-5
\$\begingroup\$

I'm looking for a way to trigger an action when approximately N pulses have been received, where N is a large number - in the range of tens of thousands. I don't want to use counters, so I'm thinking of a solution on the lines of "this operation runs in N clock pulses and returns 1". Is there a way to achieve this without using counters or programmable chips?

\$\endgroup\$
11
  • 1
    \$\begingroup\$ Can you clarify what you mean by "operation" - on what? A computer- that would seem to violate your requirement to not use programmable chips (or counters since there is a program counter in the computer). The simplest digital way is with a counter or tiny microcontroller. There might be analog ways, but it would depend on limiting the range of total time over which the pulses might or might not be received. \$\endgroup\$ Apr 13, 2014 at 15:43
  • 1
    \$\begingroup\$ Whenever someone posts a problem and states that they don't want to use what to most observers would be the obvious solution, it's a really, really good idea to explain why not. If we guess, we'll probably guess wrong. \$\endgroup\$
    – lyndon
    Apr 13, 2014 at 15:47
  • 4
    \$\begingroup\$ How fast do these pulses come? What is the minimum high and low time per pulse of the incoming logic signal? Also, without proper justification the spec of no counters is just some silly religious reason. We do engineering here, where religious hangups have no place. \$\endgroup\$ Apr 13, 2014 at 15:49
  • 1
    \$\begingroup\$ I'm looking to winning the lottery without buying a ticket. What are my chances? \$\endgroup\$
    – Andy aka
    Apr 13, 2014 at 16:29
  • 3
    \$\begingroup\$ The OP was back 1 hour ago, but hasn't provided any clarification. Time to close this mess. \$\endgroup\$ Apr 13, 2014 at 22:15

3 Answers 3

5
\$\begingroup\$

I'd consider using a strowger stepping switch like this one: -

Courtesy of wiki

Courtesy of wiki. Or maybe this one: -

enter image description here

Each pulse advances the armature (rotary arm) by one increment and a series of these would be able to eventually short two contacts together that matched the number N.

Is this a counter in the strictest sense? Maybe it's an alarm rather than a counter as there is no observable output other than when the requisite number of impulses have been reached.

Counters, in the electronic sense keep a tally of where they are. Maybe this can be regarded as doing so too?

Maybe you could modernize this a bit by using stepper motors turning a threaded circular bar so that when the thread revolved around so many times (equivalent to \$\dfrac{N}{pulses\space per\space rev}\$) a contact closed.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ +1 for valid answer that's most likely to contradict the specification that questioner hasn't provided \$\endgroup\$
    – pjc50
    Apr 13, 2014 at 17:40
  • \$\begingroup\$ @pjc50 Oooh I do hope so LOL \$\endgroup\$
    – Andy aka
    Apr 13, 2014 at 18:14
  • 1
    \$\begingroup\$ Nice solution, but which museum do you have to steal one of these from? \$\endgroup\$ Apr 14, 2014 at 14:41
  • \$\begingroup\$ @olin I have a secret stash of them waiting to be sold to the highest bidder, no doubt the OP. \$\endgroup\$
    – Andy aka
    Apr 14, 2014 at 16:51
3
\$\begingroup\$

You could flick or release an individual ball bearing into a container (perhaps using a solenoid or solenoid valve air-operated device) each time a pulse comes in. Ball bearings are manufactured to very tight tolerances, so one is much like another in weight.

enter image description here

A vibratory bowl feeder could deliver the ball bearings to the top of the container.

enter image description here

Either use a pressure switch to activate an output pulse and empty the container into the bowl feeder when the desired count is reached, or use a tipping bucket gauge (similar to the rain gauge shown below) to empty the ball bearings back into the bowl feeder.

enter image description here

Somewhat inspired by this piece of art

\$\endgroup\$
2
  • 2
    \$\begingroup\$ +1 for the amuzing drawing. Particularly the weather station cable wrapped around the funnel and the tape holding it in place. \$\endgroup\$ Apr 14, 2014 at 14:46
  • \$\begingroup\$ @OlinLathrop Gaffer tape, for sure. \$\endgroup\$ Apr 14, 2014 at 14:47
1
\$\begingroup\$

Within some limits this could be done by a charge pump, accumulating small amounts of charge in a large buffer capacitor. A level detector would detect when N is reached.

Problems:

  • accuracy (especially over time, temperature, Vcc variations, ...)
  • leakage (When N can be sperad over a long period of time)
\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.