A counter that starts and stops with a negative going pulse

I am looking for suggestions to build a counter that will start with a negative going pulse and stop with the next negative going pulse. The interval is going to be between many seconds to many minutes. The interval is to be read out, arbitrary units fine, does not have to be seconds or minutes. It can then be manually reset to start anew.
Thank you

How many bits ? And clock rate (resolution, is it 1 sec resolution) ?

What is counter output used for, a display of the interval value ?

After counter completes does it stop until a trigger starts it again or does
it just run continuously ?

Generally speaking one does that with a reciprocal counter approach.

At the counting rate you suggest an 8 pin micro would do the job.

There are many ways. Inside the box click L/H and this does what you said


or the same with a pulse

Here is basic micro coded using block language. mBlock takes your block configuration
and converts to Arduino code so you can program the micro.




Is your start and stop pulse on the same signal line or two separate inputs ? Is pulse input from
mechanical switch ? If so we would add blocks to debounce the switch.

Rough circuit -

How many bits ? And clock rate (resolution, is it 1 sec resolution) ? 1 s resolution is enough. !2 bits will give me 1 h+, thats enough

What is counter output used for, a display of the interval value ? Yes

After counter completes does it stop until a trigger starts it again or does
it just run continuously ? The next low going pulse should start it.

Generally speaking one does that with a reciprocal counter approach. Seems like getting 1/x is always a pain but I am a neophyte.

Let me state the actual problem. Am trying to make a polymer layer on the walls in side a 20 micrometer id capillary. The polymerization is photoinitiated. So we pump pneumatically the polymerization mixture through a tube coil while blasting it with 10 W 365 nm light. The difference between forming any layer on the wall and the capillary getting totally blocked is very small! But because at constant pressure the flow rate is inversely proportional to the fourth power of the diameter, the flow rate is very sensitive to the id ofthe tube and hence the polymer layerr thickness. We have flow meters that can work in this extremely small flow regime: nL/min to uL/min but polymer deposition in that flow meter is equivalent to throwing a $2K flowmeter into the toilet. So we are simply letting the solution drop out of the capillary exit and counting the time between the drops. This I am next going to do w a LED and a photodiode, as the drop falls through that light path, it will generate a momentary pulse. So we count time between the pulses. Soon as that time drops perceptibly, we will stop the 365 nm irradiation light and wash out the capillary with alcohol, which dissolves the unreacted mixture but not the polymer already formed on the wall.
So it is basically a long interval drop counter and I want to sense an increase in the interval. I wish I could think of an analog solution (as my electronics mostly belongs to that age!) but does not seem easy. Perhaps the drop pulse can be used to pulse an LED coupled to a PD and I integrate the current over time which spans at least two drops. But the needed integration time constantis likely to be too long to be practical.

If you do not want to use a Microcontroller, you can proceed with flip-flops and a binary counter IC with a clock source like 555.

You can do long integrations times with todays OpAmps and caps.
But board leakage, contamination, humidity, dust enemies but not
impossible to conquer. But digital approach mitigates all of that.

1/x trivial in todays coding tools, mBlock no exception.

The ATTINY85 or the Nano board can do I2C to an OLED or character display.

Key issue with 555 is accuracy over T and V due to electrolytic drift. Eg. timing accuracy
poor.

Block languages come in many flavors, here is example to do talking volt/freq/pulse
idth....meter. And this language has blocks for I2C and display

Text block diagram -

For the system clock, a 32.768 kHz watch crystal oscillator and 14 stage divider in a CD4060 CMOS IC. This gives you a 2 Hz timebase for twice the required resolution.

The output of the optical drop detector clocks a toggle flipflop. The flipflop output enables / resets the clock oscillator circuit. One drop and the system starts counting in 0.5 second ticks, and the next drop stops the counting. Note that this does not reset the main counter; that is done manually.

The main counter is a 4-digit BCD up-counter and display with a manual switch driving the Reset input.

ak

Note: Whether or not the clock divider is reset between operating cycles has a direct bearing of one aspect of the display accuracy. There is not enough information in the question to determine how to handle this.

PurnenduD said:

I am looking for suggestions to build a counter that will start with a negative going pulse and stop with the next negative going pulse. The interval is going to be between many seconds to many minutes. The interval is to be read out, arbitrary units fine, does not have to be seconds or minutes. It can then be manually reset to start anew.
Thank you

Click to expand...

Instead of limiting a designer to a "counter", it would be better to simply describe what you want. It sounds like you want a system that
(1) starts when a trigger is pressed (or otherwise initiated some time after a trigger press or key entry)
(2) it determines the initial flow by determining the time between drips (t seconds) with a photosensor (where t is assumed to be greater than seconds but less than 60 seconds)
(3) if the time between drips increases by 1-second (t+1 seconds), the system is stopped (either by stopping the monomer pump and/or turning off the UV light and/or starting the flow of alcohol instead of monomer through the capillary.

Is that about right?

How large is this chamber?
Can wires/cables run in/out to control the system?
What is your budget ( can you afford a $70 raspberry Pi computer that can run a Python script and almost no hardware and infinitely adjustable in code in a minutes or do you want $10 of logic chips soldered to a $5 PCB that is less adjustable/adaptable?).

With a Raspberry pi solution, you can also monitor variables (time) in real time in a standard PC monitor.

Such devices are available cheaply off-the-shelf; sometimes less than the cost of building one;

eg.