r26D

Locks and Remote Control

Preparing for the Move

The panels staying together just standing still is a given. What happens when we try to move them? In a truck. Downtown. Up elevators. The bottles are loosely held in the frame by the LED strands that are secured by the caps so from the front you can’t remove a bottle. If you pull real hard, you can probably break the wires and black out 50 LEDs at a time.

wpid-20140411-160107_BE7f_iPhone5-2-2014-04-16-22-22.jpg

Besides acting as lenses for the light, the bottles provide part of the structure. They aren’t very strong on their own but grouped together the panel is starting to get rigid. The final step for the bottles was figuring out a way to lock them in place. We looked at wire, wooden slats, and vinyl tubing. The vinyl tubing was cut into a 0.5” section. The section was sliced in one spot so the loop could be opened and slipped around the neck of the bottle extending through the back grid plane. The thickness of the vinyl tubing sat against the ring around neck where the cap sits and prevented the neck from slipping back through the plane. However, the vinyl was too soft to stay in place with the pressure holding the bottle.

The collar idea would work with a different material. After walking around the hardware store, we found a candidate in the plumbing section. Schedule 40 PVC pipe is the prefect inner diameter to fit the neck of the bottles. Using a miter saw with a blade for plastic (200 tooth), 0.5” sections were cut. Due to the stiffness of the PVC we cut out various size sections out of the ring until we found the right clamping force to hold the bottle but not strong enough to kill our hands trying to install them.

wpid-2014-04-1516.14.47-1-2014-04-16-22-22.jpg

We also experimented with the tools to cut the sections. Starting with snips they proved to be problematic because of the cutting action. When snapping through the second cut, the part and scrap would fly across the shop. The Milwaukee ProPex cutter proved the best gadget for the job since it sliced through the PVC in a more controlled manner.

wpid-2014-04-1516.18.58-1-2014-04-16-22-22.jpg

Once the bottle lock recipe was worked out, we then had to make 400 of them. Here are the first 180.

wpid-2014-04-1611.13.27-1-2014-04-16-22-22.jpg

Here’s a closeup of the bottle lock installed.

wpid-2014-04-1516.23.35-1-2014-04-16-22-22.jpg

LEDs on and bottles locked.

wpid-2014-04-1612.30.04-2014-04-16-22-22.jpg

With the first panel completed, the structure is much more rigid. Before locking the bottles we noticed some bowing outward of the grid planes. Now the panels are pulling in, exposing some of the label at the bottom of the bottle on the front of the panel. We have some 3” PVC pipe to cut cylinders to install in row 13 to maintain the distance of plywood grid planes.

Shhhh! Srsly.

The 12VDC Meanwell Power supplies we’ve used before have fans, but they only activate when the internal temperature reaches a set point. The 5VDC power supplies we’re using now have fans that run continuously. As the LEDs pull more power, the fan’s speed slows so the already annoying whir goes up and down in pitch that you can hear across the room. Not ideal for a conference room installation. To solve this problem we switched out the power supplies with a new model, Meanwell RSP-200-5, that are fanless/silent and more compact.

wpid-2014-04-1612.46.10-1-2014-04-16-22-22.jpg

How can someone over there change the active visual over here?

When we run LED projects they are usually near a computer or on the network and can be controlled. These panels are going into someone else’s building so we’re not comfortable letting this live on their network since Artnet and E1.31 are chatty. No one like it when you leave an alien artifact behind, especially in their code on on their network. Thanks to work already done for PixelController, there is a shortcut to try out. PixelController is setup to use OSC to receive commands. A TouchOSC layout was built for an iPad or Phone.

Using the TouchOSC Editor to sync a layout on an iPhone didn’t work. After Googling around, the layout did transfer to the iPhone but the TouchOSC App wouldn’t list it as an option because it wasn’t an iPhone layout type. The iPhone layout is much smaller that the phone layout we tried so we started touching it up and resizing it to fit on the screen. Here are a few of the screens.

wpid-2014-04-1622.17.44-2014-04-16-22-22.png

wpid-2014-04-1622.17.48-2014-04-16-22-22.png

wpid-2014-04-1622.18.00-2014-04-16-22-22.pngwpid-2014-04-1622.17.53-2014-04-16-22-22.png

Now that the buttons fit on the screen and can be seen (TouchOSC screens aren’t scrollable) we can use the TouchOSC App to control the panel. Tomorrow we will be completing a freestanding wireless network so remote control will be done.