Difference between revisions of "Beacon4"

From Hackstrich
(Status updates.)
 
(11 intermediate revisions by the same user not shown)
Line 2: Line 2:
  
 
= Project Status  =
 
= Project Status  =
 +
* 2013-09: Project assembled (with LED strips hung from a frame) at Mini Maker Faire 2013 as a backdrop for the HackLab.TO table. Mostly worked, but had one IO-ETH module fail, need to respin this without any QFNs on it to resolve reliability issues. Used ArtnetForProcessing instead of LightJams, and had no software issues this time.
 +
* 2012-11-24: Wrote up lessons learned and the log of what happened on-playa.  Will still continue to look into the tech issues we had scalability-wise, but other than that Beacon4 is considered to be complete.  The [[StrichLux]] hardware that was used will be reused next year for [[InvisibleMaze]].
 +
* BRC2012: Project went up (partly) at Burning Man 2012, many lessons were learned, and it looked cool.  See Lessons Learned section for more details.
 +
* 2012-08-21: Fixed StrichLux C board set (bad soldering on a level shifter on an IO-SPI shorting out a power rail), found the slow refresh rate bug was actually VMware Fusion dropping UDP packets on my laptop, won't be an issue on the real thing as there's no VMware.
 +
* 2012-08-20: Shipped all cabling, all LED strips (except the 1m test strip S33), StrichLux A and B, and the PC box w/ PC.  StrichLux enclosure C shipped but the board set is still here to try to get it fixed/working properly, and to try to fix the slow refresh rate bug.  Woot!
 +
* 2012-08-19: Another 12h workday.  All cabling terminated, all boards populated, all PWR-DC5s working well enough to use.  StrichLux C's CORE board seems to not work, as the system sends no data out.  Partial system was set up with StrichLux A and B and sort of worked, excep the refresh issue on higher channel numbers is still there and makes 3/4 almost useless.  Supposed to ship tomorrow, but I don't see how it can.
 +
* 2012-08-18: Epic workday.  All PWR-DC5s done (but one channel on one not working yet), connectors added to the 2 COREs that didn't have them and all IO-SPIs that didn't have them, all feeders, jumpers, and strips labeled, all feeders except two connectorized, power feeders purchased and connectorized.
 +
* 2012-08-17: Signal wiring terminated, but not yet inserted into the Mini-Fit plastic shells.
 +
* 2012-08-15: Foam cut to fit, Sabres terminated.
 +
* 2012-08-14: All in-box high power connectors are terminated except Sabre-side of the inputs.  Tomorrow will get signal wiring terminated and all wiring installed in boxes.
 +
* 2012-08-13: All connectors mounted on the 3 StrichLux boxes.  Wiring of B and C still needs to be done.
 
* 2012-08-12: All LED strip now cut/terminated/sealed.  Test of StrichLux A channels 1-3 are working mostly, though each higher numbered channel seems to refresh slower than the one before it.  More troubleshooting will need to be done.
 
* 2012-08-12: All LED strip now cut/terminated/sealed.  Test of StrichLux A channels 1-3 are working mostly, though each higher numbered channel seems to refresh slower than the one before it.  More troubleshooting will need to be done.
 
* 2012-08-11: StrichLux A enclosure has all connectors mounted and wired up (except only 1/2 of the power connector so far).
 
* 2012-08-11: StrichLux A enclosure has all connectors mounted and wired up (except only 1/2 of the power connector so far).
Line 27: Line 38:
 
*2012-04-23: Requested permission from NeonBunny to attach LEDs to the tower.  
 
*2012-04-23: Requested permission from NeonBunny to attach LEDs to the tower.  
 
*2012-04-22: Started putting together idea.
 
*2012-04-22: Started putting together idea.
 +
 +
= On-Playa =
 +
== Overview ==
 +
* Project took many days of work to set up, and only ever got about 2/3 of it put up
 +
** Setup took a lot more time away from enjoying BRC than I would have liked, need to emphasize quick setup time more next time
 +
* Channels 3/4 on each control box were essentially useless due to the lighting control software not scaling
 +
** Resulted in only having 6 total channels available (2 from each of the 3 boxes)
 +
* Same MAC address was accidently programmed into each [[StrichLux/IO-ETH|IO-ETH]], meaning that whatever was sent to channel X on one box went to channel X on all boxes (Art-Net is UDP broadcast)
 +
** Resulted in having only two channels of unique data total, each duplicated 3 times
 +
** Had everything we needed to reprogram the boards (as an issue like this was thought to be likely) but the diagnostic PC wouldn't talk to the PICkit3
 +
* Only one cabling issue, data/clock were backwards on one cable
 +
** Given the other issues, this segment was abandoned and others were connected instead
 +
* The parts that went up still looked cool, and got a lot of positive comments
 +
 +
== Lessons Learned ==
 +
* Way too much running around/stress/work at the last minute before it shipped
 +
** '''Lesson: Start earlier, set milestones and scale the project back/put it off until the next year if they aren't met'''
 +
* Shipping was a nightmare
 +
** Took longer than expected/guaranteed due to customs, project almost didn't get there before we departed for BRC
 +
** Cost over $1000 all together as tax/duty had to be paid both ways
 +
** '''Lesson: Ship much earlier (ideally beginning of August) and consider shipping with another large local project or the Big Ole' Truck'''
 +
** '''Lesson: Ship as Strich Labs under the program designed for shipping things down for demos that will then return'''
 +
* Lack of good documentation meant that the setup work couldn't be efficiently parallelized
 +
** '''Lesson: Put more of a priority on documentation that others can read/understand to help set the project up'''
 +
* Amount of cabling required to be sorted out/pulled/secured made the project take a huge amount of time to set up
 +
** Documentation would have helped parallelize setup if it was available, which would have made it a bit less painful
 +
** '''Lesson: Minimize cabling wherever possible by having one part attach directly to the previous part'''
 +
** '''Lesson: Minimize the number of connections between parts where possible'''
 +
* Setup diagram lead to having cables running down one vertical riser, then plugging into a box that wasn't there.
 +
** This was noted earlier on, but never recorded and was forgotten about.
 +
** '''Lesson: Document all issues found/thought about on a wiki page immediately, so they're not missed later'''
 +
* Some of the setup was near-impossible due to needing to put LED strips up where there was nothing to stand on
 +
** Resulted in a lot of overly-dangerous leaning/hanging with one hand while trying to velcro up strips with the other
 +
** '''Lesson: Think about how the project will be put up, not just the final state it will be in'''
 +
* Same MAC address was programmed into each board
 +
** Tried to fix it on-playa, but the troubleshooting/programming laptop wouldn't talk to the PICkit3 that was with us
 +
** '''Lesson: Leave more time to test hardware before it ships'''
 +
** '''Lesson: Test all troubleshooting/programming equipment that is needed before leaving for BRC'''
 +
* Lighting control software didn't scale properly, only ran two universes well per box
 +
** '''Lesson: Again, leave more time to test everything well'''
 +
** '''Lesson: Minimize software that isn't under directly under our control'''
 +
* Fit-PC 2i worked as-expected other than producing more heat than we had figured
 +
** Needed to leave the Pelican case with the control PC in it slightly open during operation
 +
** '''Lesson: Need to do better thermal calculations/testing on any high-power equipment in the future'''
 +
* Toughbook CF-U1 UMPC worked great, no issues with it other than needing one recharge during burn
 +
** Survived rain and being covered in playa with no issues at all
 +
** Wouldn't have needed recharging at all if things went better
 +
** Fluke 28 also lived up to expectations, no issues at all despite getting wet and covered in playa
 +
** '''Lesson: Reliable troubleshooting equipment is really important!  This was totally worth the research/money put into it.'''
  
