• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Triggering an IR receiver from a white object/paint

Status
Not open for further replies.

avz10

New Member
I have started to build a system/circuit to capture old cine 8/super 8 using the Cinecap programme.

I am having trouble to get the IR receiver being triggered when a white blade moves passed the IR receiver.

(There are three blades of which I have painted one white i.e. 2 black blades will pass before the white blade is supposedly required to trigger the system)

For the system I used the following-
A cine projector that can run as slow as 3 frames per second (Eumig 610D).

The moment that the frame "stops" for a split second, you need your camcorder to record the image and these frames are put together by the Cinecap into an AVI file. So when the frame stops, a trigger needs to be activated. In this case an IR receiver is the trigger.
(I also tried with a microswitch, but the capturing was erratic)

My video camera is a 3CCD Panasonic NV GS250

I have done the following: I converted a mouse (acknowledge James Rueben):



Used the following circuit (acknowledge James Rueben)::



Connected to an IR receiver:
OPB608A


http://www.datasheetcatalog.org/datasheets2/22/229105_1.pdf

The IR receiver and all my connections (converted mouse; circuit board) are working well.

My finger, or any white paper triggers a signal which captures a frame via my camcorder on Cinecap, BUT, the paper, my finger, etc need to be very close to the surface of the IR receiver to trigger the signal.

As mentioned before, I painted one of the three blades white, which will hopefully be triggered by the white blade going past.(acknowledge James Rueben):




It only works when I bring the IR receiver very close to the stationary blade.

If the blade rotates at 3fps, it does not trigger a signal, although I bring the receiver quite close.


What is the solution?:

I can see two possibilities
1. Either put something on the blade ? white paper, other material which the IR receiver will pick up (this is where I need advice and this will be the easiest),
2. Or get a stronger?? IR receiver.

I feel that there must be a way to produce a strong input by means of some material to create the trigger. I used PVA paint.

Any advice will be helpful. Will try it out in the next week or two.

Regards
Albie

I need to acknowledge the help I got from James Rueben in this regard- he initiated the project, with the initial input fro Ronypony.
Having Trouble with DIY Telecine (8mm) System - VideoHelp.com
 

avz10

New Member
Just more detail I want to add:

The distance is very close ± 1-2mm, before the switch is triggered. So if I wave a white piece of paper 1cm in front of the detector, it does not trigger it.

This is how the set-up should be:

Projector projects to a lens (the Eumig allows for speed to go down with a gear system to 3 fps), produce an ariel image which is captured at 3fps. This is not my set up.




Initially tried a microswitch, but this works erratic, especially at the start of film





Cinecap program- screenshot





The photo that I have posted of the IR receiver next to the blade, is not mine. I havent fixed mine, but this is how I painted a blade.

Someone advises that in stead of paint, I use a thin mirror or foil on the blade. Any comments on that?
 

Mr RB

Well-Known Member
I've worked with those little reflective sensors for many years in VCR spool encoders, and used some for hobby robots etc. It's no problem to detect white paper at 6" or more. You seem to be driving the LED with plenty of current so I think you have much too low a value resistor for R2, try a 100k pot there and adjust it until you get good sensitivity.

Once you get the sensitivity right, you might want to put a small cap across pins 1 and 2 of the sensor to reduce noise, its probably best to put the cap as close as possible to the video camera plug (or in the plug) which will reduce line noise a bit too.
 

avz10

New Member
Thanks for the reply. Please remember I am quite new to this and to get the mouse modified and actually working, is a major achievement for me!!

You seem to be driving the LED with plenty of current so I think you have much too low a value resistor for R2, try a 100k pot there and adjust it until you get good sensitivity
Am I correct, do you mean a 100k resistor in stead of my 7k resistor?

I am not sure what you mean by "adjust it until you get a good sensitivity"- are you referring to the IR receiver?

Once you get the sensitivity right, you might want to put a small cap across pins 1 and 2 of the sensor to reduce noise, its probably best to put the cap as close as possible to the video camera plug (or in the plug) which will reduce line noise a bit too.
You have lost me a bit here. Do you mean that I should "bridge" the wires between pin one and two as close to the modified mouse with a resistor to reduce noise? If so, what resistor strength are we talking about?

Just a last question. On other forums, people have suggested to use a thin mirror/foil on the blade; or shine a IR light towards the blade, or paint the blade with an aluminum colour. Your thoughts on this?

