CNY70 Wiring Problem with TM4C123

Hello Amit

I was thinking of supplying power to the CNY70 from the +3.3V pin (as seen in the image). Ground is also from the TM4C's GND port. I would connect the output of the CNY70 to PB0. This was the plan.

And a new idea comes into my mind:

1) I get a 5V battery, and supply 5V and GND to CNY70.

2) I replace the 10K resistor with a lower one, so I can obtain lower CNY70 output voltage. But I can't find the internal resistance value of the CNY70's transistor.

Robert Adsett said:

recommending Horowitz & Hill, they have a short section on drawing clear schematics

Roger that - and my firm's (sole) copy "disappeared" w/last year's intern crue.    (not that any accusation is (much) suggested)

Does it make sense to urge poster - prior to launching test/verify of that external circuit - to first emply a simple switch (w/proper pull-up/down Rs) and exercise the target GPIO pin - so that its operation - in response to valid signal levels - is confirmed?

Skipping that - both GPIO pin handling - input reading - AND the external circuit all rise to, "Likely Suspect!"    Two of (those three) may be eliminated - quickly/smartly - by the simple application of (drumbeat) KISS...

Thank you - thank you very much!     (think Elvis - sotto voice)

Note that our firm has "stylized these" - small, clean pcb - w/plug-in cable - used repeatedly & proves robust... (our board designs - most always - employ mating headers for, "eased - repeat" use.)

Anything so that we do not (always) have to stop/think/search for - and "re-invent the wheel!"    (death to any small biz - and (unfortunately) we see so often @ clients/others...)   [i.e. death by 1000 cuts]

After all the helpful responses, I realised that it is impossible to solve this problem without trial and error on the breadboard. So I bought and used a multimeter for the first time. Took some time to get used to it. Sorry for the delay.

In order to keep things simple as you suggest, and not to harm the MC, I have tested the CNY70 without connecting to MC. Used a 9V battery as voltage source. I have built the setup according to the figure below. The figure is screenshot from a Spanish youtube video (I have no Spanish, but the video helped alot). Please note that, I have modified the figure according to final resistance values I use. I've changed the resistance value in the circuit till lowering the AB voltage to 2V max for safety.

(Sadly, I wasn't able to find the book, and wasn't work on creating proper schematics neither.)

Hopefuly my last question comes: Since I have output voltage less than 3V, should I connect the B to a MC port set as ADC? If so, how should I select the resistance value between B and the port? I couldn't find the maximum possible allowable current limit for the GPIOs.

P.S: Adding the link of the Spanish youtube video. May help people who may have the same problem in the future.

www.youtube.com/watch

Robert Adsett72 said:

Unless you want to measure variances from different surfaces.

As responsive and descriptive as this poster has been - point Robert brings - flew (far) over this reporter's dull head.    Indeed - the "Anti-KISS" use of the ADC may be justified - if signal level gradations - rather than simple logic levels - is "in play."

Poster may do himself - and helpers - more good by (bit) detailing his application.    Having to "guess" is very much second to a clear, complete project description...

Dear Friends

In my test setup is a vertical transparent plastic pipe, with tennis balls inside. The balls can be in any color (yellow, orange, blue etc). Example display of the pipe and balls can be seen in photo below:

I will drill a small hole on plastic pipe from side, and open the pipe from top and bottom. Will use CNY70 like a level sensor.
All I care is if there is a ball at the predefined level or not. The colors of the balls are not important. This is the most easy to use sensor that seller in the radioshack offered, ofcourse there may be much easier/better applications.

I must admit that, it is easier for me to find/search a random safe spot for both the circuit and MC, and dealing the output signal via ADC. Because I'm quite a beginner in electronic components and circuits, but i'm abit better and more practiced in programming. The system doesn't have to be very robust, and efficient at the momement. But I hope to create better circuits as I read, learn, and practice more.

Substantial application detail now arrives - good that - thank you.

