Long wires to transistor base a problem?

[fourfathom]

I'll move and vary the resistor. The darlington was added per the suggestion of fourfathom.

Well, I *did* recommend the Szilaki pair over the Darlington to get the advantage of a single base-emitter junction.  But, not having the photodiode specs, I suspect that the extra gain of the Darlington will still provide a big boost.

[LKO Railroad]

I'll move and vary the resistor. The darlington was added per the suggestion of fourfathom.

Well, I *did* recommend the Szilaki pair over the Darlington to get the advantage of a single base-emitter junction.  But, not having the photodiode specs, I suspect that the extra gain of the Darlington will still provide a big boost.

I have never built anything with a Szilaki pair. Had to Google it. I have made things with Darlington pair so was more comfortable with that option.

[SilverSolder]

I'll move and vary the resistor. The darlington was added per the suggestion of fourfathom.

Well, I *did* recommend the Szilaki pair over the Darlington to get the advantage of a single base-emitter junction.  But, not having the photodiode specs, I suspect that the extra gain of the Darlington will still provide a big boost.

I have never built anything with a Szilaki pair. Had to Google it. I have made things with Darlington pair so was more comfortable with that option.

They really work the same as a Darlington, just with opposite polarity transistors.

The cool thing about a breadboard is that you can try all the crazy ideas and see which works the best!

The bad thing about a breadboard is that you can try all the crazy ideas and see which works the best...   (you end up playing a lot! - but it's kinda fun)

[LKO Railroad]

Can you share the circuit of yours, I am in sort of designing the similar circuit 

It is on page 1 of this thread.

[LKO Railroad]

Thanks for being patient.

Tried both the Darlington pair and, after rearranging the circuit a little bit, the Szilaki pair. Essentially same result with both. The circuit becomes far too sensitive to even trivial amounts of ambient light. The installation area is generally dark but not pitch black. The ambient light level varies somewhat.

Tried series R up to 10M between diode and base. No joy. I then tried a divider network on the base. This worked but by the time I biased it down enough to solve the room light problem with sufficient margin, my object / no object voltage swing was quite narrow, less than 1v. No good.

All this lead me to ask why boost the signal if I just turn around and clamp it down? Must be the tea cup thing. That made me wonder just how much of a jackhammer my tea cut is resting on. I rebuilt the original single transistor circuit and incorporated your suggestions except the clamping diodes. I ordered some 1N5817. I used 30cm of shielded cable between diode and Q base. I tried every device I could find - cell phone, cordless phone, CFL, electric drill, cordless drill, AC grinder, even an old B&D router that throws so many sparks from its brushes it should be in a 4th of July parade. Nothing. No false triggers. The circuit carried on working seemingly unfazed by any of the devices operating within inches of the cable.

Is it possible the shielded cable is sufficient? It sure seems to be. Should I throw a couple clamping diodes on it when they arrive and call it done?


 

[floobydust]

Go with what works on your breadboard- but leave provisions on the PCB for drama.
A 10-47nF cap at the receiving end of the long cable run, maybe an LED indicator for troubleshooting, extra ground pad for cable shield if you use it. I don't know where the diode is you are mentioning or what it's for.
You can improve immunity to ambient light with additional IR filters (I'm assuming the photo-diodes are not clear lens) or using a tube, and make sure the reflective angle is optimal. I doubt all the cars are the same size on a layout.

[LKO Railroad]

Here is the circuit I described in the previous post.

The clamping diodes are a result of @fourfathom comment and the subsequent search that led me here: https://www.digikey.com/en/articles/protecting-inputs-in-digital-electronics. They would go between the 1M and 1K resistors on Q base.

The emitter LEDs and detector diodes are already in tubes (uncollapsed heat shrink). Have been from the get go. Needed so they could sit side-by-side.

The three E/D pairs are in a stack perpendicular to the cars. Lowest one for depressed center flatcars, middle one for standard flatcars and couplers, top one for tank cars. These turned out to be the picky cars. Boxcars, hoppers, etc. hit more than one E/D.

[SilverSolder]

If the circuit works in the environment it will operate in - you are on to a winner!

[floobydust]

I would not use a Schottky diodes there because they have high leakage current relative to the small photo-current, so I imagine the circuit losing sensitivity.
A 1N4148 across E-B will protect the transistor from -ve spikes, and the E-B junction will take the +ve spikes. I would ensure the photo-diodes would not see any transients or overload, if ESD hits happen on that node as you are connecting the cable. That is the usual damage, happening during installation.

[LKO Railroad]

I would not use a Schottky diodes there because they have high leakage current relative to the small photo-current, so I imagine the circuit losing sensitivity.
A 1N4148 across E-B will protect the transistor from -ve spikes, and the E-B junction will take the +ve spikes. I would ensure the photo-diodes would not see any transients or overload, if ESD hits happen on that node as you are connecting the cable. That is the usual damage, happening during installation.

Got it. Will take the necessary precautions during installation.

[dietert1]

Nobody suggested a shielded cable for signal transfer from photo diode to amplifier/controller?