Thanks for the help

Albie
 

Mr RB

Well-Known Member
Ok, that sensor will EASILY pick up white paint. Given a choice i'd use a matte white paint, no need for fussing with mirrors or additional IR lights etc.

A distance from the IR device to the white surface should be about 2mm to 3mm, about 1/16" to 1/8" or so.

Replace the 7k resistor with a 100k adjustable resistor, normally called a "trimpot" you can get one from radio shack or digikey etc. This is how I tune those sensors, just adjust the trimpot (which controls the sensitivity) until it triggers reliably with the paint surface and distance you require.

Note! the trimpot has 3 legs, connect to only two legs; and it must be the middle leg and ONE outside leg.

Then once working well you can put a spot of hot glue on the trimpot to secure it at that position, or remove it and measure the "adjusted" value, and replace it with a normal resistor of the right value.

The people on other forums seem to have complicated this, all you need is to adjust R2 to the right value of resistance to give you the desired sensitivity.

What I said about the capacitor, forget that for now until you get it triggering reliably, then you can measure the final value of R2 and we can suggest a value of capacitor suitable for noise filtering.
 
Last edited:

avz10

New Member
Thanks for that. I will be home this weekend, but might only be able to work again on the project in 2 weeks time.

Thanks

Albie
 
Last edited:

avz10

New Member
I'm not going to have time today to buy resistors, but this what I'm gonna do, as said before:
Buy the resistors and a photo interrupter at the same time. Will try the resistors, but if I struggle, will try to install the interruptor.







From the side:





This is basically the available space. I think if I glue rigid plastic to a blade, it will work quite well.

Please help me to choose a photo interrupter from this site, as this is where I can buy:

Optoelectronics and Displays | Optoelectronics | Slotted Optical Switch

Thanks for the responses
Albie
 

duffy

Well-Known Member
Ok, that sensor will EASILY pick up white paint.
...
Replace the 7k resistor with a 100k adjustable resistor, normally called a "trimpot" you can get one from radio shack or digikey etc. This is how I tune those sensors, just adjust the trimpot (which controls the sensitivity) until it triggers reliably with the paint surface and distance you require.
I suggested he remove that 7k completely when he posted this item the first time. Optos don't have much sink capability, the mouse's internal pull-up resistor (actually a current source, I looked up the chip for him) is about in the range he needs for that opto.
 

Mr RB

Well-Known Member
Good point, I was under the impression that the "out" ont he schematic went straight into a digital input pin on the chip.

Hey AVZ10, I wouldn't use a opto interruptor. :( There is more chance of snagging on the rotating parts, clearances are tighter, mechanical installation is fussier, they also are much harder to inspect the lenses and clean, and they can trap dirt, grease globs etc in the gap, etc.

All round they are less of an ideal solution than the reflective device you already have. There is no reason to think the refective device won't work or will be "harder" to get working. It is ideal for the task you have.

I think you are only one resistor change from a reliable working design!

Why switch to a worse choice of opto, then have all the development time again measuring more voltages, choosing more resistors etc etc??
 

ccurtis

Well-Known Member
It only works when I bring the IR receiver very close to the stationary blade.

If the blade rotates at 3fps, it does not trigger a signal, although I bring the receiver quite close.


That's the crux of the problem as I see it. The sensor is sensitive enough to detect the blade. It's reasonable to conclude that the response time is not good enough.

I experimented with my computer mouse. If I wave a 3 in wide envelope up against the flat surface, above a certain speed, the mouse pointer on the computer screen doesn't move. Slower speeds, the pointer moves. Maybe that observation is related to your experience with the rotating blade. Maybe, but maybe not.

A graph on the data sheet for the sensor shows that a lower load resistor gives a better rise and fall time, with 1000 ohms being the highest value shown, and the rise and fall times are increasing at a very high rate from that point.
 

duffy

Well-Known Member
At 3fps, that's 333ms, which is a very long time. With a 100k resistor that would give you a time constant for 3.3µf worth of capacitance, nothing on there even close to that. In spite of the moving vs. stationary results I don't think it's an issue with too much resistance causing a slow rise time - it looks like a simple current sink issue because the resistance is too low.
 

ccurtis

Well-Known Member
Yes, the interval between pulses is 333mS. Looking at the photos, the blades have a considerable gap between them, a gap appearing to be well wider than the width of a blade, giving a pulse width of, say, roughly 30mS, maybe less. I can't tell any better than that from the photos and I may be wrong about the gap, as the other blades appear to be hidden except for what looks like a piece of one.