Some quick observations/comments:

a) will the "real" application actually detect tennis balls?    (your photo/illustration may (again) display your resourcefulness in, "gathering source material.")    Ask that as there must be minimal/marginal "reflectance" from such surface - but surely near total blockage of an imposed, "light-beam."    Reflectance vs. Light Beam blockage is most likely to impact, "Rules of design engagement."

b) You note, "Seller in the RadioShack" - in the US - RadioShack is (effectively) "no more" and some have suggested that (sometimes) components supplied - had not met all performance test requirements.    I am simply reporting (thus hearsay) but such may signal that your (assumed) single CNY70 may perform "outside" of normal/customary device specifications.    Multiple devices - likely bearing newer date codes - may present better & more consistent operation & recovered signal performance.

c) ball velocity may prove an issue - as will the separation - between adjacent (or successive) balls in transit.    And note that gravity's "pull" will accelerate ball descent - which may be further influenced by the rate of "launch."   (launch = initial speed of propulsion)

Surely more issues lurk.    (I leave (some) meat on the bone for Robert, Veikko, Luis)

Running a small tech firm - while your "vertical pathway" is bit unexpected - your search for, "Conveyor's,  "missing or improper object" detection methods" is sure to provide very fertile ground for your exploration/exploitation...

Dear Friend

Thank you for all suggestion and help, especialy about the "Clear Object Detection" part. I have found quite a good industrial sensors in the internet after your suggestion, which can be/should be used in the final product if all goes well. But I don't have much chance to buy one of them at the moment. I see your concerns, I will try to answer them:

a) Trust me this small sensor is a non-critical component in the current design and most of the critical parts had already been designed. It is only a small part in the design. I'm thinking of CNY70 and its circuit as a "replecable" sensor or design in the future, as I (hopefuly) learn more about electronics. At the moment all I need is a some howworking sensor, that doesn't have to be very efficient, that doesn't have to detect balls %100 all the time. I'm trying to avoid much money/time spend on this prototype #0. Resources are very criticaly scarce, until I build a dirty but somehow working device, and some how get funded.

I have checked the reflectance of the tennis balls with CNY70, altough they don't have as much as reflection as the white plain paper, there is some. At least I have detected some voltage change (in the order of 1.5V) with the multimeter when doing the tests. As the balls get dirty and darken, I guess I will have some problems with the detection. But it is not the a concern at the moment, because the sensor would already been replaced with a better one such as clear object detection photoelectric sensors at that time. Bytheway if I have a problem with the reflection from the pipe, I will replace it with dark plastic pipe of similar diameter.

b) You are probably right about the seller, and the device. But I got a feel that I'm very close to operating this sensor. It is sad, but that will be the only operating electronic sensor in my tiny little arsenal :)

c) The balls will stationary during the sensor check. I will design another dirty and cheap mechanical gate for the bottom of the pipe. (I had already checked some old patents). I will not count the number of balls. All I need is if there are enough number of balls inside the pipe before the operation starts.

Hope this answered most of your concerns. Thanks for all the assist my friend :)

Thank you - as well - always nice to receive quick/detailed feedback - just as you have supplied.

My sense - just working w/the facts presented herein - you'll speed, ease & enhance your design by switching from, "Reflective" to Light Blockage."    It should not be hard for you to locate properly matched, "Light Transmitters & Receivers" (often IR - think TV remote) and to simply detect the "LOS" (from NASA - "Loss of Signal") when a ball blocks the receiver.     (you do not have to utilize the TV code modulation scheme - simple, pulsed On/OFF should be sufficient.

As for funding - when we started my small firm (dinosaurs ruled the earth) we faced similar issues - most were solved by our approaching qualified users (i.e. buyers)  of our proposed projects - who eagerly provided financing.

I suggest that you some way/how consult legal & accounting experts - to assist you in such implementation - prior to (and if) you make such an approach...   And again - your study of "Conveyor Detection Products" will prove most helpful.

emrah erden said:

I have found quite a good industrial sensors in the internet after your suggestion, which can be/should be used in the final product if all goes well.

Definitely the approach to take if you can.  They will be more expensive in parts but less expensive to integrate and they will have taken care of a lot of concerns so you don't have to anticipate and work around them.

emrah erden said:

I have checked the reflectance of the tennis balls with CNY70, altough they don't have as much as reflection as the white plain paper, there is some.

Consider sanding the balls (less fuzz on the ball to absorb/scatter) and/or painting the ball to be more reflective.

Another item to consider is that you may get reflection from the other side of the tube. 

cb1, I think mentioned using an interruption sensor instead. That may be a good idea and for some the other 1/2 of the sensor is simply a mirror. Some sensors will be easier to align than others.

Robert

Robert Adsett said:

cb1, I think mentioned using an interruption sensor instead. That may be a good idea and for some the other 1/2 of the sensor is simply a mirror

Precisely why I suggested that experimenting poster make good effort to see how the industry - in general - attacks.    Re-invention of the wheel - while suggested by some here - really makes little sense.

