100 posts, and still not nearly enough information about the sensors and/or signals. Back to basics . . .

Two sensors, T (top) and B (bottom). An object passes both. It is clear that the leading edge of the sensor output signal is the critical timing edge. I see three sensor signal options:

1. Each sensor makes a fixed-width pulse when triggered, and we use the leading edge to control the circuit outputs.

2. Each sensor makes an output state as long as it is blocked/touched/whatever. The output pulse width is inversely proportional to object velocity. This is typical of a "broken beam" type of sensor.

3. Each sensor's output is latched in its "detect" state by the sensor, and stays at that level until it is manually reset.

4. Other.

Each of these options has a different effect on the circuit, especially the part that indicates a "zone" event. Without getting into the actual sensor technology, object type, blah blah - which option best describes the two signals?

ak

 

This goes back to something I did for an Experimental Psychology lab in the early 1980's. I've got a demo circuit down to three chips plus one stupid little XOR gate that is bugging me.

Three indicators:

1. Red - forward tilt
2. Red - backward tilt
3. Green - near-zero tilt

Two latches (one chip), one for the tilt indications and one to show when the tilt is almost zero.

Two monostables (one chip), one to discriminate near-zero tilt and one to enable all outputs for 5 seconds. The output timer can be replaced with a manual Reset switch.

Three output enable gates (one chip). A manual Reset switch eliminates this chip.

One stupid little XOR gate (one chip). This turns two input signals into a single pulse whose width is the time between the inputs. The more narrow this pulse width, the closer the swing is to zero tilt.

WAIT - just had a thought about a different approach with fewer parts. More later.

ak

OK, and here, for A.K., is the alternative to the XOR gate: define that "sweet spot"as the time both sensors are ON, since they wil probably not be zero width outputs anyway. Then a simple AND gate gives the length of the overlap, which will include actual vertical. That might help solve the need for the XOR.

 

100 posts? Wow. I lost track but I see the number is now 101. Not sure how many of these were among yourselves but this has consumed too much time already. Remember the tennis analogy is just that - an analogy. I will not be dealing with that kind of speed. Before this number gets up to 200 I have to say enough and thank all contributors. I will look further into some of the existing suggestions. I thought I would be lucky to get a few responses over several weeks. You guys are passionate about what you do. I envy that. I have created a list of user IDs (but I'm sure I missed a few) so that if I do complete this I can re-post to all at a later date. It's too bad that on a forum like this or any platform allowing the anonymity of a User ID, a certain level of hostility eventually shows itself. Our local "Nextdoor" site is starting to see it as well. Can't help feeling like I triggered it but from some of the feedback you are offering perhaps I shouldn't. Not sure how many of you are engineers. When I was at university (actually started in electrical engineering then switched to math - good thing right?) I worked as a helper to electricians & mechanics at INCO in northern Canada - I was 18. While having coffee/lunch with Mature Man 1 I heard nothing but bitching about Mature Man2. Next day with MM2 I got the opposite about MM1! Will never forget it. The common (ok NEUTRAL) rant we all had was bitching about the engineers who designed the frickin' thing we had to install/fix/remove! Who would put a fastener back there? Where are the hooks to lift this 800 pound sucker? They were good lessons.

Keep up the passion and Thanks
Brewster

 

OK, and here, for A.K., is the alternative to the XOR gate: define that "sweet spot"as the time both sensors are ON,

Based on the information in this thread at the time of writing, there is nothing to support that definition. And yes, I did take that possibility into account when trying to reduce the possible sensor options to a digestible three. In different ways, it is a possibility in options 1 and 2.

ak

 

3 chips? Why? I can do it with one. I know you all hate my micro for everything solution, but it is certainly best solution for this problem. Everything that has been discussed here can be done with the same simple hardware and different programming. The TS based his product on what he could do with switches and relays, accepting a far from optimal solution. When you make the decision early to use a micro, the solution space expands immediately.

Bob

 

No hate expressed nor implied; it is a tool in the box. But generally speaking, a synchronous machine is uniquely *un*qualified to interface with multiple, high-speed, asynchronous inputs. I have run into a variation of this exact problem in each of the last six decades.

The two asynchronous inputs could be microseconds (or less) apart in time, I'm guessing too fast for a polled input. I suppose if you pick the right uC it could use an input pin to gate an internal counter directly, but that is semi-cheating. Even with a 20 MHz clock, could an interrupt routine really capture what is needed? One way would be to dumb-down the requirements, such as declaring that any time difference less than a millisecond is a tie. Probably good enough, but feels yucky.

ak

 

Input capture on PICSs will stop a timer on an external event with the resolution of the clock frequency., no interrupt required. And I doubt that you need 1sec resolution fir this project. Say the swing is 100 KPH. In 1 usec it moves 30 microns.

Edit: change nanometers to microns

Bob

 

Using any micro device takes both program creation skills and a fair amount of hardware that I do not presume everybody to have on hand.
Besides that, the TS asked for the very simple scheme, and there is nothing very simple about using a micro for the very first time. In addition, the other considerations already mentioned do apply.
So for those folks who are already skilled in using micros, and already have al of the required tools, and who are are already experienced at adapting assorted inputs, a micro could be an OK scheme.
For this particular thread, the TS has already commented about not having ANY of those prerequisites, and so in this instance using a micro is not an option.
Aside from all of that, it is the whole "nuts and Volts" "do everything with a micro and learn no electronics at all" mindset that I find so offensive.
Hate is way to strong a concept for anything here.

 

Input capture on PICSs will stop a timer on an external event with the resolution of the clock frequency., no interrupt required. And I doubt that you need 1sec resolution fir this project. Say the swing is 100 KPH. In 1 usec it moves 30 microns.

Edit: change nanometers to microns

Bob
.

Using any micro device takes both program creation skills and a fair amount of hardware that I do not presume everybody to have on hand.
Besides that, the TS asked for the very simple scheme, and there is nothing very simple about using a micro for the very first time. In addition, the other considerations already mentioned do apply.
So for those folks who are already skilled in using micros, and already have al of the required tools, and who are are already experienced at adapting assorted inputs, a micro could be an OK scheme.
For this particular thread, the TS has already commented about not having ANY of those prerequisites, and so in this instance using a micro is not an option.
Aside from all of that, it is the whole "nuts and Volts" "do everything with a micro and learn no electronics at all" mindset that I find so offensive.
Hate is way to strong a concept for anything here.

Hey MB2. Big change of topic from lamp circuit. But today I was looking at a course for the Arduino Uno which alleges to teach a beginner basic electronics and use of a microprocessor. Was never into electronics but was a programmer years ago. Thinking of robots like the old Robots Wars show. Is that a good one for a beginner to learn? It has a dozen or so projects to teach the basics of different devices you can attach and control.
Thanks
Brewster

 

Just another note of Thanks to all of those who participated in this lamp circuit stream the last several days. It was certainly enlightening for me. There was mention of a Game Show Buzzer circuit - I think it was Yaakov. After accepting the relay concept that I started with and the faults you identified I looked up just what Game Show Circuit meant (don't watch game shows). I think that is exactly it. There is really no speed involved here and the switches won't be push buttons but I think the rest of the circuit is just what the doctor ordered. Thanks again to all. Will post a result some day after my first $million! (and after I win the Masters when I am lucky to break 90). Eventually I also will score my age.
The Brewster