My Gizmo--Update, Acid Test
Posted: Sat Dec 06, 2014 4:36 am
I’ve been designing and fabricating a gizmo in my spare time at work for the last several months. It's a tape applicator, used to adhere a double sided tape to aluminum and plastic bars used in framing of canvas prints. I finally finished it late yesterday, and put it in service this morning. My boss was so impressed he took it and me to the senior VP of the company to show it off. The leads, manager and director of the area I made it for all came and watched it work. There were lots of oohs and ahhhs.
I won’t bore you too much with technical details, but basically aluminum and plastic stretcher bars, used to assemble canvas prints (rather like how artist canvas is attached to conventional wooden stretcher bars), are assembled, and then an incredibly sticky double sided tape is laid along the outside edges and inside diagonal corners of the bars. The tape is nasty stuff to work with, and sticks to nearly everything, including the assemblers. The tape had been applied to the bars by hand, which is laborious, messy, and time consuming. My task was to design and build a machine that would apply the tape automatically. It does so, and it works rather well. The real trick was to establish the spacing between bars as they travel through the machine—the three proximity sensors (little black rectangles along the front side) sense the bars, start and stop the machine at certain times, and establish the spacing between bars along the running tape. The bars space at twice the length of the diagonals on the bar ends. There’s a small guillotine type blade, hidden inside the machine, that cuts the midpoint of the tape between pairs of bar ends, thus leaving just enough tape at each bar end to be folded back along the diagonal faces.
The machine’s various functions are controlled by a PLC (programmable logic controller), which receives inputs from the sensors, and turns the drive gearmotor on and off, and also fires the cutter. The cutter is powered by a small pneumatic cylinder via a solenoid valve, run off the house compressed air supply. The PLC, along with AC/DC voltage supply and DC gearmotor, are visible on the back of the machine. The bars are driven through the machine by pairs of polyurethane rollers either driven directly by the gearmotor, or via a timing belt (visible on the front), or gears (visible in the middle, from the ends or back). The electric motor is quite small; the gear reduction allows it to drive the machine without much load on the system.
When I began the project, I looked into commercial machines that would do the job. Though there are all sorts of tape applicator machines around, none were capable of dealing with the incredibly sticky, nasty tape we use. I talked with one company that thought one of their machines MIGHT work, but had no guarantee of it. Their machine was $15,000. I built my machine for less than $400, using mostly scrap metal and stuff we had on hand, such as the aluminum frame, PLC, etc. The gears and timing pulleys are plastic, the cheapest that could be found. The 12V DC gearmotor is the cheapest I could find, ~$50. The machine should eliminate three or four unskilled laborers doing it by hand, and do it several times as fast, with better quality. The bosses all have stars in their eyes—the VP mentioned passing it on to our corporate owners. Oh boy. Will I see any monetary reward for it? I highly doubt it. Since I made it for the company, they own any rights to it.
The aluminum front and rear frames are 5 1/8"x1/2"x12" long (scrap we had on hand), spaced 1 1/2" apart, just a hair wider than the bars. The overall footprint is 12"x12" or so. This is just a prototype, it’s rough around the edges. The clear Lexan covers in particular don’t look too hot—I’d never heat formed Lexan before, and struggled with it a bit. I'm waiting on some special adhesive to glue the edges together. Overall it looks pretty cool, though, and it does what it was meant to do. I put a lot of thought and time into this thing. After this prototype, I may wind up doing a Mk II design. The VP said they may want a bunch of them made.
Pics follow: several views of the front, sides and back. Later pics show the bars before and after assembly, then before and after taping. Pic 1114 shows that the tape must be manually applied to the nose end of the first bar, then the bar is inserted into the machine. Thereafter all that the operator needs to do is feed assembled bars into the machine, the taping and cutting happens automatically. The operator catches the bars as they exit the machine, then must fold the tape ends along the diagonals. Taped bars of various lengths are shown in 1118—the machine is insensitive to bar length. They do 12”, 16" and 20” bars, along with 30” and 40” bars for larger prints. They may go to bars as long as 60” in the near future. The machine will spit out a 12" bar in seven or eight seconds, and keep doing it as long as the operator keeps feeding bars into it.
