Pulse Generation Problem

[ecid]

hallo !
is there any circuit that can narrowed a pulse?
for example, if there's a 50% duty cycle, it means that, 50% high and 50% low, isn't it?
the problem is, i wanted to, for example, make a 10% high pulse and 90% low pulse. Is that possible? with what? A PWM? how

would be the circuit look like?

The application that i want to make is, a rpm counter that will be interfaced to a computer which display the rpm's speed.
SO, i wanted to change the single pulse's width that the optocoupler produced to ACK pin of parallel.

thanks...

 

[phalanx]

From a hardware standpoint, the simplest setup would be a basic microcontroller. All you would have is 3 wires. Power, Ground, and the PWM output. Timing is adjusted in software and is very easy to do.

 

[mramos1]

What range (speed) is this going to run? The parallel port is not super fast.

Maybe a micro like palanx said, and have the microcontroller send the count to the serial port on the PC.

 

[Hero999]

A 555 timer would do and it's far simpler than a complex microcontroller.

 

[phalanx]

A 555 timer would do and it's far simpler than a complex microcontroller.

A PIC10F or 12F can do everything he needs with no external components other than wires and will be far more versatile than a 555 timer. The software is simple enough to be a good project for a novice programmer.

 

[Hero999]

Depends on your experiance, it'd take me less than 5 miniutes to put together a 555 timer circuit on vero board, but it'd take me a lot longer to learn to program a PIC also I'd have to purchase a programmer and the componants will cost less too.

 

[stevez]

As part of a larger project I am putting together a pulse width modulator circuit. It will consist of a 555 timer to generate a ramp or sawtooth voltage that will run between 2 volts and 4 volts. The ramp goes to an LM339 comparator that is also fed with a reference voltage. The comparator output will change (either on or off) as the ramp voltage rises above or drops below the reference voltage. If you visualize a ramp or sawtooth voltage then draw a horizontal line that cuts thru it (representing the reference voltage) you can then visualize adjusting the ramp voltage up or down cutting thru longer or shorter periods of the ramp/sawtooth.

This is only one way - as others have suggested there are many other ways to do it. Note that a general purpose (low cost) op amp can be configured as a comparator.

 

[stevez]

I realize that I made an error here - "you can then visualize adjusting the ramp voltage up or down cutting thru longer or shorter periods of the ramp/sawtooth" - this should say " you can then visualize adjusting the reference voltage up or down cutting thrur longer or shorter periods of the ramp/sawtooth"

 

[neon]

an LM555 can do thet better then any other described scheme trust me on that. if you are doing modulation look at pin 5 as a souce of input also usea current source to linearize your ramp rc time is not linear. An LM555 can be one shot nonresatble ,astable and modulator. of course you could use a PC to do that and don't forget the kitchen sink. take care

 

[hyedenny]

How about using a couple inverter gates and a transistor? You can get pulse widths of just a few nanoseconds (not necessarily 10% duty) quite easily this way. If you want adjustability, the easiest way is probably with the 555.

 

[hyedenny]

There are 2 ways to measure RPM: counting the number of pulses during a certain time interval, or measuring the period of the pulse.... Which are you doing - frequency or period??

 

[ecid]

complete problem

in this reply, i added an attachment of the block diagram/schematic...
hope that'll helps

 

[Nigel Goodwin]

I don't really understand what you want at all?, why not just input the pulses directly? - the pulse width is pretty irrelevent. Assuming you do want to do it (perhaps you'd explain why?) you just need a monostable - 555's are commonly used in rev counters fed from pulses like this.

A simple google find loads of examples, you might find this one interesting? **broken link removed**.

 

[Dr.EM]

I have implemented the the triangle wave into a comparator idea recently, and have found it to work very well indeed. It has a full 0 - 100% adjustment, is very linear in operation and has the nice feature that its voltage controllable. With 555 PWM, I always find that as you alter it, the frequency drifts aswell, sometimes quite a lot. Using it as a variable duration monostable following the oscillator can work, but then it is poor with a varying frequency input.

 

[ecid]

problem design

Nigel Goodwin:
The pulse width is actually to speed up the respon of the holes of the disc, that's why i have to change the pulses from the optocoupler.
To change it to down pulses, an inverter could done it, isn't it?
By the way... thank's for the link that you provided, it's a good reference....

Dr.EM:
Impelementing triangle wave into a comparator? How?

Hydenny:
Between frequency and period? I'm doing period.

[/b]

 

[hyedenny]

I think using the period in a tach is usually reserved for low speed measurement. Yours looks more like a high speed signal...

An example of how an inverter can be used to shape your pulse is above, in my previous post.

If you want to use a discrete monostable multivibrator ("one-shot") circuit as shown below, use t=CRln2 to set the duration (t) of your pulse.

Its simpler (and way more stable) to use a 555, and simpler still to use a PIC.

 

[Nigel Goodwin]

Re: problem design

Nigel Goodwin:
The pulse width is actually to speed up the respon of the holes of the disc, that's why i have to change the pulses from the optocoupler.

Sorry, I still don't see the need? - how are you measuring the frequency?.

 

[ecid]

more...

hyedenny :
An output reading from flying wheel to measure rpm, is it a high speed signal?

In your previus post (pulse.jpg), how can i determine the value of the components (resistor) ? And what type of transistor would be used?
Also in your newly post (monostable.jpg) what kind of transistor(s) is used?
what's the difference between both configuration?

The parallel interface to the pc is only for displaying the speed in the pc's monitor. What is actually the PIC needed for?
Is the PIC also for controlling the motor movement ? (although for my project, it is not necessary) ?

If i use 555, how is the configuration? (I think it's provided in the 555's datasheet, isn't it?)


Nigel Goodwin :
The goal of my project?

Just like the block diagram in my previous post, i want to build a rpm counter that will be interfaced to pc using parallel connection. The rpm will be displayed at the pc's monitor.

The main purpose of my project is the implementation of interrupts from a rpm counter module to the pc.

What i needed now is a circuit design that can change the output pulse from the optocoupler (IR receiver) (just like in my previous post iwa2.jpg), since the original output pulse had 50% duty cycle.

And then the output'ed pulse will be directed to ACK pin of parallel connections.


Thank' s.

 

[Nigel Goodwin]

Re: more...

hyedenny :
The main purpose of my project is the implementation of interrupts from a rpm counter module to the pc.

What i needed now is a circuit design that can change the output pulse from the optocoupler (IR receiver) (just like in my previous post iwa2.jpg), since the original output pulse had 50% duty cycle.

You seem to be missing my point?, what difference does altering the pulse width make? - a 50/50 pulse is as easy to measure the frequency of as anything else. You appear to be wanting to add an extra stage that isn't needed?.

 

[ecid]

more...

Nigel Goodwin :

That's the problem.... it has been specified in my project that i must altered the pulse width to fullfill the interrups requirements, so the pulse will be 100us.