Thought I would finally get around to showing off a Halloween project that was built for the 2018 Haunted Halloween Trail of Tears (link here). Those that checked out the trail walkthrough probably noticed a couple of interesting adds that I didn’t really talk much about. Was looking at my post queue and noticed I STILL had not posted on that project over a year later. This must be remedied immediately (as in before this year’s Halloween hits the history books). So without further delay I bring you Project Nightwings.
The night shots of the Nightwings didn’t come out the best because.. well, it was dark. Last year there was close to zero ambient light on the trail making it a bit difficult on the final camera round. Now, on the other hand, definitely upped the spook factor on the trail! If you look closely you will see a giant pair of wings adorning the back of the gargoyle. Now, making one pair of wings is fun enough, however, TWO wings pretty much puts this guy on cloud nine. The second pair is even harder to make out as the black wings blended into the darkness – unlike the zombie silhouette that looked awesome after an upgrade was made to the backing board that reflected the light better)
Hit jump to see how these new props for 2018 came to be!
Sure, pictures of the final project are cool, but if you are still reading this post then a) you like Halloween as much as we do and b) you want to know more about the build details. Before we go any further, full credit goes to Kraven Z. who posted work she had done on her mechanized wings on Youtube back in October 2015 (https://www.youtube.com/watch?v=eLzaqj-PQmk). I found that video while i was scouring the Internet looking for ideas on how to realize a concept I had been working on for several Halloween seasons. It is her linkage design and linear actuator concept that was at the heart of the project. I did make some linkage improvements and a few upgrades that fit my purposes better. Kraven built hers to wear – I was building stationary props. One thing Kraven didn’t provide details on was the linkage dimensions which turned out a lot more difficult than expected trying to figure that out from scratch. Before you ask, I am leaving those details off as well since I am not sure how she originally figured them out for her project.
First stop was the hardware store. Picked up some 4′ aluminum stock, 3′-4′ aluminum rods and the complete genius connectors that pose as electrical box lugs. Note, it took me forever to find those in the store since I didn’t have the real name for them and the store clerks didn’t know what mechanical wing bar connectors were – imagine that.
Having no idea how this was going to work, built a prototype rig to prove out the linkages. Grabbed a 1x board for the base and attached a linkage support board with some angle brackets. I have a large number of mending bars of different sizes (translate that to every size the hardware store carries) to help with linkage design. Note, this linkage ended up being so sensitive to the relative dimensions that I eventually gave up on the pre-drilled bars in the full build and cut new linkage test bars out of the aluminum stock – I put a hole every half inch in the new bar designers giving me full freedom to adjust as needed. After studying the Kraven video for hours, came up with the following prototype.
Problem was I didn’t have enough hands to move the linkage up, still check measurements and test the wing rod placements. Added a piece of PVC to simulate the future linear actuator movement. Attached a string to the bottom of the straight bar in the middle that when pulled would push the linkage up out of the PVC tube. Even put a hole in the tube so I could pin it up in the raised position. Pull the string and this is what you get when you attach the required 9 linkage bars. Take note that the support bars are attached on opposite sides of the bar to compensate for the downward V bars being gapped by the center bar. The top bars are pegged to the top of the support structure providing the critical pivot points for the upward movement of the center bar. Words of caution, make sure each side is IDENTICAL in their dimension or chaos will reign in the later parts of the build. Additionally, make sure the vertical bar stays straight as it travels up. Ended up adding a couple of L brackets at the base to keep the bar from moving off center.
With core linkage working, switched to aluminum stock that allowed me to put the wing bar connectors where they needed to be. This was the most frustrating part of the build. Test after test after test failed to work. Still a bit baffled on how Kraven got hers to work and not bind up as the wings were raised – either I am missing something or there was a design change not shown in her video. Most of the issues came from the bottom holders of the rods. The lug design had a hole on the flat side and then a set screw to hold the rods in place. Initially thought only the bottom of the rod was locked in place by the set screw with the upper one allowed to move freely – mistake. Both lugs are set tight to the rod. Problem is, the connection points are off center to the bars causing weird irregularities as it was raised up and down.
Decided to fix this before I was driven completely nuts. Simply sawed off the tabs on the lugs and then drilled a hole underneath the set screw. Countersunk the hole and then used tapered bolt heads so they stayed flush with the bottom of the holder. That allowed me to put the wing rods in and put their fix points centered to the linkage bars. Sooooo much better. You can also see the bottom rod attached to the lower bar.
Now for the other end of the wing bars. The initial configuration of the lugs worked well for the ends of the rods. They didn’t need to travel through so having the screw protruding through didn’t cause any issues. Now, the backside did have issues as the rod did need to pass through – simply used the same mod that used for the other end.
Okay, one side of the wing prototype complete. Are you getting the idea of just how large these wings are? Definitely bigger than my lab is able to handle.
Pull the string, raise the middle bar and magic, we have ourselves a working wing…well at least one side of it. This is where the trial and error comes in. Every dimension has an impact on the functionality of the wing. how close the lug connectors are on the bar stock impacts how far it extends up, the distance of the two lugs on the outer bar dictate how far the wing tips extend out. Go a bit too far or too little and the wing will either collapse or bind – you can already see some of the tension placed on the rods with the weight they are having to handle at the ends. Caution – KEEP YOUR FINGERS OUT FROM BETWEEN THE TWO LINKAGE BARS. If the set screws are not tight enough on the lugs you might end up putting the ends of your fingers in cooler and driving to the local ER – you have been warned.
