timing system help

[APalmer]

Helloooo all!!!!!!
So I have this timing system.. using UV sensors with UV Lights. Race coordinator software. Everythingworks great when its on and no cars are running... shortly after I start running a car, phantom laps start counting away. I have double checked and made sure no wires are near each other to help eliminate noise interference. . I have changed 3 different wal-wart power supplies for the light gantry.. I even wrapped and rf noise filter around the power for the light gantry.. any ideas??????.. anyone??

[Kemtronracer]

I've had that problem when a UV light was going bad, or was too weak. Try looking at the light output using a digital camera with the light bridge on and the lights in the room off (has to be done at night).

[APalmer]

Thanks will look at that tonight..
also would it be better to run the lights in series or parallel. . And each light getting its own resistor?

[RichD]

I'll bet that you mean IR emitters and sensors. A count is triggered when something blocks the light, so my best guess is that you are getting false counts because the amount of light getting to the sensors is right at the point where there is enough to trigger them. If the light output of the emitters goes down by just a tiny amount you would get a false count. It does not matter whether the emitters are wired in series or parallel as far as the actual counting goes. If the emitters are wired in series and one goes out that will knock out the rest of them. Finding which emitter is bad when they are wired in series is more difficult and it is recommended that you wire them in paralel with a separate dropping resistor for each one.
Besides needing enough output from the emitters the alignment of the emitters and detectors can be critical. You had already considered that there could be some sort of electrical interference and taken steps to avoid that. I presume that the power supply for the emitters is regulated, if not the voltage to the emitters could drop once in a while causing the emitters to dim. Possibly vibrations from the cars going around the track could cause a small misalignment between the emitters and detectors or just jiggle a bad connection someplace. You might consider using a cluster of three emitters for each sensor. That would eliminate the problems of poor alignment and marginal light levels.

[APalmer]

RichD.... yes IR. is what I meant. That's good info you gave. I am in the process of running them in parallel. I am following Greg Brauns info from hoslotcarracing but I keep coming up with different resistor ohm requirements than what he suggest. I have a 12v unregulated wal-wart. That strangely puts out 18.33v at 200math. (4 IR led's @ 1.2fwd v, 100ma) he is suggesting 470ohm resistors... I keep using online calc's and come up with anything from 120ohm down to about 75ohm..in parallel.
Any further insight??

[RichD]

If you are trying to size the dropping resistors you first have to know what voltage the power supply puts out when it is under a load. For many other applications you would only need a ball park figure, but for powering LEDs, especially IR LEDs, you need a more exact figure. With LEDs an extra half of a volt over its rated value will shorten its life, an extra volt will burn it out right away. A low voltage would greatly reduce the output of the LED. If the power supply actually puts out 18.3 volts under load, the LEDs drop 1.2 volts and use 100ma and you wire the LEDs in parallel with a separate resistor for each one then you would have to use 180 ohm resistors. If you were to build the circuit and measure the voltage at the power supply and it was really less than 18.3 volts you could go to lower value resistors to be sure that the LEDs had their full output. Here is the calculator that I used: http://ledcalc.com/ . Since I have a good background in electronics I don't need a calculator of this sort, but using it gave me the nearest resistor to the actual calculated value of 171 ohms. You only need to do the calculation for one LED/resistor combination when they are all wired in parallel.
You should be aware that the sensors that you use should be matched in frequency to the emitters. Even a small mismatch would result in the sensor seeing only a fraction or the emitter's output. If you had used the emitters and sensors that Radio Shack sells in the same package those do not match. The detectors are rated at 850nm and the emitters are rated at 950nm. With that combination you might loose 3/4ths of the emitter's output.

[APalmer]

I will pick up a pack of 180 resistors and give it a whirl. The sensors are flat with a very small bubble where the sensor actually is. They do not resemble an LED at all. Now the emmiters I did pick up are the ones from RS that MR. Braun has listed on his site. I appreciate the information that has been given. Once I try all this and get better numbers I will post yays or nays... may also try just plain incandescent light just for trial sake as well.

[RichD]

The sensors will be looking for a longer wavelength than regular incandescents put out, but those do have some output in the IR range, so they will probably work if they are bright enough.

[Kemtronracer]

I have always found the photo-transistors and IR emitters from Radio Shack to be problematic. Their emitters vary in quality and as such, don't always trigger laps effectively. I have also seen their photo-transistors fail over time, probably due to using emitters with the wrong frequency. I have used matched components from DigiKey with much success: RPT-37PB3F for the photo-transistor and SIR-34ST3F for the IR emitter. If you read the spec sheet for the photo-transistor you are using, it should tell you which emitter to use.

[RichD]

Thanks Kemtronracer, I have saved those part numbers for future reference. The sensors will not be damaged if the emitters are the wrong frequency, they just won't be sensitive enough unless the emitters and sensors are very close to each other. The intensity of the light drops off in proportion to the square of the distance between the emitters and the sensors, so a small change in the distance can have a bigger effect than you would expect. The sensors can also have a narrow viewing angle, making the alignment more critical.

[Kemtronracer]

As long as you're taking down part numbers, I'll throw in QSE113 and QEE113 for a photo sensor and emitter that work across the slot (under the braid). These work really well unless your guides are translucent like white Parma guides.

[APalmer]

Again thanks for all the input... I have, what appears to be a successful change. From serial wiring to parallel, using 180ohm resistors in place for each emitter. So far, no phantom laps!!