Robert's suggested, "mirror at the tube's side - opposite of the source transmitter" indeed demands a fine initial adjustment - and forces both a stronger & more narrowed light beam - to prove robust & adjustment-free over time.   

In our book - separate transmitter & receiver (one at each tube side) is far simpler, faster to install & align - and proves far more able to, "maintain that needed alignment."   No matter how well a system performs, "post critical alignment" - if it fails later - clients become most unhappy!    Long term robustness should be a major goal - and separate TX & RX-  proves best means to achieve that.

As I promised I'm back again to inform you that I found the solution, and it was tested and works. Hope this post helps disoriented newbies like me in the future.

Steps I Have Followed:

1) Built up some electronics background with the help of Edx courses. Electronic Interfaces course helped me a lot to gain hands on experience.

 https://www.edx.org/course/circuits-electronics-mitx-6-002x-0 (Circuits and Electronics)

 https://www.edx.org/course/electronic-interfaces-bridging-physical-uc-berkeleyx-ee40lx-0 (Electronic Interfaces: Bridging the Physical and Digital Worlds)

2) I have asked stack exchange forums, how CNY70 works and how to read the data sheet. And got tons of good information:

http://electronics.stackexchange.com/questions/179115/understanding-the-cny70-reflective-optical-sensor

Solution to Problem:

I have built the CNY70 circuit according to data sheet specifications, and connected the CNY70's output and 1.65V to LMC6482 Operational Amplifier (Op Amp) as Robert Adsett72 stated. Op Amp is railed to 0 and 3.3V. Now I get 0 or 3.3V, no need to use ADC of TM4C.

Plese note that I have obtained 1.65V by using voltage divider on LF33ABV 3.3V voltage regulator. Energy is suplied from battery.

Thanks for all messages and help.

Robert Adsett72 said:

A couple of posts seem to have disappeared here   

In the (now) immortal words of (WW2 prison guard) Sergeant Shultz, "I know nothing!"

(note all declarative sentences - w/improperly appended "?" - are duly flagged...)

I had come up with a solution after weeks of study, and my post seems like got deleted..

Edit: Good that my post is back. All fine now

emrah erden said:

2) As you see I'm quite noob at electronics, and electronic components. I've started taking Edx: 6.002x Circuits and Electronics. Which courses, or books do you suggest further to study?

 Second question you are on a good boat so choose course there and try graduate.

 When you are more acquainted with all devices and how to do energy calculation evaluate a different approach:

 Don't drive infrared emitter by DC 20mA CW but use an High energy near 1A (check data sheet)  short pulse averaging to CW 20mA, when you drive with ten time the current of CW your energy is the same but pulse carry 10 time more energy and S/N noise is improved by 10dB (20dB in voltage).

 Again you can reveal pulse by measuring width than using a complex ADC evaluation and clearly modulation also reveal if light come from reflection or sun exposure. TV and appliance transmitter IRDA too use an high power narrow pulse to improve transmission over distance.

 Don't use GPIO to drive LED, must use a mosfet driver to obtain lower impedance needed by. Use capacitor close to pulse generating circuits and dont use spare wire, energy of short pulses radiate away from circuits.

Roberto Romano said:

...or you get out of business in a few stroke.

As do many tennis players when up against hard (and line hitting) superior players...    (that's clever word-play Roberto - not to be missed...)

Great detailed tech advice as well.    We note that the "plastic tube" itself is a substantial static generator.   I believe there are special chemical sprays & methods to reduce the undesired ESD which such tubes may create...

emrah erden said:

That was quite a nice and technical approach to reduce the noise, and to get more reliable results. Never thought pulse inputs have such amazing uses. Your previous post was also quite important for protecting the circuit against ESD. I will add the capacitors of the previous post, but I'm just a n

... Snip...

Hi Emrah, thank for selecting as correct answer,  as Amit wrote I forever supported this policy, please check all answer are close to solving your problem not just one. I am sure none of us who helped you, are there just to collect point like some useless super guru. I trust on correct honest sharing of income to all contributed to solution.
 This is how scientist perceive knowledge and technology, who come here to collect point or do simple "social network" activity is a big loser wasting his and community time.
When you can take some time invest it to learn some old scientist phrases, they are forever a guide to correct living but forever take care of you against actual world.
 Have good time as I wish for all good people are here sometimes hit by worst policy of insane commercial "bean counters".