Had to move out of the lab to put the other side on. Wanted to get the full effect, so took some black sheeting and quickly covered the wings. The plan was to use that as a template to eventually cut the cloth material for the final covering. Ended up starting over later in the build cycle. Finally get to see the fruits of my labor after 4 long months of working on it.
Thought I would give you reference scale – can you say huuuuuugggeeee!?!
With linkage prototype done, attention moved to the linear actuator. Picked up one off of Amazon that looked like it would do the trick. Also picked up a heavy duty switch so I could manually control the up and own movement of the actuator. Found a 12 volt power supply, added a few quick connects to bring in the switch (the manual switch would not be in the final design and wanted to be able to remove it easily).
Built a stronger stand for the wings as it was evident that the thin piece of wood I used for the prototype would not stand up to the weight and forces of the final product. If you look at the shot above, you will notice that the base of the actuator has a protrusion out the bottom. That prevented me from connecting it to the base directly. Added a block of wood that was laying around to offset for the protrusion and then locked it in with another angle bracket to keep it from shifting as it raised and lowered. Also added another angle bar towards the top to keep it upright. Ended up adding another angle bar to the side of the vertical piece of wood (see left side of wood straight up from the left cylinder on the actuator). That allowed me to screw it into the side for strength and then use the holes to put the zip ties through to hold the actuator in place. Not the prettiest frame I’ve built, but it sufficed. I also drew a line so I could make sure the actuator was raising straight up and provided a measurement point to affix the top linkage bars.
Next up was to build the final linkage bars – these must be precise between the two sides and the holes perfectly aligned or weird things start happening, It cost me several weeks tracking down an issue only to find out I had one of the holes a quarter inch off – ugh. Note, I had to drill bigger holes in ends of the V bars so they would fit over the pin in the top of the actuator.
Starting to finally come together!
I wanted maximum span on this set of wings so the outside rods actually extend past the base. Simple fix, just raise the base. Everything back together now with the new linkage bars and rods all in place.
I hit the actuator and had an oh shit moment as the wings kept raising up towards the ceiling. A huge sigh of relief when they eventually stopped about an inch before putting holes in my brand new ceiling tiles. Also became a bit dangerous walking around when they were up so stuck tape on the ends to make them easier to see.
Now that the hard part was figured out, decided a second set of wings was in order. This time I wanted a more ornate set similar to the ones Kraven had built – did sacrifice the span. Kept all the linkages exactly the same, but bent the rods in an elongated ‘S’ curve. I definitely underestimated how hard it was to build 4 rods in the exact same shape. Good luck on that.
Curve wings down….
…curved wings up!
Now time to build the electronics to control these bad boys. Had to bring my brother Ron in to help me on this part. I am okay around small circuits in the 5v DC range and comfortable in the 120V AC range. What I am not particularly good at is the space in between that. In this case I needed to switch on the 12V power to the linear actuator and then reverse the polarity to bring the actuator back down. I had the sensor and triggers built on the Arduino side but wasn’t familiar with the larger voltage relays. Picked up a 4 way relay (4 banger). Ron taught me how to use two of the relays to manage the up and down of the actuator based on the 5V signals from the Arduino. Since we had an extra relay, ended up controlling the gargoyle prop’s eyes as well. Thanks for the help Ron!
Okay, wing structure built, electronics built – all that is left is to cover the wings. Earlier in the week I dragged Linda to JoAnne’s Fabric to pick up some lightweight black material to cover the wings as I didn’t want to add more stress to the linkage than I had to. It was 2 am in the morning of the trail and it was getting to be crunch time to get my signature props done. A full year’s worth of work came down to the witching hour to complete. Ron came up with the idea of building a material template directly from the wings. Went back to the black plastic and taped it to one side of the wing. Took blue tape pieces and hand made a wing template.
Moved them around a bit until we got the look we wanted. This is the look of exhaustion, not to mention concern as I had already earned Linda’s wraith thanks to not being out of the basement yet and she still had work to do to get that ready for the guests that were due 13 hours from then.
Quickly went to work to cut out the pattern.
Laid out the new template on the fabric and cut out two wing coverings. Put carpet tape around the upper wing rods and then attached the fabric being careful not to impede the movement points.
Then did it all again for the curved wing. By this time Ron and I were spent. Wings were finally done and there was still time to get a couple hours of sleep before starting to build the trail – this time with two awesome mechanical wings sure to impress our guests.
Thought I would point out a key design change we made over the original Kraven version. With the large wingspan, there was a lot of stress being put on the linkage bars. This stress was causing strain on the actuator and when fully outstretched, the bars were twisting a bit causing some random behavior. Thought about it a bit and opted to double up the bars to strengthen the support and keep the top bars straight. Basically doubled up the fixed points between the bolts. This did the trick. Also recommend using lock nuts to keep the bars in place, but still slightly gapped to allow them to move freely.
Here is a better view of them as we were staging the large wings on the trail.
In closing, big thanks to Kraven for the original concept and to Ron for helping me get them all covered and wired up. They ended up working fantastic the entire night. Had some issues with the large wing this year and clearly need to work on a better way to store them during the off season. They basically sat in my basement all year much to the angst of Linda. Pretty sure I am not going to get away with it this year. Hope you enjoyed reading about the details of last year’s signature pieces.