 
= Outstanding Issues  =
 
= Outstanding Issues  =
Line 631: Line 691:
 
== Feeders and Jumpers ==
 
== Feeders and Jumpers ==
 
* Feeders
 
* Feeders
−
** Feeder A-1 - 310cm
+
** Feeder A-1 - 310cm (F01)
−
** Feeder A-2 - 310cm
+
** Feeder A-2 - 310cm (F02)
−
** Feeder A-3 - 310cm
+
** Feeder A-3 - 310cm (F03)
−
** Feeder A-4 - 310cm
+
** Feeder A-4 - 310cm (F04)
−
** Feeder B-1 - 360cm
+
** Feeder B-1 - 360cm (F05)
−
** Feeder B-2 - 360cm
+
** Feeder B-2 - 360cm (F06)
−
** Feeder B-3 - 360cm
+
** Feeder B-3 - 360cm (F07)
−
** Feeder B-4 - 711cm
+
** Feeder B-4 - 711cm (F08)
−
** Feeder C-1 - 576cm
+
** Feeder C-1 - 576cm (F09)
−
** Feeder C-2 - 576cm
+
** Feeder C-2 - 576cm (F10)
−
** Feeder C-3 - 516cm
+
** Feeder C-3 - 516cm (F11)
−
** Feeder C-4 - 687cm
+
** Feeder C-4 - 687cm (F12)
 
* Jumpers
 
* Jumpers
−
** Jumper A-1 - 308cm
+
** Jumper A-1 - 308cm (J01)
−
** Jumper A-2 - 262cm
+
** Jumper A-2 - 262cm (J02)
−
** Jumper A-3 - 308cm
+
** Jumper A-3 - 308cm (J03)
−
** Jumper A-4 - 262cm
+
** Jumper A-4 - 262cm (J04)
−
** Jumper B-3 - 92cm
+
** Jumper B-3 - 92cm (J05)
−
** Jumper B-4 - 152cm
+
** Jumper B-4 - 152cm (J06)
−
** Jumper C-1.1 - 73cm
+
** Jumper C-1.1 - 73cm (J07)
−
** Jumper C-1.2 - 142cm
+
** Jumper C-1.2 - 142cm (J08)
−
** Jumper C-2.1 - 73cm
+
** Jumper C-2.1 - 73cm (J09)
−
** Jumper C-2.2 - 142cm
+
** Jumper C-2.2 - 142cm (J10)
−
** Jumper C-3 - 92cm
+
** Jumper C-3 - 92cm (J11)
−
* Pinout
+
 
−
** Pin 1 - White - Ground
+
=== Pinouts ===
−
** Pin 2 - Black - +5V
+
==== Power+Data Cables ====
−
** Pin 3 - Red - Data
+
{| class="wikitable sortable" border="1"
−
** Pin 4 - Green - Clock
+
|-
 +
! Signal
 +
! Wiring Colour
 +
! MX150L Pin
 +
! RT360 Pin
 +
|-
 +
| GND
 +
| White
 +
| 1
 +
| A
 +
|-
 +
| +5V
 +
| Black
 +
| 2
 +
| B
 +
|-
 +
| Data
 +
| Red
 +
| 3
 +
| 1
 +
|-
 +
| Clock
 +
| Green
 +
| 4
 +
| 2
 +
|}
 +
 
 +
==== Power Only Cables ====
 +
{| class="wikitable sortable" border="1"
 +
|-
 +
! Signal
 +
! Wiring Colour
 +
! RT360 Pin
 +
|-
 +
| Bus A+B Negative
 +
| Black
 +
| A
 +
|-
 +
| Bus A Positive
 +
| Green
 +
| B
 +
|-
 +
| N/C
 +
| N/C
 +
| C
 +
|-
 +
| Bus B Positive
 +
| White
 +
| D
 +
|}
  
