Hi guys, hopefully somebody can help me out with this. My dad likes to work on watches and would like to know the various rates of unknown hairsprings. I thought I could accomplish this with an IR LED receiver/transmitter circuit. When the spoke inside the hub of the hairspring breaks the IR signal it would serve as the input for a counter program. Long story short, I have to use a 555 timer for the basic stamp microcontroller because the freqout command will not allow for any other simultaneous actions. With the 555 circuit I can simply use a hi/low command and get the job done. My problem is my 555 circuit is not working properly. If I take the IR LED transmitter out of the circuit I can get a debug counter to my computer which allows me to tune the potentiometer to desired frequencies. If I place the IR LED transmitter back into the circuit I end up with a 0 output to the debug screen which essentially means I don't have anything coming out of the output pin of the 555 chip. So I was wondering what might be causing this? I have reworked this circuit and checked it probably 20 times over and get the same results every time. I attached the diagram along with my circuit, hopefully somebody can see something or might have an idea as to why it is behaving this way. I am also open to suggestions if somebody has another idea which they think might be easier and/or more efficient for the project at hand. Thanks in advance for any help and sorry for the long winded description.

 

What are you using to detect the 38kHz? Most likely the oscillator is not quitting, just shifting frequency to a degree that your detector is not seeing it. An o'scope would tell the tale.

A couple of things on the posted circuit. There needs to be a big bypass cap from Vdd to Vss, right across the 555. The modulation input, pin 5, needs to be bypassed.

 

For short range I would not use modulated IR. IR beam- receiver needs to be narrower than interupting spoke. A reflected signal might work better as it might not be as sensitive to beam width.

 

I would take the output for the debug/counter from pin 3 of the 555 timer and not across the led.

 

What are you using to detect the 38kHz? Most likely the oscillator is not quitting, just shifting frequency to a degree that your detector is not seeing it. An o'scope would tell the tale.

A couple of things on the posted circuit. There needs to be a big bypass cap from Vdd to Vss, right across the 555. The modulation input, pin 5, needs to be bypassed.

It is the IR receiver for the boe-bot project sold through parallax. The IR LED is an IR T1 3/4 LED. When I run the debug program to see the output the oscillator is just outputting 0 which means it stops, right?

 

For short range I would not use modulated IR. IR beam- receiver needs to be narrower than interupting spoke. A reflected signal might work better as it might not be as sensitive to beam width.

That is interesting, are you familiar with the basic stamp at all? Do you know what components might be useful in accomplishing this? Sorry, I am a complete newb when it comes to electronics. I wonder if simply placing the IR receiver behind the transmitter if that would pick up any deflection from the spoke which is placed in front of the IR LED?

 

Basic Stamp-- has inputs & outputs & requires programming. Depending on physical arrangement of spoke, a slot detector might work or from one side a reflective detector. Pictures of same at Electronic Goldmine, # G 20133 or G 18740 & a bunch of slot detectors. Do I know how to hook them up--Yes.

 

how about just using the 555 to power a visable light led? use it like a strobe light., led's are fast, and would let you see that the light path is ok.

 

Basic Stamp-- has inputs & outputs & requires programming. Depending on physical arrangement of spoke, a slot detector might work or from one side a reflective detector. Pictures of same at Electronic Goldmine, # G 20133 or G 18740 & a bunch of slot detectors. Do I know how to hook them up--Yes.

This looks like an interesting option

Basic Stamp-- has inputs & outputs & requires programming. Depending on physical arrangement of spoke, a slot detector might work or from one side a reflective detector. Pictures of same at Electronic Goldmine, # G 20133 or G 18740 & a bunch of slot detectors. Do I know how to hook them up--Yes.

This is interesting, how does the slot detector work? Is it something akin to having lets say a pendulum swing back and forth between the slot and it senses it? How does the photoreflector operate? It seems after reading the description on the website that everything is packaged into one little unit. So would it be possible to just place the spoke of the hairspring in front of the photoreflector. The photoreflector would send a signal as well as sense the reflection correct? Sorry for all these questions, I am just new to all of this.

