Infrared Communication for Light Shows

[lightman]

I decided to open a new thread regarding my work on IR communications. I have hijacked mrpacketheads thread on Ear to Ear Communication long enough...thanks again.

I will post a quick history from the other thread for reference after providing a quick review of key elements here.

Purpose of this Thread:
1) To discuss and share Infrared (IR) communications as it relates to light shows and other communuication needs.
2) To provide updates on the progress of my portable TX/RX IR system for use with lighted headsets/ necklaces, or other lighting elements.

Project Description:

My goal is to develop a IR transmitter and receiver that will take several channels from my main light show and convert them to an IR transmission that is received by a portable device (like a headset worn by a person that contains lights). The portable unit will contain a minimum of 2 GECE bulbs as the lighting effect. This way my guests can share in the immersive nature of the light show with blink flashy fading lights on their head sychronized to the music and other lights on the house.

For the sake of this discussion.... the remote portable unit will be refered to as a headset... even though I am still deciding on final configuration (i.e., Necklace, headband with lights, magic wand, lighted baseball cap...etc).

I plan to build about 20 headsets to hand out (on temporary basis) every night to kids and adults alike. The headset needs to portable, so it will be battery operated that will last at least 5 hours. Headsets will be recharged every night. I plan to replace the underlying head band for sanitary reasons as the unit gets moved from guest to guest. This will be accomplished with velcro attached to bulbs and electronics pack.

Several transmission stations will be built to provide proper coverage of IR beams. I currently believe 3 stations will be sufficient and placed on trees and portions of building to get proper spread.

Transmitters will sniff the 38.4kBaud main lighting Vixen bus and pull off 8 channels (red, green, blue, control for each GECE bulb). An SX processor will perform the sniffing and reformatting of data for IR transmission. The SX will generate the requisite 38kHz (now 57kHz) carrier and then modulate it at 2400 Baud (now 4800 Baud) for output to the IR LED emitter. I plan to use 160mW emitters in groups of 3 for each transmitting station.

The Headset electronics will then detect the incoming 2400 Baud (now 4800 Baud) IR signal with an IR detector and send that inverted RS232 signal to the RX SX processor for processing. The RX SX will immediately reformat the 8 incomming channels for retransmission to the 2 GECE bulbs. This entire process is proving out reliable at a 100 ms update period. However, the new faster IR Detector no longer requires a rest GAP time of 25 msec, so updates are well within the 50 msec period with 4800 Baud. Running at full 20 Hz update rate.

History to Date

Here are post links to previous threads describing progress and findings (See Ear To Ear Communications Thread)...

Distance Test: http://doityourselfchristmas.com/for...978#post211978
Battery Test: http://doityourselfchristmas.com/for...032#post212032
IR TX/RX Test: http://doityourselfchristmas.com/for...241#post212241
GECE Control with IR Test: http://doityourselfchristmas.com/for...253#post212253
Transmitter Range Test: http://doityourselfchristmas.com/for...352#post212352
Inversion of Logic Increases Range & Schematic: http://doityourselfchristmas.com/for...386#post212386
Increaing Baud Affect on Reliability: http://doityourselfchristmas.com/for...420#post212420
Headset Protoype: http://doityourselfchristmas.com/for...432#post212432
Outside Range Testing: http://doityourselfchristmas.com/for...481#post212481
Complete System with Vixen - End to End Test : http://doityourselfchristmas.com/for...505#post212505
4800 Baud Achieved: http://doityourselfchristmas.com/for...384#post213384
Range/Distance Outside Test w/57kHz Carrier: http://doityourselfchristmas.com/for...914#post213914
Multiple Transmitters: http://doityourselfchristmas.com/for...961#post213961
Prototype I Manufacturing -Headband Style: http://doityourselfchristmas.com/for...981#post236981
Coin Cell Battery Test: http://doityourselfchristmas.com/for...337#post237337
Weight Breakdown of Headset: http://doityourselfchristmas.com/for...183#post238183
New Rudolf Headset with Baseball Cap: http://doityourselfchristmas.com/for...979#post238979
Preliminary Real Show Test of Headset/Transmitter: http://doityourselfchristmas.com/for...500#post239500
System End to End Test in Live Show; Flicker Elimination: http://doityourselfchristmas.com/for...471#post240471
IR Reindeer Headset Debut a Success: http://doityourselfchristmas.com/for...963#post241963
IR Headset Receiver / Transmitter SX Code:http://doityourselfchristmas.com/for...492#post242492
Sequencing the Headsets - Easy Method: http://doityourselfchristmas.com/for...638#post242638
Timing Diagram for IR Communication: http://doityourselfchristmas.com/for...406#post243406
Logistics of Managing Headsets: http://doityourselfchristmas.com/for...691#post253691

Video of END TO END test (W/Vixen):
Video of Headset in Action: http://youtu.be/b5dlFordOls
Video of All Three Prototype Headsets in Action: http://www.youtube.com/watch?v=tBn5vSnY7AA
Video of IR Reindeer Headset Debut: http://youtu.be/9In1cqq8Nyk