 
= StrichLux Enclosures  =
 
= StrichLux Enclosures  =
Line 889: Line 998:
 
|-
 
|-
 
| 12 sets of StrichLux IO-SPI Parts
 
| 12 sets of StrichLux IO-SPI Parts
−
| Future
+
| Obtained
−
| align=right | ~$165.00
+
| align=right | $190.39
 
|-
 
|-
 
| Enclosure A
 
| Enclosure A
Line 927: Line 1,036:
 
| Obtained
 
| Obtained
 
| align=right | $113.00
 
| align=right | $113.00
 +
|-
 +
| Shipping (outbound)
 +
| Done
 +
| align=right | $455.92
 +
|-
 +
| Tax/Duty (outbound)
 +
| Done
 +
| align=right | $104.40
 +
|-
 +
| Shipping (return)
 +
| Done
 +
| align=right | $224.29
 +
|-
 +
| Tax/Duty (return)
 +
| Done
 +
| align=right | $443.06
 
|-
 
|-
 
| '''Total'''
 
| '''Total'''
 
| '''Ordered'''
 
| '''Ordered'''
−
| align=right | $5250.84
+
| align=right | $6668.90
 
|-
 
|-
 
| '''Total'''
 
| '''Total'''
 
| '''Future'''
 
| '''Future'''
−
| align=right | ~$346.45
+
| align=right | $0.00
 
|-
 
|-
 
| '''Grand Total'''
 
| '''Grand Total'''
 
|  
 
|  
−
| align=right | ~$5597.29
+
| align=right | $6668.90
 +
|-
 +
| '''Hours Spent'''
 +
|
 +
| align=right | 577
 
|}
 
|}
  
Line 946: Line 1,075:
 
</gallery>
 
</gallery>
  
−
[[Category:Current_Projects]]
+
[[Category:Completed Projects]]

Latest revision as of 00:23, 27 September 2013

Beacon4 will be the fourth incarnation of the flagpole beacon (after Beacon2 and Beacon3), for Burning Man 2012. This year the plan is to take a different approach, and light the entire scaffold tower rather than just a beacon on top of the flagpole.

Project Status

  • 2013-09: Project assembled (with LED strips hung from a frame) at Mini Maker Faire 2013 as a backdrop for the HackLab.TO table. Mostly worked, but had one IO-ETH module fail, need to respin this without any QFNs on it to resolve reliability issues. Used ArtnetForProcessing instead of LightJams, and had no software issues this time.
  • 2012-11-24: Wrote up lessons learned and the log of what happened on-playa. Will still continue to look into the tech issues we had scalability-wise, but other than that Beacon4 is considered to be complete. The StrichLux hardware that was used will be reused next year for InvisibleMaze.
  • BRC2012: Project went up (partly) at Burning Man 2012, many lessons were learned, and it looked cool. See Lessons Learned section for more details.
  • 2012-08-21: Fixed StrichLux C board set (bad soldering on a level shifter on an IO-SPI shorting out a power rail), found the slow refresh rate bug was actually VMware Fusion dropping UDP packets on my laptop, won't be an issue on the real thing as there's no VMware.
  • 2012-08-20: Shipped all cabling, all LED strips (except the 1m test strip S33), StrichLux A and B, and the PC box w/ PC. StrichLux enclosure C shipped but the board set is still here to try to get it fixed/working properly, and to try to fix the slow refresh rate bug. Woot!
  • 2012-08-19: Another 12h workday. All cabling terminated, all boards populated, all PWR-DC5s working well enough to use. StrichLux C's CORE board seems to not work, as the system sends no data out. Partial system was set up with StrichLux A and B and sort of worked, excep the refresh issue on higher channel numbers is still there and makes 3/4 almost useless. Supposed to ship tomorrow, but I don't see how it can.
  • 2012-08-18: Epic workday. All PWR-DC5s done (but one channel on one not working yet), connectors added to the 2 COREs that didn't have them and all IO-SPIs that didn't have them, all feeders, jumpers, and strips labeled, all feeders except two connectorized, power feeders purchased and connectorized.
  • 2012-08-17: Signal wiring terminated, but not yet inserted into the Mini-Fit plastic shells.
  • 2012-08-15: Foam cut to fit, Sabres terminated.
  • 2012-08-14: All in-box high power connectors are terminated except Sabre-side of the inputs. Tomorrow will get signal wiring terminated and all wiring installed in boxes.
  • 2012-08-13: All connectors mounted on the 3 StrichLux boxes. Wiring of B and C still needs to be done.
  • 2012-08-12: All LED strip now cut/terminated/sealed. Test of StrichLux A channels 1-3 are working mostly, though each higher numbered channel seems to refresh slower than the one before it. More troubleshooting will need to be done.
  • 2012-08-11: StrichLux A enclosure has all connectors mounted and wired up (except only 1/2 of the power connector so far).
  • 2012-08-06: 35m of LED strip cut/terminated/sealed.
  • 2012-08-05: Reels A-I marked for cutting, so far all measurements add up correctly. Plan for tomorrow is to pick up StrichLux C enclosure, customize it, and get all LED strips terminated.
  • 2012-08-04: All LED strip wire cut, terminated, and inserted into connectors. StrichLux B enclosure has all connectors drilled. Thermal tests failed running the PC inside a box, will need to add an external radiator.
  • 2012-07-29: All feeders and jumpers now have female MX150Ls, and StrichLux A enclosure has output connectors drilled. Plan for tomorrow is to get the LED strip wire (832 inches total, ideally 208 inches/18 feet each of 4 colours) and get it cut into 4 inch pieces.
  • 2012-07-28: 9 feeders connectorized with MX150Ls, power and Ethernet jacks installed in StrichLux A enclosure.
  • 2012-07-25: All feeders and jumpers cut to length.
  • 2012-07-24: Finalized all strip IDs and created an empty project in LightJams that has all fixtures added in their correct places. Built the other two required Core Boards and they power/program OK.
  • 2012-07-23: Work on the installation helping system (BITS) is coming along, currently can scan a part and get a 3D model of where it goes. Inter-strip connectors, bulkhead M12 connectors, and Ethernet switch arrived. Cable ordered.
  • 2012-07-19: fitPC is here!
  • 2012-07-15: Ordered inter-strip connectors and remaining PWR-DC5, CORE, and IO-ETH parts.
  • 2012-07-14: First successful end-to-end test of the whole StrichLux system driving 5m of LED strip from the PWR-DC5.
  • 2012-07-11: Ordered the switch.
  • 2012-07-09: Ordered the majority of enclosure connectors.
  • 2012-07-07: First functioning test of StrichLux system driving 5m of LED strip.
  • 2012-06-24: LED strips all obtained. Lighting design continues, as does firmware development.
  • 2012-06-17: Hardware bring-up and firmware writing continuing. Lightjams is almost certainly going to be the software of choice. Lighting design is coming along.
  • 2012-06-07: Work on StrichLux continues. Evaluated lots of software, Lightjams is looking the best at this point, though would require Windows.
  • 2012-05-27: Completed rev. 1 StrichLux system design of all modules required for this project and sent all boards off for manufacturing.
  • 2012-04-30: Started planning the power architecture for the project.
  • 2012-04-29: Work continuing on IO-DMX module.
  • 2012-04-24: Permission received!
  • 2012-04-23: Requested permission from NeonBunny to attach LEDs to the tower.
  • 2012-04-22: Started putting together idea.

