motorized wigwag

[peteski]

peteski
That is a broad statement - what do you consider detection? a low of < 0.2v?
What is your IR transistor load resistor? How is it connected common emitter -driving an opamp etc?
Bread board my circuit with just the IR diode and IR transistor and let me know at what distances you get a true low   ie <0.2 volts
at various diode currents.

Al

The statement was broad but valid.

I have to dig up the design paperwork.  I was using a 10k load resistor and feed the output to a Low power Shottky TTL buffer input.

Looking at your circuit, it seems to me that increasing the R1 value to much higher value (maybe 22 or 47k) should increase the sensitivity.  You could do the same to the other 4.7k resistors in that circuit. Right now I don't have time to breadboard your circuit and experiment.

[alhoop]

The statement was broad but valid.

I have to dig up the design paperwork.  I was using a 10k load resistor and feed the output to a Low power Shottky TTL buffer input.

Looking at your circuit, it seems to me that increasing the R1 value to much higher value (maybe 22 or 47k) should increase the sensitivity.  You could do the same to the other 4.7k resistors in that circuit. Right now I don't have time to breadboard your circuit and experiment.

"The statement was broad but valid."

I don't consider it valid under all circumstances. Consider:
From 'Texas Instruments Optoelectronics and Image Sensors' 1990
Applications -Interfacing discrete phototransistors to digital logic circuits:
Assuming 95% of ambient light is blocked the transistor load resistor calculates to be
between 11 kohms and 19 kohms.
assuming 80% of ambient light is blocked  the transistor load resistor calculates to be
between 2.7 kohms and 6.5 kohms.


"Looking at your circuit, it seems to me that increasing the R1 value to much higher value (maybe 22 or 47k) should increase the sensitivity.  You could do the same to the other 4.7k resistors in that circuit."

True and I have done that and it does increase sensitivity but when designing a circuit that you have
no control over where it is going to be used one had better use the formulas from the above referenced book or similar ones.
 So my choice of 4.7 kohms was a tradeoff.
I have used a load resistor as high as 22 kohms for a circuit that was used exclusively in my house.
My train building has windows that allow the great IR source in the sky to affect these circuits unless
ambient light is blocked to some extent. Look how Boulder Creek Engineering blocks ambient light
on their "TrainBoss" IR detectors. You think they hung those big black  things there for looks?
Regards
Al

[peteski]

"The statement was broad but valid."

I don't consider it valid under all circumstances. Consider:
From 'Texas Instruments Optoelectronics and Image Sensors' 1990
Applications -Interfacing discrete phototransistors to digital logic circuits:
Assuming 95% of ambient light is blocked the transistor load resistor calculates to be
between 11 kohms and 19 kohms.
assuming 80% of ambient light is blocked  the transistor load resistor calculates to be
between 2.7 kohms and 6.5 kohms.


"Looking at your circuit, it seems to me that increasing the R1 value to much higher value (maybe 22 or 47k) should increase the sensitivity.  You could do the same to the other 4.7k resistors in that circuit."

True and I have done that and it does increase sensitivity but when designing a circuit that you have
no control over where it is going to be used one had better use the formulas from the above referenced book or similar ones.
 So my choice of 4.7 kohms was a tradeoff.
I have used a load resistor as high as 22 kohms for a circuit that was used exclusively in my house.
My train building has windows that allow the great IR source in the sky to affect these circuits unless
ambient light is blocked to some extent. Look how Boulder Creek Engineering blocks ambient light
on their "TrainBoss" IR detectors. You think they hung those big black  things there for looks?
Regards
Al

Which specific digital logic is the book talking about?  TTL, LS-TTL, CMOS?  They all have very different input current requirement which plays big role in determining the load resistor.   Whatever it i, my choice to user 10k resisttor seems to follow the book's recommendation. 

Then there are book-smarts and actually building a circuit and experimenting.  I choose to get my theory from the book, then experiment and come up with a circuit which works.  Books like that are for reference, not absolute gospel which has to be blindly followed.

I never leave the phototransistors uncovered and exposed to ambient light. I always place them in some sort of enclosure which narrows its view angle and aim it at the light source. Sometimes all it takes is a piece of black heat-shrink tubing and some black paint.  It doesn't have to be bulky.

Your circuit works (with high-power IR LEDs) and my circuit also works (using IR LEDs with less than 20mA going through them). I guess we'll just leave it at that.

[alhoop]

Which specific digital logic is the book talking about?  TTL, LS-TTL, CMOS?  They all have very different input current requirement which plays big role in determining the load resistor.   Whatever it i, my choice to user 10k resisttor seems to follow the book's recommendation. 

Then there are book-smarts and actually building a circuit and experimenting.  I choose to get my theory from the book, then experiment and come up with a circuit which works.  Books like that are for reference, not absolute gospel which has to be blindly followed.

I never leave the phototransistors uncovered and exposed to ambient light. I always place them in some sort of enclosure which narrows its view angle and aim it at the light source. Sometimes all it takes is a piece of black heat-shrink tubing and some black paint.  It doesn't have to be bulky.

Your circuit works (with high-power IR LEDs) and my circuit also works (using IR LEDs with less than 20mA going through them). I guess we'll just leave it at that.

For Posterity.
Al

[peteski]

For Posterity.
Al

You just quoted my previous post. I don't get it.

[bdennis]

N scale Wig Wag taken from a layout I visited last night.
<iframe width="560" height="315" src="

frameborder="0" allowfullscreen></iframe>

[peteski]

N scale Wig Wag taken from a layout I visited last night.

Interesting.  But I think that I like Al's mechanism better.

BTW, was that H0 scale layout?  And shouldn't the signal be facing the other way (or we are viewing the back of it)?

[bdennis]

Petski.
Nope this is N scale. But is Australian prototype.. So we drive on the left so the cross backs and signal are on the "Correct" side..

[Cajonpassfan]

Brendan, whose layout?
Peteski, at least the ones I'm familiar with here in California, they had lights on both sides, see pic.
Don't know about the cross bucks on the left though...
Ah, Aussie, that would explain it.
Otto K.

[ Guests cannot view attachments ]

[bdennis]

Otto,
Darren French in Melbourne owns the layout.

[peteski]

Petski.
Nope this is N scale. But is Australian prototype.. So we drive on the left so the cross backs and signal are on the "Correct" side..

Impressive!  The track looked like it was H0 (which means it looked really fine).  I had no idea that they used WigWags in Australia.  But that sure explains its location in relation to the road.

Otto: does the light flash or stay steadily lit on the American WigWags?

[bdennis]

Petski,
The track is Peco Code 55.

[Cajonpassfan]

Otto: does the light flash or stay steadily lit on the American WigWags?

I can only speak to the ones I know on the old Santa Fe, and those stayed on; no flash. Very quaint
And thanks for the info Brendan. Some very nice modeling down under
Otto K

[peteski]

Petski,
The track is Peco Code 55.

That just goes to show you that good ballasting and weathering job makes a huge difference in the track's appearance.  Not that there is anything wrong with the peco track. 

Otto: thanks!

[C855B]

Brendan, whose layout?
Peteski, at least the ones I'm familiar with here in California, they had lights on both sides, see pic.
...

Otto, is that Hwy 138?

And yes, I can confirm the light did not flash. We had several in the nearby town where I grew up. All were gone by the mid-1960's.