Status to Date

- Battery test complete - over 7 hours of battery life with 4 AA batteries; Final unit will use 4 AAA batteries
- IR distance test completed with 60 ft using standard TV remote and 30 ft with TX/RX 2400 Baud serial transmission (2 -15 mW emitters)
- TX module prototype is done and working receiving VIXEN serial stream and retransmitting on IR band (includes hardware and software)
- RX module prototype is done and working receiving IR transmission at 2400 Baud and controlling GECE bulbs properly (includes hardware and software)
- Vixen Light Show successfully integrated into the loop; Complete end to end testing works!
- Communication is robust and reliable. Interruption to IR beam is managed well and system recovers quickly and cleanly
- Upgraded to 57kHz carrier and improved detector... running at 4800 Baud
- Implemented 300 mW/sr IR transmitter circuit (57 kHz carrier)... working good with increased range (>30 ft in lab, reflected, indoor test).
- Completed Range/Distance test with outstanding results...over 108 ft reception at 4800 Baud. Good swath of coverage
- Succcessfully ran dual transmitter test (2 carriers) with proper IR Detection...even with small phase shift between carriers
- Succesfully fabricated Production Prototype that fits in 2" x 2.25" board; Runs at 4800 Baud with Vixen sequence
- Performed Coin Cell Battery Test: They are not adequate in capacity....Will stay with AAA batteries
- Weighed headset components and determined Battery and case comprise 40% of total weight; GECE bulbs and mounts comprise 47% of total weight
- Built Transmitter enclosure for loft. It will be added to the end of the data line.
- Designed and fabricated a new baseball cap style headset... Rudolf... using coroplast as endo skeletal structure
- Preliminary test of Transmitter and Headset performed; Headset responded with good range; Some noise flicker exists in lights
- Eliminated flicker with split data bus and incorporation of checksums
- Tested and Debuged IR Headset Transmitter in real live show; Corrected IR Transmitter#1 AC power issue with Surge Protector in loft
- Sequenced all 15 songs to include Headsets: Recommend using pre-exsiting background channels in your sequenece... copy/paste
- Posted SX code for both TX and RX units on forum
- Manufactured 10 Reindeer headsets total for this season
- Logistics of managing headsets require babysitting them; Requires stamina and inclimate weather can have an affect on the success of your show

Next Steps
1) Evaluate and Develop a low cost PCB surface mount and DIP versions of IR RX; LabRat leading PCB layout and prototype tests

[LabRat]

I'd like to assist if there's an opportunity. Protocol design, perhaps DMX to IR bridge, or XBee to IR bridge etc.

[lightman]

I'd like to assist if there's an opportunity. Protocol design, perhaps DMX to IR bridge, or XBee to IR bridge etc.

You got it buddy. Lets chat.

Lightman

[lightman]

TX Timing - IR Detector Needs a Break...

I ran some tests on re-timing the TX interface to see if I could increase the update rate from 10 Hz to 20 Hz. Got some interesting results.

I discovered that the IR Detector has limit to how soon you can send bursts of data. They require about a 25 msec downtime called a "Signal Gap" between continuous 38 kHz carrier bursts. Without the gap, the detector refuses to process any more information.

Now why did they design that in? It has to do with their filtering algorithms that reject spurious noise like Flourecent light or other TV remotes broadcasting in the area. Some of their filters require 7 msec of rejection processing and other things that require the quiescent period.

When honored, the little detector does great... When violated the detector goes dark and gives out nothing. Bottom line... respect the gap time.

What does this mean to my design... I just have to account for the 25 msec somewhere in the processing timeline. With a 50 msec Vixen update rate, that doesn't leave a lot of wiggle room. I can fit everything in stock at 100 msec update periods. If I combine the GECE color hues of 4 bits and stack them into bytes.. I can shave off more TX time. Its a game... take some here put some back there.

This is complicated by my recent discovery that my laptop and netbooks both output Vixen with a varying Break-Idle time. I can lose up to 10 msec of that processing period when the Break-Idle creeps up and steals that away. Good news is my design is tolerant and simply waits for next clean signal burst (no spurious flickering ...YEAH!)...but it slows down update rate to 10 Hz (100 ms periods).

Still in work...
Lightman

[lightman]

Just completed video of the IR TX/RX End to End test (Vixen to IR TX to IR RX to Headset to GECE Lights)...

Sequence: Amazing Grace



Regards,
Lightman

[rush44oh1]

Great job lightman I can't wait to get my hands on some of those if they are any thing I can do to help give me shout

[lightman]

Great job lightman I can't wait to get my hands on some of those if they are any thing I can do to help give me shout

Thanks rush440h1. I will keep publishing schematics and timing diagrams...keep a look out... send me comments.

Hopefully the final form factor will be something we can all use... or at least understand so it can me customized to your application.

Later...
Lightman

[Barnabybear]

Fantastic work – I’ve only been a member since January (I think) and can’t believe how fast things develop on this site.
I was reading this new thread this morning and wondering if there would be any sources of IR interference that may affect the ears, in particular the fact that I have CCTV that defaults to IR in low light levels. Then I got to wondering if the built in IR of the cameras could be hacked to form the transmitter as they probably not be needed due to the overall light level of the show.
May be a non-starter – but just a thought.

[lightman]

Barnabybear...

The IR detector I use filters out unwanted noise from other IR sources. That way the detection is clean and robust. It does by first only accepting a narrow carrier frequently of 38 kHz. So if the stray IR is not cycling at 38 kHz, the detector ignores it. Constant IR signals are ignored too.

Second, it has built in algorithms that filters out noise from flourescent lights, natural sources of IR,...reflections from sunlight, miscellaneous TVs remotes, etc. So the IR source in your CCTV should be filtered out with possible exception of overdriving the detector with too much radiation.

Could your IR source be used to transmit your communication signal..I believe so. It just needs to be modulated. This would require hacking into unit and adding power transistors to gate the IR at carrier freq.

Hope this helps,
Lightman

[Barnabybear]

I must admit - sunlight was one of the other things I thought of. Looks like you've got this well covered, I shall continue to follow the thred with intrest.