On-Playa

Overview

  • Project took many days of work to set up, and only ever got about 2/3 of it put up
    • Setup took a lot more time away from enjoying BRC than I would have liked, need to emphasize quick setup time more next time
  • Channels 3/4 on each control box were essentially useless due to the lighting control software not scaling
    • Resulted in only having 6 total channels available (2 from each of the 3 boxes)
  • Same MAC address was accidently programmed into each IO-ETH, meaning that whatever was sent to channel X on one box went to channel X on all boxes (Art-Net is UDP broadcast)
    • Resulted in having only two channels of unique data total, each duplicated 3 times
    • Had everything we needed to reprogram the boards (as an issue like this was thought to be likely) but the diagnostic PC wouldn't talk to the PICkit3
  • Only one cabling issue, data/clock were backwards on one cable
    • Given the other issues, this segment was abandoned and others were connected instead
  • The parts that went up still looked cool, and got a lot of positive comments

Lessons Learned

  • Way too much running around/stress/work at the last minute before it shipped
    • Lesson: Start earlier, set milestones and scale the project back/put it off until the next year if they aren't met
  • Shipping was a nightmare
    • Took longer than expected/guaranteed due to customs, project almost didn't get there before we departed for BRC
    • Cost over $1000 all together as tax/duty had to be paid both ways
    • Lesson: Ship much earlier (ideally beginning of August) and consider shipping with another large local project or the Big Ole' Truck
    • Lesson: Ship as Strich Labs under the program designed for shipping things down for demos that will then return
  • Lack of good documentation meant that the setup work couldn't be efficiently parallelized
    • Lesson: Put more of a priority on documentation that others can read/understand to help set the project up
  • Amount of cabling required to be sorted out/pulled/secured made the project take a huge amount of time to set up
    • Documentation would have helped parallelize setup if it was available, which would have made it a bit less painful
    • Lesson: Minimize cabling wherever possible by having one part attach directly to the previous part
    • Lesson: Minimize the number of connections between parts where possible
  • Setup diagram lead to having cables running down one vertical riser, then plugging into a box that wasn't there.
    • This was noted earlier on, but never recorded and was forgotten about.
    • Lesson: Document all issues found/thought about on a wiki page immediately, so they're not missed later
  • Some of the setup was near-impossible due to needing to put LED strips up where there was nothing to stand on
    • Resulted in a lot of overly-dangerous leaning/hanging with one hand while trying to velcro up strips with the other
    • Lesson: Think about how the project will be put up, not just the final state it will be in
  • Same MAC address was programmed into each board
    • Tried to fix it on-playa, but the troubleshooting/programming laptop wouldn't talk to the PICkit3 that was with us
    • Lesson: Leave more time to test hardware before it ships
    • Lesson: Test all troubleshooting/programming equipment that is needed before leaving for BRC
  • Lighting control software didn't scale properly, only ran two universes well per box
    • Lesson: Again, leave more time to test everything well
    • Lesson: Minimize software that isn't under directly under our control
  • Fit-PC 2i worked as-expected other than producing more heat than we had figured
    • Needed to leave the Pelican case with the control PC in it slightly open during operation
    • Lesson: Need to do better thermal calculations/testing on any high-power equipment in the future
  • Toughbook CF-U1 UMPC worked great, no issues with it other than needing one recharge during burn
    • Survived rain and being covered in playa with no issues at all
    • Wouldn't have needed recharging at all if things went better
    • Fluke 28 also lived up to expectations, no issues at all despite getting wet and covered in playa
    • Lesson: Reliable troubleshooting equipment is really important! This was totally worth the research/money put into it.

Outstanding Issues

  • How do we attach strips to toe boards?
    • Can't ziptie/cable tie like we do everywhere else as the toe boards are wide

Ideas/Plans

  • Main idea is to cover most of the scaffold tower in addressable LED strips, driven by multiple StrichLux modules
  • Art-Net will be used between the controller and the StrichLux box
  • Fit-PC 2i is being looked at to control the whole thing
    • As nice as ARM would be power-wise, x86 will be much more flexible software-wise
    • Need to make sure it will be powerful enough before we order
  • Need to find a fanless very low power 4 or 5 port Ethernet switch
    • Allied Telesis AT-FS705LE V5 might work, it's rugged, low power (1.5-2.5W), and cheap
      • But it takes 7.5V which would be another regulator and is a potential point of failure that would bring the entire system down
    • Alternative option is the GarrettCom S14H-Hi, it takes 8-15V in via screw terminals and is ruggedized, but $140 and draws 4W
    • Best option may be the Sixnet SL-5ES-1, it's $100, rugged, takes 10-30VDC via screw terminals, and only draws 2W
    • Ended up going with an M12 ethernet switch, the Octopus 5TX EEC

System Components

  • Fit-PC 2i (controller)
  • Octopus 5TX EEC 5-port Ethernet switch
  • 3 StrichLux boxes (driver), each with:
  • 60m of LED strips
  • Power setup
  • Debugging terminal (XO-1? Toughbook?)

