first.

The canopy orientation was again reversed to complete the polishing stages on the exterior surface.

The polishing compound was rubbed into the surface until it began to disappear as a smudge across the plexiglass. It was then rinsed off with water and re-applied as necessary to problem areas.

The final result was a canopy that appeared at first glance undamaged, although some distortion was noticeable when the original scar was examined closely. The removal of material during the polishing process made such distortion unavoidable. Additionally, the scar was opaque and not by any means as clear as the surrounding acrylic. Nevertheless, the canopy appeared structurally sound and worthy of flight.

A view from the interior revealed a few areas where further application of polishing compound was needed.

The canopy was ready for installation and would be flight tested at altitude and under typical 3 to 4-G aerobatic maneuvering to stress the integrity of the bond. It was decided that, since the process was not nearly as labor-intensive or expensive as full canopy replacement, three attempts to repair the canopy would be a warranted option under the circumstances should flight testing cause the bond to fail. If nothing else, the process was educational and worth attempting.

I hoped it would prove successful in the end.

wjb 01-11-2018 07:17 AM

Great writeup Scott! All the minute detail is very informative. Looking forward to hearing how it stands up for your next 40 day cross-country trip!

9GT 01-11-2018 07:29 AM

Fantastic write-up and documentation of your repair Scott! A post that will get referenced and studied for years to come I am sure.

Saber25 01-11-2018 07:29 AM

Excellent repair description and helpful pictures. How many labor hours involved?

Every time I look at my 30 year old canopy I knock on was possibly the most difficult part of the construction.

Cheers, Hans

All times are GMT -6. The time now is 02:11 AM.