30 mS is still a long time, but the rise and fall times are going quickly off the datasheet chart at beyond 1000 ohm load resistance and beyond. That's without any external load capacitance. If the rise and fall times increase by a factor of 100 at the load resistance in the OPs circuit, a 30mS pulse becomes a sliver.

Just my two cents. I can't say what the cause of the problem is. I don't know the width of the blade and I don't know what the sensor is driving. I don't know that what the sensor is driving can even respond to a 30mS pulse if the sensor does reproduce it.
 

avz10

New Member
I need serious help. This is my set-up so far:




On request, I tested values at different points, using the original configuration (R1 180 Ohm; R2 6k8 Ohm)

Here are the results:

With the IR detector facing away from the blade:

A: 1.3V
B: 3.4 V
C: 3.4 V
D: 4.25 V

With the IR detector facing towards the stationary blade:

A: 1.3V
B: 1.55 mV
C: 3.4 V
D: 4.25 V

Projector off, blades stationary and turned away from the IR receiver: measure the 5 V (point E), 3.3 V (point F) and signal (point G)
.


E: 4.25
F: 3.4
G: 3.4

Projector off, blades stationary and white painted blade in front of the IR receiver: measure the 5 V (point E), 3.3 V (point F) and signal (point G).

E: 4.25V
F: 3.4V
G: 155mV

As a final check on the cables, measure between the zero volt point on the detector circuit and the zero volt point on the mouse. The projector state doesn't matter for this one. This should of course be zero or very close to it.
0.02mV

So it looks as my point E just produces 4.25V.

With R2 replaced by 100k Ohm (the resistor is not a "trimpot"; it just says 100k

B (Output) Stationary: 3.1V (facing away from white blade)
Facing white blade: 0.07 mV
Moving: 2.4Volt
Does not trigger “Capture” on Cinecap

With R2 replaced by 100kOhm and R1 replaced by 68 Ohm:

B: Stationary: 3.38V (facing away from white blade)
Facing white blade: 0.37 mV
Moving: 1.68Volt
Does not trigger “Capture” on Cinecap

So, I am not sure what I am doing wrong.

I bought this Slotted IR optical switch yesterday (OPB365T), just in case I get stuck: Datasheet:
Details and datasheet on part: OPB365P55



Does the beam need to be broken in that first 1mm transparent part, or does the object need to go deeper in the slit?



If so, I can try to mount it close to two movable parts, that move only when a frame advances (so it is stationary and then either move down- that white piece)


Or to the right:


These two possible trigger points are very consistent.

So, should I try this optical switch, I need to know the following
1. How deep should the object need to go in the slit (because if it needs to go deeper, it might be difficult)II thought of attaching a rigid plastic to the "moving" side of one ot the trigger points.
2. What wires should be connected to what pins (For me this is very important, being a newbie to electronics)
3. Do I need resistors?

Thanks
Albie
 

duffy

Well-Known Member
What's your "double click" speed set for on the mouse?
 

Mr RB

Well-Known Member
Hmm now you got troubles...
E: 4.25
F: 3.4
G: 3.4v (signal out when dark)

and

E: 4.25V
F: 3.4V
G: 155mV (signal out when light)

The voltages seem good at the sensor end. So changing the sensor won't help you.

It really looks like a speed issue, can you confirm that the sensor will cause ONE capture if you manually move it over the white blade?

I think duffy is on the right track with your mouse click speed. It is also possible the microcontroller in the mouse does software 'debouncing" that will limit the speed that you can make successive clicks.

What speed are you expecting to get? I was under the assumption that this projector would be going at 3 frames per second? Surely the mouse click should work at 3 times a second as you can casually tap your finger at that speed.
 

duffy

Well-Known Member
I think your switch is working, but maybe you have the double-click speed set too fast and it is triggering the wrong event handler. Go to settings -> control panel -> mouse -> buttons -> double click speed and set it to the SLOWEST setting.
 
Last edited:

avz10

New Member
I have used my IR detector and mounted it to the rear trigger.

R1: 68 OHM
R2: 100k

Picture of the triggering system- works wel, up to now:



The silver paper moves up and down and trigger the IR detector

Any advice?

Albie
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top