Power Details

  • Power requirements (peak)
    • Fit-PC 2i - ???
    • Ethernet Switch - ???
    • StrichLux box(es) - ???
    • LEDs - 10W/m (~1.8A/m at 5V measured, so 10W/m after losses is reasonable)

Power Architecture

http://www.asciiflow.com/#4007824742034929919/13747898[YOB-2]
                                                                              +--------------+
                                                                         +--->| LED Strip  0 |
                                                                         |    +--------------+
                                                                         |
                                                                         |    +--------------+
                                                       +---------------+ | +->| LED Strip  1 |
                                +--------------+       |               +-+ |  +--------------+
                                |              |       |   StrichLux   +---+
                                |     Dome     |    +->| Controller A  +---+  +--------------+
                                |   Lighting   |    |  |               +-+ +->| LED Strip  2 |
                                |              |    |  +---------------+ |    +--------------+
                                +--------------+    |                    |
                                        ^           |                    |    +--------------+
                                        |           |                    +--->| LED Strip  3 |
                             +----------+---------+ |                         +--------------+
                             |                    | |
                             |                    | |                         +--------------+
                             |                    +-+                    +--->| LED Strip  4 |
                             |                    |                      |    +--------------+
                             |                    |                      |
                             |                    |                      |    +--------------+
                             |                    |    +---------------+ | +->| LED Strip  5 |
    +---------------+        |                    |    |               +-+ |  +--------------+
    |               |        |    Distribution    |    |   StrichLux   +---+
    |    Battery    +=======>|                    +--->| Controller B  +---+  +--------------+
    |               |        |        Box         |    |               +-+ +->| LED Strip  6 |
    +---------------+        |                    |    +---------------+ |    +--------------+
                             |                    |                      |
                             |                    |                      |    +--------------+
                             |                    |                      +--->| LED Strip  7 |
                             |                    |                           +--------------+
                             |                    +-+
                             |                    | |                         +--------------+
                             +--+-------+---------+ |                    +--->| LED Strip  8 |
                                |       |           |                    |    +--------------+
                      +---------+       |           |                    |
                      v                 v           |                    |    +--------------+
              +--------------+   +--------------+   |  +---------------+ | +->| LED Strip  9 |
              |              |   |              |   |  |               +-+ |  +--------------+
              | Convienience |   |    System    |   |  |   StrichLux   +---+
              |   Outlets    |   |  Controller  |   +->| Controller C  +---+  +--------------+
              |              |   |              |      |               |   +->| LED Strip 10 |
              +--------------+   +--------------+      +---------------+      +--------------+

Lighting Layout

Random Notes

  • Decided on 60m length total (3 StrichLux boxes)
    • That's 1920 RGB LEDs and ~600W peak!
    • Will need to come up with a lighting layout/design that will fit within this

Dimensions

Main Module (4 of these vertically, 2 of which will be lit)
-----------------------------------------------------------

+-----D-----+------E------+       |\                                   /|
|           |             |       |    \                           /    |
|-----D-----|             |       |        F                   /        |
|           H             |       |            \           /            |
|-----D-----|             |       |                \   /                |
|           |             |       |                /   \                |
|-----D-----+------E------|       |            /           \            |
|                         |       |        F                   \        |
A                         A       A    /                           \    A
|-----------B-------------+       |/                                   \|


Top Module
----------

|                         |       |                                     |
|                         |       |                                     |
|                         |       |                                     |
|                         |       |                                     |
|-------B-----------------|       |------------------C------------------|
|            |            |       |  /                               \  |
|            G            |       |------------------C------------------|
|            |            |       |                                     |
A============B============A       A                                     A
|                         |       |                                     |
  • A - Module Height (5ft 1in)
  • B - Module Width (5ft)
  • C - Module Depth (7ft)
  • D - Ladder Width (2.5ft)
  • E - Beside-Ladder Width (B - D) (2.5ft)
  • F - Cross-brace Length (7.37ft)
  • G - Top Rail Support Height
  • H - Ladder Section Height (3.5ft)
  • n - Number of Modules (3)
  • x - Number of Main Modules (2)

Idealized Layout

  • Vertical Mains: A * 4 * n = 5.083 * 4 * 3 = 61ft (done piecing)
  • Beside-Ladder Boxes: (E * 2 + H) * 2 * x = 34ft (done piecing)
  • Cross-braces: F * 2 * 2 * x = 7 * 2 * 2 * 2 = 59ft (done piecing)
  • Top Rail: B * 2 + C * 2 = 24ft (done piecing)
  • Top-Middle Rail: C * 2 = 14ft (done piecing)
  • Toe Board: B * 4 = 10ft (done piecing)
  • Total: 202ft / 61.5m

Actual Layout

  • Vertical Mains: 4.5m each * 4
  • Cross brace tubes: 1.125m each * 8
  • Toe board: 1.5m each * 2
  • Top middle rail: 2.1m each * 2
  • Top rail long side: 2.1m each * 2
  • Top rail short side: 1.5m each * 2
  • Ladder box vertical: 1m each * 4
  • Ladder box horizontal: 0.75m * 8
  • Total: 60.00m

Universe <-> Controller Mapping

Since physical locations of the controllers changed after all the channels were laid out, the universe numbers don't map linearly to the controller channels anymore.