Regards, Dieter

[LKO Railroad]

Nobody suggested a shielded cable for signal transfer from photo diode to amplifier/controller?

Regards, Dieter

Yes. SilverSolder did. The reason I tried it. Appears to have been a magic bullet.

Will the 20cm run inside a shielded cable or something like that?

[dietert1]

No, you still did not understand. I mean the signal should be inside a shielded cable all the way (3 m). You could try an audio cable.

Regards, Dieter

[LKO Railroad]

No, you still did not understand. I mean the signal should be inside a shielded cable all the way (3 m). You could try an audio cable.

Regards, Dieter

I'm confused. The signal is inside shielded cable all the way from photo diodes to the breadboard. Alternate placement allowed original wire run of 3m to be reduced to 20cm. My test setup uses a 30cm cable for margin.

Perhaps I drew the schematic incorrectly. This better illustrates the shielded cable connections.

[dietert1]

My fault, did not get all details of this thread. Anyway, once you use a shielded cable, a length of some meters doesn't matter. Your signal receiver is low impedance.
For pulse oximeters people use some meters of double shielded cable that have the LED signals between inner and outer shield. Inside the inner shield there are two wires for the photo diode to keep noise as small as possible.Those photometers use modulation at a bandwidth of about 100 KHz.

Regards, Dieter

[Manul]

I read through this thread and I do not understand, why complicate so much. This system has serious drawbacks, high sensitivity to ambient light, sensitivity to EMI. Presense / obstacle detector should have a carrier. Common ones are 38 and 56 kHz. Then you can take a ready made CW IR receiver which outputs nice low impedance digital signal. Solves most problems about ambient, noise and general reliability. For example, if I remember correctly, TSSP4038 works with CW.

[floobydust]

If you can get away with shielding ambient light using filters or tubes, shutters etc. then it's good enough.

OP could use consumer IR remote receiver modules with say 38kHz LED drivers. That gives high immunity to ambient light. But it costs a lot with 60 nodes unless you go offshore. The IR modules are wide-angle and optimized for serial data - not for object detection. You have to buy special IR rx modules with the correct AGC timing for object detection, for light curtains or garage door use . Or get around it making a more complex IR emitter pulse train but with detection delays.

Vishay TSSP4056 Sensor for Fast Proximity Sensing Arduino example
Vishay App Notes on IR receivers

[LKO Railroad]

The ambient light problem is specific to the two transistor circuit. The single transistor circuit does not have an ambient light problem. At least not in the intended location.

Undoubtedly there are any number of clean slate ways of going about this most of which I am sure would be much better than my solution. Earlier in the thread I mentioned completing this project with parts I have on hand if at all possible. The IR pairs I have many. Same for the CD4584. That is how they became the basis of the design. I'm an old man that already has way too many bins of electronic junk that my kids will just throw away when I'm gone. If I can build a working system while bleeding down the parts cabinets then that is a double win. My total investment so far is $0 and 0 parts added to the bins. (I cancelled the Schottky diodes order. Thanks @floobydust)

At this point the circuit appears to be working great even if of an inferior design. This is not a production item, merely a one-off detection system for a model train. The consequence of failure is limited to nothing more than a sigh and "Aww, that's a shame."

 

[Manul]

I'm an old man that already has way too many bins of electronic junk that my kids will just throw away when I'm gone.

I'm sorry for off topic, but this sentence was sadly touching. All the best to you. And keep the thread updated

[SilverSolder]

[...]  I'm an old man that already has way too many bins of electronic junk[...]

Is there such a thing??? 

I'm with you on "use what you have", it is very modern to be conscious of avoiding the build-up of junk.  For many (most) hobbyists, a scope and a Digilent Analog Discovery would be worth as much as an entire rack full of period test equipment...  just too easy to pile it on...  ask me how I know!

[LKO Railroad]

At what point did it go from solving a problem to simply having fun?

The original spacing between sensors was 17". I tried doubling up IR diodes on each sensor. Dark signal stays well below the threshold. Actually, very little difference at Q collector. This means I can reduce the sensor to sensor distance to a little over 8" which solves yet another issue - lone car sitting between sensors.

I used the remaining gates to alternate the IR LEDs so doubling of sensors didn't double the load on my 7805.

Also made the display panel LEDs all on at low brightness. Object detect makes them bright. That will make the panel more aesthetically pleasing when no objects are detected.

Building a prototype now. Will let you know how it goes.

[LKO Railroad]

Started building out. LED pair mounting first order of business.



120 of them plus a few extras.



Mounting points for them are installed.

[SilverSolder]

Interesting - What's the purpose of the 5 stacked tracks?  - and where do the sensors you're making fit in the picture?

[floobydust]

LED's emit some light out their side, so I'm hoping you have something between the emitter and detector rows to prevent this. Like a small opaque piece of thin plastic. Or maybe do a quick test to see if LED on/off with no reflection makes too large a difference.
Are the track runs in a tunnel or something, where's the re-railers lol.

[gcewing]

Are you making a railway version of the Large Hadron Collider?