help regarding triggering comparator LM393

Thread Starter

devalvyas

Joined Nov 11, 2008
80
I am using the LM393 as a comparator to give monostable output for a particular time.

The comparator is working fine if i directly trigger it by connecting it to +Vcc.

The comparator also works fine if I trigger it by using a transistor.

But the moment i try to trigger it by a pulse trigger circuit made of R13, R14 and C7, as shown in the attached diagram, its not triggering.

Can some one please help.
 

Attachments

Audioguru

Joined Dec 20, 2007
11,248
When you turn on the switch (or push the pushbutton) then C7 gets backwards polarity but still might work for a while and pull up the base of the transistor. But the transistor is in series with the 10k resistor R12 which slows down the charging of C5. If R4 is set to 10k or less then the (-) input of the comparator will never be high enough for the comparator's output to change. Also, the time for C7 to charge might be faster than the time for C5 to charge so the comparator will not work.

Try it with 120 ohms for R12. Then the LED will light almost immediately and stay on for the R4/C5 delay.
 

Thread Starter

devalvyas

Joined Nov 11, 2008
80

Thanks AG, its working now.

However the problem is that the charging of C7 dictates the output of the comparator.

The output comparator changes when i close the switch, but it again changes back as soon as C7 charges up.

I have shown the new values in attached sheet.

I have chaged the polarity of c7. however there is no effect on the output.

MY MAIN INTENTION IS TO MAKE R4/C5 DETERMINE THE OUTPUT TIME AND NOT C7. HOW DO I DO IT?

I am trying to build a circuit that can detect when the motor shaft comes to a halt and want to use this circuit for the same.

The idea that I am toying around is this:

I have a small contactor attached to the shaft of the motor. As the motor rotates this contactor touches two pins A and B there by giving pulse to the comparator through a trigger circuit.

As soon as the connection is established the output of the comparator changes and remains at that state for some time say for 0.25 seconds, before the output can change back, the contactor touches the pins again and again triggers the comparator. This continues as long as the shaft is moving.

But the moment the shaft stops, the comparator no-longer gets any pulse for 0.25 seconds and it changes state, hence i get to know that the motor has stopped rotating.

The reason why I need the trigger circuit is that Even if the shaft comes to halt exactly at the point where it touches pin A and B, the comparator receives only one trigger through the trigger circuit and does not get second trigger for 0.25 seconds.

My expected speed of rotation will be around 1500 rpm. this means that the contactor will touch pins A and B 1500 time in a minute or 25 times a second.

The time between each pulse will be around 40 msec and the duration of the pulse will be for 4 msec only.

Will this circuit work? or is there any better simpler way to do it?


I have posted a new thread for the same...please help with this as well.
 

Attachments

Last edited:

Wendy

Joined Mar 24, 2008
23,415
Cool, learned a new part number. I was thinking you were incorrectly writing LM339, til I looked up the datasheet.

IMO mechanical is usually iffy, especially at high speed. I am aware you don't want to add a magnet to this, but I suspect there is a way to do it magnetically, or optically. If this were a shiny shaft, how practical would it be to take a magic marker to it (as in is it in a dark area)?

If there were a way to add a really small magnet then everything is simple, since the pulse would tell the tail.
 

Thread Starter

devalvyas

Joined Nov 11, 2008
80
I dont mind adding magnets...as a matter of fact i was trying out hall sensors that i bought and a 16 pole ring magnet that a company gave me as a sample..but more important than the contact part or pulse generating part, i am not getting the trigger circuit right.

I have basically 2 questions.

Que 1: (Fig no. 2)

What i want is a trigger circuit that

a. generates only one pulse irrespective of the time the switch is closed (that part you have already shown earlier and i have used it here)
b. This trigger circuit should not interfere with the RC timing of the comparator...this I am not able to get. How do i get this trigger ckt to work (Fig 2 of the attached file)

Que 2: (Fig no. 3)

i am also thinking about the 555 missing pulse detector..but how do i build a trigger circuit that generates only one pulse irrespective of the time the switch is closed?....can you suggest on how to use the trigger circuit for missing pulse detector...? In the attached file, fig no. 3 i want to replace switch S1 with a trigger circuit that gives small pulse irrespective of the time the switch is closed. The RC values in this trigger circuit should not interfere with Rt Ct of the 555 timer.
 

Attachments

Wendy

Joined Mar 24, 2008
23,415
What you are after is a retriggerable monostable. Where did you get those schematics?

From what I can tell you are trying to charge a cap, turn this around, discharge it instead, 100% for every pulse.

BTW, using word to show pictures is like using excel to do word processing. I strongly suggest you learn M/S Paint, Gimp, or any other of a large host of graphics packages. If you can do this using Word I think you'll find the others a snap. You can do things with graphics that you can't do this Word, such as this.







All of these schematics work by integrating the signal. If you totally discharge a cap using the collector of a transistor (resetting the timer) then I think you will get what you want.



This might work better with Schmitt Trigger in front, and it doesn't have to be a 555.
 

Attachments

Last edited:

Thread Starter

devalvyas

Joined Nov 11, 2008
80
fig 1 and 2 i got from this forum in my earlier threads and did some modification myself...but its not doing exactly what i want it to do though...fig 3 i got from net.

But i am not sure what you are trying to say....I want to replace the switch s1 of 555 in fig 3 with a trigger circuit...are you suggesting schmitt trigger for that? A schmitt trigger will basically clean up the pulse, which i can use after the contact points A AND B to get better output pulse train to the comparator or 555 missing pulse detector..

but once this pulse arrives at pin 2 of the 555 timer in fig 3 working as missing pulse detector, i want it to pass through a trigger circuit...and i want help in designing this trigger circuit which can be used with pnp transistor that is being used in 555...can you plz help with schematics of this trigger circuit that can replace switch s1?

OTHER WAY ROUND if i dont use a 555 but just comparator circuit, how do i design a trigger circuit for the comparator that does not interfere in the RC timing components of the comparator...
 
Last edited:

Wendy

Joined Mar 24, 2008
23,415
An integrated signal looks at an average of all your circuits, whereas a retriggerable monostable only cares about the last signal. All of your schematics uses the emitter to bring the cap up to the value on the base, and then slowly discharges the capacitor through an adjacent resistor. A retriggerable monostable references off the last pulse, period.
 
Top