  • Controller A (Base of VM2)
    • A-1 - U0
    • A-2 - U1
    • A-3 - U5
    • A-4 - U6
  • Controller B (Base of VM3)
    • B-1 - U2
    • B-2 - U7
    • B-3 - U9
    • B-4 - U10
  • Controller C (Base of VM4)
    • C-1 - U3
    • C-2 - U4
    • C-3 - U8
    • C-4 - U11

LED Strip List

ID Piece Location Length # of LEDs Reel Universe DMX Channels LightJams IDs
S01 Vertical Main 1 Vertical Main 1 450cm 144 C U0 1-432 BAA-BFN
S02 Vertical Main 2 Vertical Main 2 450cm 144 D U1 1-432 CAA-CFN
S03 Vertical Main 3 Vertical Main 3 450cm 144 E U2 1-432 DAA-DFN
S04 Vertical Main 4 Vertical Main 4 450cm 144 F U3 1-432 EAA-EFN
S05 Cross brace tube 1 MF1/2 B Right 218.75cm 70 G U4 1-210 FAA-FCR
S06 Cross brace tube 2 MF1/2 A Right 218.75cm 70 G U5 1-210 GAA-GCR
S07 Cross brace tube 3 MF1/2 A Left 218.75cm 70 H U6 1-210 HAA-HCR
S08 Cross brace tube 4 MF1/2 B Left 218.75cm 70 H U6 211-420 HCS-HFJ
S09 Cross brace tube 5 MF3/4 A Right 218.75cm 70 I U8 1-210 JAA-JCR
S10 Cross brace tube 6 MF3/4 B Right 218.75cm 70 I U9 1-210 KAA-KCR
S11 Cross brace tube 7 MF3/4 B Left 218.75cm 70 J U10 1-210 LAA-LCR
S12 Cross brace tube 8 MF3/4 A Left 218.75cm 70 J U10 211-420 LCS-LFJ
S13 Toe board 1 1 (Close) 150cm 48 A U11 349-492 MEM-MGH
S14 Toe board 2 2 (Far) 150cm 48 A U11 1-144 MAA-MBV
S15 Top middle rail 1 Left 212.5cm 68 K U11 145-348 MBW-MEL
S16 Top middle rail 2 Right 212.5cm 68 K U8 307-510 JDY-JGN
S17 Top long rail 1 Right 212.5cm 68 L U9 307-510 KDY-KGN
S18 Top long rail 2 Left 212.5cm 68 L U7 145-348 IBW-IEL
S19 Top short rail 1 TF1 150cm 48 B U7 1-144 IAA-IBV
S20 Top short rail 2 TF2 150cm 48 B U7 349-492 IEM-IGH
S21 Ladder box vertical 1 MF1 100cm 32 A U4 283-378 FDQ-FEV
S22 Ladder box vertical 2 MF2 100cm 32 A U5 283-378 GDQ-GEV
S23 Ladder box vertical 3 MF3 100cm 32 B U8 211-306 JCS-JDX
S24 Ladder box vertical 4 MF4 100cm 32 B U9 211-306 KCS-KDX
S25 Ladder box horizontal 1 MF1 Top 75cm 24 C, D U4 211-282 FCS-FDP
S26 Ladder box horizontal 2 MF1 Bottom 75cm 24 E, D U4 379-450 FEW-FFT
S27 Ladder box horizontal 3 MB2 Top 75cm 24 K U5 211-282 GCS-GDP
S28 Ladder box horizontal 4 MF2 Bottom 75cm 24 L U5 379-450 GEW-GFT
S29 Ladder box horizontal 5 MF3 Top 75cm 24 G, H U0 433-504 BFO-BGL
S30 Ladder box horizontal 6 MF3 Bottom 75cm 24 I, H U1 433-504 CFO-CGL
S31 Ladder box horizontal 7 MF4 Top 75cm 24 J, H U2 433-504 DFO-DGL
S32 Ladder box horizontal 8 MF4 Bottom 75cm 24 F, H U3 433-504 EFO-EGL

Reel Assignment

  • Reel A (100% used)
    • Toe Board 1: 150cm
    • Toe Board 2: 150cm
    • Ladder box vertical 1: 100cm
    • Ladder box vertical 2: 100cm
  • Reel B (100% used)
    • Top short rail 1: 150cm
    • Top short rail 2: 150cm
    • Ladder box vertical 3: 100cm
    • Ladder box vertical 4: 100cm
  • Reel C (100% used)
    • Vertical Main 1: 450cm
    • Ladder box horizontal 1a: 50cm (chunk 1)
  • Reel D (100% used)
    • Vertical Main 2: 450cm
    • Ladder box horizontal 1b: 25cm (chunk 2)
    • Ladder box horizontal 2b: 25cm (chunk 2)
  • Reel E (100% used)
    • Vertical Main 3: 450cm
    • Ladder box horizontal 2a: 50cm (chunk 1)
  • Reel F (100% used)
    • Vertical Main 4: 450cm
    • Ladder box horizontal 8a: 50cm (chunk 1)
  • Reel G (100% used)
    • Cross brace tube 1: 218.75cm
    • Cross brace tube 2: 218.75cm
    • Ladder box horizontal 5a: 62.5cm (chunk 1)
  • Reel H (100% used)
    • Cross brace tube 3: 218.75cm
    • Cross brace tube 4: 218.75cm
    • Ladder box horizontal 5b: 12.5cm (chunk 2)
    • Ladder box horizontal 6b: 12.5cm (chunk 2)
    • Ladder box horizontal 7b: 12.5cm (chunk 2)
    • Ladder box horizontal 8b: 25cm (chunk 2)
  • Reel I (100% used)
    • Cross brace tube 5: 218.75cm
    • Cross brace tube 6: 218.75cm
    • Ladder box horizontal 6a: 62.5cm (chunk 1)
  • Reel J (100% used)
    • Cross brace tube 7: 218.75cm
    • Cross brace tube 8: 218.75cm
    • Ladder box horizontal 7a: 62.5cm (chunk 1)
  • Reel K (100% used)
    • Top middle rail 1: 212.5cm
    • Top middle rail 2: 212.5cm
    • Ladder box horizontal 3: 75cm
  • Reel L (100% used)
    • Top long rail 1: 212.5cm
    • Top long rail 2: 212.5cm
    • Ladder box horizontal 4: 75cm