 

For short range I would not use modulated IR. IR beam- receiver needs to be narrower than interupting spoke. A reflected signal might work better as it might not be as sensitive to beam width.

Optics may be the hard part in this project. How big is the spoke and the space between spokes? A picture with a good size reference would be very useful.

 

Sounds like you have basic understanding of slot detector. There are many varities, depth & width of slot & shapr of IR transmition path. Generally the reflective sensor is close range, maybe up to 1 cm to read patterns in black & white or a shiney spoke.
To be more consistant,I would measure time only when spoke is moving in same direction for a full cycle. Were you planning to use 555 as a gated pulse generator to serve as a EPUT meter, events per unit time??

 

I found a no name reflective sensor in junk box; checked conduction between all 4 leads with analog VOM on X 10k ohm range. Pair that conducted is emitter, + lead to anode. Connected 5 V battery via 150 ohm R to anode , cathode to -. Guessed that cathode & emitter were on same side, grounded emitter?? , connected collector ?? to + battery via 10k R. Monitored collector V & passed white paper over sensor, V fell confirming connections. Passed various objects over sensor, 1 mm pin, black lines on white paper, white on black & tooth pick, all were detected. Best spacing was 5 mm to 15 mm. To use as an input signal, I would connect to schmitt trigger or a 555 to square up pulse.

 

I found a no name reflective sensor in junk box; checked conduction between all 4 leads with analog VOM on X 10k ohm range. Pair that conducted is emitter, + lead to anode. Connected 5 V battery via 150 ohm R to anode , cathode to -. Guessed that cathode & emitter were on same side, grounded emitter?? , connected collector ?? to + battery via 10k R. Monitored collector V & passed white paper over sensor, V fell confirming connections. Passed various objects over sensor, 1 mm pin, black lines on white paper, white on black & tooth pick, all were detected. Best spacing was 5 mm to 15 mm. To use as an input signal, I would connect to schmitt trigger or a 555 to square up pulse.

Wow, I am going to have to read this a couple times over as it went over my head, lol. But I would say the way I would interpret that post is that you were successful in creating a circuit which could detect an object as small as a 1mm pin. And you have a hunch as to how to use this "output/input" for a microcontroller such as the basic stamp using a 555 timer or Schmitt trigger. I am not familiar with the Schmitt trigger but I am going to do some research on it. Thanks again for all the help as I have been stuck in a rut with this project. What is good to know is that you dropped the voltage down to 5 volts using a resistor if I am reading that correctly because the output voltage for the basic stamp is 5 volts. Coincidence? At this point I don't think it is, so thanks.

 

As an aside, I would have done it differently. I would build a strobe light using a white LED. Then you do not need to detect the signal. What you need is a digital readout of the strobe frequency, i.e. a frequency meter.

 

Erik, how do you plan to process the detected signal? Do you have something with a numeric display?

 

As an aside, I would have done it differently. I would build a strobe light using a white LED. Then you do not need to detect the signal. What you need is a digital readout of the strobe frequency, i.e. a frequency meter.

Would this work for low end frequencies? I get what you are saying, much like a doing the timing in your car. The only problem here is the hairspring oscillates back and forth at maybe 100 rpm's

Erik, how do you plan to process the detected signal? Do you have something with a numeric display?View attachment 78630

Right, I have a 2x16 LCD Display which is easily programmed with the basic stamp.

 

Maybe this would work: Adjust U 2 to overlap second pulse so that measurements are with spoke always moving in same direction. U3 is toggle flip- flop with Q giving a gate signal for input to stamp; assuming an internal clock of around 1 k Hz can be programed. U4 probably internal to stamp. stamp.
R4 should be around 1M ohm.

 

Yes, you're right. 100 rpm is like 2Hz which is way too slow and would give anyone a headache.