Loading...
 

Troubleshooting Infrared Repeater Systems


by Jeff Fisher

Server does not support image manipulation.
Tip 1: If you don't have "blink" infrared emitters, get at least one 283D for testing. (Blink emitters emit a visible-as well as invisible infrared-light.)

Tip 2: The "talk back" led on most infrared pickups (all of Xantech's) will NOT light up if there is not a complete current path. This means that if any connection comes loose or the thin emitter wire breaks, the talk back led will not flash. (If you have more than one emitter and one emitter wire breaks, the talk back will still flash.)

Tip 2a: Those thin emitter wires break quite easily. And there is often no visible evidence of the break.

Tip 3: The problem is not the connecting block itself. Connecting blocks are just connectors and a few resistors. We've never seen a defective one.

Tip 4: A common failure mode we've seen: Because of the way the infrared pickup's 1/8" three-conductor plug is laid out, if it is not fully inserted in the jack, or it is inserted in a two-conductor jack, the power supply will be shorted out. Most power supplies will run very warm for a little while then quit entirely. Often the power supply will have visible indications of overheating, or even smell "burnt". Make sure your receivers are plugged all the way into the jacks. Test the power supply. Make sure it isn't getting very warm.

How Infrared Distribution Systems Work

They are really very simple. People tend to think of them as being more complex and sophisticated than they are. There are three conductors involved. (When a fourth is involved, it is usually a superfluous signal.)

  • Ground
  • +12V
  • and Signal


The connecting block simply connects everything together.

The emitter is just an infrared led (blinks include a visible led also) and usually a current limiting resistor. Note that emitters only use the signal and ground connections.

The infrared receiver does most of the work. (That's why they are the most expensive part.) They consume the power in the system. They turn received infrared flashes into a 0 to 12 volt signal impressed on the signal line. This signal line (and the ground return) directly drives all the emitters plugged into the connecting block.

The connecting block has divider resistors to try to route a similar current to each emitter plugged into it. Since different emitters have different resistances, there is a tendency for some emitters to get more current (and thus flash "brighter") than others. The dividing resistors are only partially effective at equalizing the current. That's why we recommend using the same type emitter in each port.

The talk back led on most emitters is in series with the signal line, so it will only flash if there is someplace for the current to go.

Troubleshooting

We have found that the best way to troubleshoot completely dead infrared systems is to simplify everything. Get a small test-bed working. Then add the real-world pieces back in one at a time until it quits. The last piece added is the broken piece! Here are those three steps spelled out:

Step 1: Create a bench-top infrared repeater system using the components you have. This means taking a blink emitter, connecting block, power supply, and infrared receiver out of your system and setting it up on a bench. Grab a remote and see if you can make the emitter blink. If it doesn't, swap components until you get a blink. If you're still stuck here, see "Using a Voltmeter" below. Otherwise you're going to have to send everything back to us for testing.

Step 2: Add real-world pieces back in. Take your pile of "known-good" stuff to where your equipment to be controlled lives. One at a time, "install" the pile. Test after each step. Usually, at some point things quit working. Stop!

Step 3: The last thing added is a problem. It may not be the only problem, but it is one of them. If all you did was remote the infrared receiver through a cable, the cable (and the connections at each end) are suspect. If you switched to a different emitter, the emitter is suspect...pull it out and try it on the bench.

We have troubleshot many a "dead" IR system using this method. It has yet to fail. (Although we always had a voltmeter available.) IR systems that are just not very reliable (I.E. they sort of work) are harder to troubleshoot and are the subject of a different article.

May all your troubleshooting be simple and productive!

Troubleshooting With a Voltmeter

Server does not support image manipulation.
A simple voltmeter is very useful for troubleshooting infrared distribution problems. Here are a few of the uses:

Measure Power Supply Voltage

Many infrared receivers are surprisingly picky about their supply voltage. Xantech receivers, in particular, don't like over voltage. Get much over 12 VDC and they start to create "noise" on the signal line. So we like to see 11 to 12 VDC at the receiver. Xantech used to ship an unregulated power supply. Unregulated power supplies, when lightly loaded, put out a much higher voltage than they are rated for. So Xantech shipped a 9 volt unregulated power supply. As you can imagine, this cause no end of confusion. Fortunately, regulated power supplies have come down in price and Xantech switched to a 12 VDC regulated version. Measure your power supply and make sure it is 11 to 12 VDC with a small load. Some regulated supplies get very unhappy with no load at all (which is an undefined situation as far as they are concerned) so test the supply with at least one receiver connected.

See The Infrared Signal

You can actually "see" the signal with a voltmeter. Put the voltmeter on Volts - DC and flash a remote into the receiver. You should see the reading go up and fluctuate. How much it goes up is dependent on the code being sent, frequency, distance, etc. This test is more to see if the signal is there at all than to measure the quality of the signal. To do that, you need an oscilloscope (see below.)

Test Emitters

Infrared emitters are one or two LEDs in series with a current limiting resistor. LEDs are a special type of diode. Diodes pass current in only one direction. If you voltmeter has a diode measuring mode (many do, look for the diode symbol
Server does not support image manipulation.
, you can use your voltmeter to test infrared emitters! Just put the meter into the diode test mode and connect the probes to the two terminals of the emitter. If you don't get a change off of infinity, reverse the leads. (Remember, diodes only pass current one way.) If you don't get a non-infinity reading after switching the leads, the emitter is "blown".

Troubleshooting With an Oscilloscope

As explained earlier, infrared receivers are pretty simple. They "watch" for an infrared signal and impress that signal onto the signal output. (The complexity involves weeding out noise and detecting weak signals.) The wavelength of the infrared light is around 850nm. Fortunately, we don't have to deal with those speeds. The infrared is essentially treated as binary. It is either there or it isn't. The remote control switches this 850nm light on and off in some pattern that the receiving device can understand. Most remotes (there isn't really a standard) switch a "carrier" frequency on and off in a run-length-limited style of encoding. The frequency of the carrier is normally 30 to 50 kHz.

If you connect an oscilloscope to the signal and ground wires of a properly functioning infrared distribution system, you should see a series of square waves pulsing on and off. Ideally, the trough of the square wave (in the middle of the carrier) should drop below 0.7 volts and the peak should be above 5 volts or so. The cleaner the square wave, the better. We have seen situations where the infrared source was overdriving the signal. (The bottom of the square wave never got below 3 volts.) Adding a "dropping resistor" between the signal and ground pulled the signal down enough to make it work. This is the reason for the emitter with the built-in trim-pot.

Jeff

Register (under "Log In" above) to leave a comment.


   

Share

Newest Blog Posts

  1. A new home.
    2014-10-30 22:44