Cable Routing

  • Overview
    • When viewing the tower end-on with the ladder on the left:
      • Vertical Main 1 is on the left of the close end
      • Vertical Main 2 is on the right of the close end
      • Vertical Main 3 is on the left of the far end
      • Vertical Main 4 is on the right of the far end
      • Main Frame 1 is the lower close frame of the 2 lit ones
      • Main Frame 2 is the lower far frame of the 2 lit ones
      • Main Frame 3 is the higher close frame of the 2 lit ones
      • Main Frame 4 is the higher far frame of the 2 lit ones
      • Top Frame 1 is the close top frame
      • Top Frame 2 is the far top frame
      • Cross Brace A are the tubes running low-to-high from MF1/3 to MF2/4
      • Cross Brace B are the tubes running high-to-low from MF1/3 to MF2/4
      • Toe Board 1 is the close board
      • Toe Board 2 is the far board
  • Controller A
    • Channel 1
      • Feeder (up VM1) -> Vertical Main 1 -> Long Jumper (via Toe Board 1, down VM2) -> MF3 LBH Top
    • Channel 2
      • Feeder (up VM2) -> Vertical Main 2 -> Long Jumper (down VM2) -> MF3 LBH Bottom
    • Channel 3
      • Feeder (up VM3) -> Vertical Main 3 -> Long Jumper (via Toe Board 2, down VM4) -> MF4 LBH Top
    • Channel 4
      • Feeder (up VM4) -> Vertical Main 4 -> Long Jumper (down VM4) -> MF4 LBH Bottom
  • Controller B
    • Channel 1
      • Feeder (up VM4) -> CB MF1/2 B Right -> MF1 LBH Top -> MF1 LBV -> MF1 LBH Bottom
    • Channel 2
      • Feeder (up VM2) -> CB MF1/2 A Right -> MF2 LBH Top -> MF2 LBV -> MF2 LBH Bottom
    • Channel 3
      • Feeder (up VM1) -> CB MF1/2 A Left -> Short Jumper -> CB MF1/2 B Left
    • Channel 4
      • Feeder (up VM3) -> Short Jumper (via TF2 Top Short Rail) -> TF2 Top Short Rail -> Left Top Long Rail -> TF1 Top Short Rail
  • Controller C
    • Channel 1
      • Feeder (up VM4) -> CB MF3/4 A Right -> Short Jumper -> MF3 LBV -> Long Jumper (via MF3 LBH Top, VM2) -> Right Top Middle Rail
    • Channel 2
      • Feeder (up VM2) -> CB MF 3/4 B Right -> Short Jumper -> MF4 LBV -> Long Jumper (via MF4 LBH Top, VM4) -> Right Top Long Rail
    • Channel 3
      • Feeder (up VM3) -> CB MF3/4 B Left -> Short Jumper -> CB MF 3/4 A Left
    • Channel 4
      • Feeder (up VM4) -> Toe Board 2 -> Left Top Middle Rail -> Toe Board 1

Feeders and Jumpers

  • Feeders
    • Feeder A-1 - 310cm (F01)
    • Feeder A-2 - 310cm (F02)
    • Feeder A-3 - 310cm (F03)
    • Feeder A-4 - 310cm (F04)
    • Feeder B-1 - 360cm (F05)
    • Feeder B-2 - 360cm (F06)
    • Feeder B-3 - 360cm (F07)
    • Feeder B-4 - 711cm (F08)
    • Feeder C-1 - 576cm (F09)
    • Feeder C-2 - 576cm (F10)
    • Feeder C-3 - 516cm (F11)
    • Feeder C-4 - 687cm (F12)
  • Jumpers
    • Jumper A-1 - 308cm (J01)
    • Jumper A-2 - 262cm (J02)
    • Jumper A-3 - 308cm (J03)
    • Jumper A-4 - 262cm (J04)
    • Jumper B-3 - 92cm (J05)
    • Jumper B-4 - 152cm (J06)
    • Jumper C-1.1 - 73cm (J07)
    • Jumper C-1.2 - 142cm (J08)
    • Jumper C-2.1 - 73cm (J09)
    • Jumper C-2.2 - 142cm (J10)
    • Jumper C-3 - 92cm (J11)

Pinouts

Power+Data Cables

Signal Wiring Colour MX150L Pin RT360 Pin
GND White 1 A
+5V Black 2 B
Data Red 3 1
Clock Green 4 2

Power Only Cables

Signal Wiring Colour RT360 Pin
Bus A+B Negative Black A
Bus A Positive Green B
N/C N/C C
Bus B Positive White D

StrichLux Enclosures

  • Each StrichLux controller with its 4 output modules, 1 input module, and 1 power module will need an enclosure of some kind.
    • Pelican case? Could do one larger one with all 3 controllers in it, or 3 smaller ones
    • Assuming each StrichLux box has its own enclosure, would need a bunch of connections made through it:
      • Power - 2x12V @ 20A each - 4 20A power pins
      • Ethernet input - RJ45 - 1 RJ45
        • Will use M12 ethernet connectors (either from L-com or wherever else we're getting connectors) rather than RJ45s as they're cheap and IP67
          • Not RJ45-compatible, but an M12->RJ45 cable will be brought for troubleshooting, and they're not expensive
      • Strip outputs - SPI+5V - 2 signal pins, 2 10A power pins each x 4
        • JAE SRCN2A16-7S / SRCN6A16-7P would be good and extremely rugged, but the cost may be too high
          • Not weatherproof, doh!
        • Amphenol SINE MotionGrade M23/M40-series connectors look really nice, aren't too badly priced (compared with 5015 connectors anyway) and are weatherproof
          • They're really tall though, sticking out would negatively affect durability, so RT360 connectors might be better (but a bit more expensive)
        • Pricing is Receptacle, Sockets, Plus, Pins, Grip/Shell (* 4 per box * 3 boxes)
        • SINE M23: (6.46 + 0.728*4 + 8.430 + 0.640*4 + 3.61) * 4 = $95.88 * 3 = $287.64
        • SINE M40: (8.09 + 1.660*4 + 11.05 + 1.780*4 + 7.83) * 4 = $162.92 * 3 = $488.76
        • SINE RT360: (4.85 + 0.351*4 + 12.28 + 0.351*4 + 6.62) * 4 = $106.23 * 3 = $318.69
  • Molex MX150L will be used for inter-strip connections
    • Reasonably priced
    • Tooling is reasonably priced
    • Matte sealed

