I built this I/R sensor circuit a while back. I had everything working perfectly till I had a tragic accident with my CNC machine (it was to be the RPM sensor for the machine). So I had to rebuild the thing from scratch.

Now I am not getting a transition at all on pin on of the LM2903.

My supply is +5V. I have verified supply is at pin 8 of the op amp

Pin 1 sits at a solid 0.975V no change reflecting or not reflecting.

Pin 3 is at 0.495 volts.

The strangest thing is on Pin 2. Not reflecting it is at around 4 volts but then I see these periodic transitions to around a few milli volts. Sometimes it stays there for less than a second. Other times it stays there for several seconds.

I have turned off any potential source of I/R like my florescent light.

Reflected pin 2 is just a few milli volts.

I was able to capture one transition right in the middle of the transition and included it below.

If the op amp is pulled out of the circuit then pin 2 (of the socket) stays at just a little over 3V not reflected and goes to just a few milli volts when it is reflected.

I have tried 2-3 opamps all with the same result.

This is very frustrating since I already had the thing working.

My only difference between what I have now and what I had working is I am fairly certain R1 used to be a 20k. I don't know why I did it but I salvaged the old board for parts. It was stupid. Somehow I thought it would be easier swapping out parts. I should have left everything intact.

The other change was R6 was originally left out of the design. After realizing my mistake I simply soldered a 10K from pin 1 of J1 to pin 1 of J2. I then updated the pcb design in the event I was going to make more boards. But I can't see how that could cause an issue.

But could too high of a value for R1 cause those strange oscillations?

 

Isn't that a bit overkill for a I/R RPM sensor?
What is the sensor?
Or what about a slot opto?
Max.

 

You appear to have some sort of intermittent connection or cold solder joint.

Do you have a good power and ground connection to U2?

 

Also what is recieving the I/R signal for processing?
Max.

 

The voltage at comparators + pin is quite low < 0.5V
Will the opto output go below it?
You could increase R1 to 1K.

Edit. should be: increase R4

 

Does the sensor circuit input to a Pic-micro?
Max.

 

The connections to U2.2 are weird. The output is grounded. That might be interfering with the other amplifier. Connect the output only to the inverting input and connect the non-inverting input either to ground or to the 0.5V reference.

 

The connections to U2.2 are weird. The output is grounded. That might be interfering with the other amplifier. Connect the output only to the inverting input and connect the non-inverting input either to ground or to the 0.5V reference.

As per the datasheet.

 

Does the sensor circuit input to a Pic-micro?
Max.

Yes I was not getting a good response out of the I/R sensor so I added the opamp.

The I/R sensor worked fine as a simple switch but not when using it to sense rpm.

 

Also what is recieving the I/R signal for processing?
Max.

I don't understand your question. It is both emitter and receiver. I/R reflects off of reflective tape.

But the issue occurs when it is NOT reflecting.

 

As per the datasheet.

I can see nowhere in the datasheet where it suggests that arrangement.
https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjjhuWuy7LXAhVmB8AKHbp0Az8QFggoMAA&url=http://www.ti.com/lit/gpn/LM193&usg=AOvVaw2gVV6jhFiJBeB-3lvnZUD1

 

I can see nowhere in the datasheet where it suggests that arrangement.
https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjjhuWuy7LXAhVmB8AKHbp0Az8QFggoMAA&url=http://www.ti.com/lit/gpn/LM193&usg=AOvVaw2gVV6jhFiJBeB-3lvnZUD1

All input pins of any unused comparators should be tied to the negative supply.

I might be wrong but I thought I also read the output pin should be tied to ground too.

But there is a clarification here.

https://e2e.ti.com/support/amplifiers/etc_amplifiers__other_linear/f/18/t/608157

 

I was right. Here it is in this datasheet.

http://peshkin.mech.northwestern.edu/datasheets/Comparators/Comparator_LM193.pdf

All pins of any unused comparators should be grounded.

 

I don't understand your question. It is both emitter and receiver. I/R reflects off of reflective tape.
.

What I meant was if inputting into a Picmicro, a comparitor could be used.
Is it a OPB609 or a CNY70 for e.g.?
Max.

 

What I meant was if inputting into a Picmicro, a comparitor could be used.
Is it a OPB609 or a CNY70 for e.g.?
Max.

It is a pic. I decided to go this route. I will have to look into using comparitors on the pic in the future. I wanted to exteriment with opamps anyway so this was a good exercise.


I got it working. Swapped out R1 for a 30K and went over all my solder joints. It is working like a champ now.

My only issue now is a software glitch. I have am RPM percentage graph that isn't working right. Need to look into that one of these days.

 

Incidentally, what sensing medium are you using? Light to dark e.g.?
Max.

 

Incidentally, what sensing medium are you using? Light to dark e.g.?
Max.

I painted the spindle black matte. I have a strip of that reflective tape to reflect the IR.

 

It is OK to ground the output of a comparator because it is open collector and without a pullup resistor it will never go high.
Why is the +5V wire of J1 labelled GND?

 

.
Why is the +5V wire of J1 labelled GND?

Its not. Look again.

 

Its not. Look again.

It is confusing.