I won’t bore you too much with technical details, but basically aluminum and plastic stretcher bars, used to assemble canvas prints (rather like how artist canvas is attached to conventional wooden stretcher bars), are assembled, and then an incredibly sticky double sided tape is laid along the outside edges and inside diagonal corners of the bars. The tape is nasty stuff to work with, and sticks to nearly everything, including the assemblers. The tape had been applied to the bars by hand, which is laborious, messy, and time consuming. My task was to design and build a machine that would apply the tape automatically. It does so, and it works rather well. The real trick was to establish the spacing between bars as they travel through the machine—the three proximity sensors (little black rectangles along the front side) sense the bars, start and stop the machine at certain times, and establish the spacing between bars along the running tape. The bars space at twice the length of the diagonals on the bar ends. There’s a small guillotine type blade, hidden inside the machine, that cuts the midpoint of the tape between pairs of bar ends, thus leaving just enough tape at each bar end to be folded back along the diagonal faces.
The machine’s various functions are controlled by a PLC (programmable logic controller), which receives inputs from the sensors, and turns the drive gearmotor on and off, and also fires the cutter. The cutter is powered by a small pneumatic cylinder via a solenoid valve, run off the house compressed air supply. The PLC, along with AC/DC voltage supply and DC gearmotor, are visible on the back of the machine. The bars are driven through the machine by pairs of polyurethane rollers either driven directly by the gearmotor, or via a timing belt (visible on the front), or gears (visible in the middle, from the ends or back). The electric motor is quite small; the gear reduction allows it to drive the machine without much load on the system.
When I began the project, I looked into commercial machines that would do the job. Though there are all sorts of tape applicator machines around, none were capable of dealing with the incredibly sticky, nasty tape we use. I talked with one company that thought one of their machines MIGHT work, but had no guarantee of it. Their machine was $15,000. I built my machine for less than $400, using mostly scrap metal and stuff we had on hand, such as the aluminum frame, PLC, etc. The gears and timing pulleys are plastic, the cheapest that could be found. The 12V DC gearmotor is the cheapest I could find, ~$50. The machine should eliminate three or four unskilled laborers doing it by hand, and do it several times as fast, with better quality. The bosses all have stars in their eyes—the VP mentioned passing it on to our corporate owners. Oh boy. Will I see any monetary reward for it? I highly doubt it. Since I made it for the company, they own any rights to it.
The aluminum front and rear frames are 5 1/8"x1/2"x12" long (scrap we had on hand), spaced 1 1/2" apart, just a hair wider than the bars. The overall footprint is 12"x12" or so. This is just a prototype, it’s rough around the edges. The clear Lexan covers in particular don’t look too hot—I’d never heat formed Lexan before, and struggled with it a bit. I'm waiting on some special adhesive to glue the edges together. Overall it looks pretty cool, though, and it does what it was meant to do. I put a lot of thought and time into this thing. After this prototype, I may wind up doing a Mk II design. The VP said they may want a bunch of them made.
Pics follow: several views of the front, sides and back. Later pics show the bars before and after assembly, then before and after taping. Pic 1114 shows that the tape must be manually applied to the nose end of the first bar, then the bar is inserted into the machine. Thereafter all that the operator needs to do is feed assembled bars into the machine, the taping and cutting happens automatically. The operator catches the bars as they exit the machine, then must fold the tape ends along the diagonals. Taped bars of various lengths are shown in 1118—the machine is insensitive to bar length. They do 12”, 16" and 20” bars, along with 30” and 40” bars for larger prints. They may go to bars as long as 60” in the near future. The machine will spit out a 12" bar in seven or eight seconds, and keep doing it as long as the operator keeps feeding bars into it.