Enclosure Connectors

Function Description Part # Qty Req'd per Enclosure Status
Ethernet Connector Female Panel Mount Receptacle CON-M124PF 1 Ordered
Power Connector Male Flanged Receptacle RT00144PNH 1 Ordered
Power Connector Male Pins SP12A1T 4 Ordered
Power Connector (mating) Female Cable Plug RT06144SNH 1 Ordered
Power Connector (mating) Female Sockets SS12A1T 4 Obtained
LED Connector Female Flanged Receptacle RT00142SNH 4 Ordered
LED Connector Female Power Sockets SS12A1T 8 Obtained
LED Connector Female Signal Sockets SS20M1T 8 Obtained
LED Connector (mating) Male Cable Plug RT06142PNH 4 Ordered
LED Connector (mating) Male Power Pins SP12A1T 8 Ordered
LED Connector (mating) Male Signal Pins SP20M1T 8 Obtained


Inter-Strip Connectors

Description Part # Qty Req'd Status
Female Receptacle 19418-0019 51 Obtained
Female Crimp Socket 19420-0001 204 Obtained
Male Plug 19419-0004 51 Obtained
Male Crimp Pin 19417-0047 204 Obtained

Test Setup

  • Want to be able to power the whole setup from AC power for configuration/testing/programming
    • ATX power supplies provide plenty of +12V which could feed the PWR-DC5 modules
      • A modular supply like the OCX ZT-series would be cleanest, could modify the cables with Sabre connectors on the ends
      • Need 100W per channel, 200W per board, 600W total @ 12V
      • 100W per channel @ 12V means 8.33A
      • PCI-E and 8-pin CPU power connectors on ATX PSUs provide 6 amps per pin
        • PCI-E connectors have 3 pin pairs each
        • 8-pin CPU power connectors have 4 pin pairs each
      • Need four pins per board
        • Two boards get 8-pin CPU power -> Sabre connectors
        • The last board gets a 2xPCI-E power -> Sabre connector
      • OCZ ZT-series 750W power supply has all the required connectors

Lighting Programs

Specific Programs

  • Program 1 - Moving Rainbow
    • Track: ???
    • Rainbow pattern moves along the track
  • Program 2 - Wandering LEDs
    • Track: 4 parallel tracks, each following:
      • Counterclockwise along top rail to VM
      • Down VM to top MF bottom LBH, loop around LB
      • Down VM and repeat the loop at the bottom MF
      • Down VM to the very bottom then up the CB
      • Not included: middle rails, toe boards, top CBs
    • A number of LEDs wander around the track randomly
  • Program 3 - Random Flashes (programming complete)
    • Brief flashes of random LEDs/groups of LEDs
  • Program 4 - Random Fades
    • Fading in/out of random groups of LEDs

Program Ideas

From Jammies

  • a quick-moving plasma effect scaling upwards, with LEDs at the top glowing more and more brightly white, that makes the tower look like it's "powering up"

LightJams Snippets

  • Random stepped values over time: if(randpulse(1, 75, 0.01), gmem(0, rand(0, 100)), grecall(0))
  • Bouncing (still needs work at the top): fadein.linear(1.5, fadeout(2.5, pulse(3.5, 1)))

Cost

Item Status Cost
IO-DMX + IO-BO Rev. 1 PCBs Obtained $40.69
IO-DMX + IO-BO Parts Obtained $18.31
IO-SPI Rev. 1 PCBs Obtained $49.65
CORE + IO-ETH + PWR-DC5 PCBs Obtained $454.66
CORE + IO-SPI + IO-ETH + PWR-DC5 Parts Obtained $184.40
Test LED Strip Obtained $25.00
OCZ ZT 750W Test PSU Obtained $107.34
60m LPD8806 LED Strip Obtained $1320.00
Fit-PC 2i Obtained $455.00
Lightjams License (design PC) Obtained $78.59
Lightjams License (installation PC) Obtained $78.59
Ethernet Switch Obtained $363.85
2 sets of StrichLux IO-CORE + IO-ETH + PWR-DC5 Parts Obtained $512.83
12 sets of StrichLux IO-SPI Parts Obtained $190.39
Enclosure A Obtained $62.15
Enclosures B-D Future $186.45
Enclosure Connectors (order 1) Obtained $563.09
Enclosure Connectors (order 2) Obtained ~$0.00
Inter-Strip Connectors (order 1) Obtained $320.55
Inter-Strip Connectors (order 2) Ordered $30.01
250ft 14/4 SJOOW Cable Ordered $371.44
SSD for Controller Obtained $101.69
Windows License for Controller Obtained $113.00
Shipping (outbound) Done $455.92
Tax/Duty (outbound) Done $104.40
Shipping (return) Done $224.29
Tax/Duty (return) Done $443.06
Total Ordered $6668.90
Total Future $0.00
Grand Total $6668.90
Hours Spent 577

Photos

  • Beacon4 Scaffold 3D Model

Retrieved from "http://hackstrich.sarahemm.net/wiki/index.php?title=Beacon4&oldid=1826"