Driving 555 IC with a pulse stream

[Jag Man]

I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the decade
counter.

My guess is I have to use the decade counter output to both reset the 555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the reset
want,
high or ground?

TIA

Ed

 

[John Fields]

On Sun, 14 Nov 2004 23:55:49 GMT, "Jag Man"
<Jag_Man653R-E-MOVE@hotmail.com> wrote:

I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the decade
counter.

My guess is I have to use the decade counter output to both reset the 555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the reset
want,
high or ground?

---
The reset wants to see ground, and so does the trigger, but the width
of the trigger pulses needs to be shorter than the output pulses from
the 555, so you'll probably need to differentiate the counter outputs
before you send them to the 555.

--
John Fields

 

[Bill Bowden]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote in message news:<9gSld.8844$zx1.6054@newssvr13.news.prodigy.com>...

I have a decade counter (NTE4017B) that generates a pulse
stream on each decoded output, a couple of which I will use.
However, the width of these pulses are too wide at lower
frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what I want,
regardless of the frequency and pulse width coming from
the decade counter.

My guess is I have to use the decade counter output to
both reset the 555 (pin 4) and trigger it (pin 2). Is that
right? I know the triggering is by dropping pin 2 to ground,
perhaps by switching with a transistor, but what does the reset
want, high or ground?

TIA

Ed

The reset pin is active high but it doesn't need to
be used. Just connect it to ground.

The 4017 pulse is probably positive going, and the
555 needs a negative going edge, so you will need
a NPN transistor to invert the pulse and a cap
to trigger the 555 on the leading edge. Something
like the drawing, but you may also need a diode from
the transistor base to GND if the input is more than
6 volts.

+V
|
\ 555
/ +------+
5.1K \ | |
| | 3|--> Out
+--------|2 |
C | |
0.01uF |/ +------+
In --/\/\/---| |--+--B| NPN
5.1K C | |\
\ E
5.1K / |
\ |
| |
GND GND

-Bill

 

[Terry Pinnell]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote:

I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the decade
counter.

My guess is I have to use the decade counter output to both reset the 555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the reset
want,
high or ground?

TIA

Ed

Other respondents seem to have no trouble with your spec, so the odds
are that I've missed something - but I'm darned if I can make sense of
your description. Perhaps if you could describe your end objective?

Meanwhile, as I understand it, you have a 555 astable clocking a 4017,
yes? You want the 4017's outputs to have a fixed duration? Then what's
the problem with just fixing the 555 astable period accordingly, which
then fixes each 4017 high output to the same duration?

In which case, where does the need for a 555 (or any other) monostable
arise?

You don't mention any specific values, but FWIW here's a 5V circuit
with a 555 clocking a 4017 so that each of its outputs flashes an LED
for about 4 ms.

http://www.terrypin.dial.pipex.com/Images/555+4017.PDF

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[John Fields]

On 14 Nov 2004 23:08:09 -0800, wrongaddress@att.net (Bill Bowden)
wrote:

The reset pin is active high but it doesn't need to
be used. Just connect it to ground.

--- _____
It's active low; RESET

--
John Fields

 

[Jag Man]

John,

I now have the input to the 555 going to BOTH TRIGGER
and RESET, without differentiation (i.e., no capacitor in series),
and it seems to work. I can definitely get pulses out of the 555
that are shorter than the inputs. As I see it, I am resetting and triggering
at the same time. Is that antithetical?

Ed

The reset wants to see ground, and so does the trigger, but the width
of the trigger pulses needs to be shorter than the output pulses from
the 555, so you'll probably need to differentiate the counter outputs
before you send them to the 555.

--
John Fields

 

[Jag Man]

Thanks for coming into the discussion, Terry.

The project was first introduced about a month ago, 10/10/04,
as "Special dual pulse generator circuit." I'm implementing a
circuit basically as proposed by Chris Foley. Briefly,
I'm trying to achieve a tester for the EFI system on a Jaguar
V12. the car has a trigger board in the distributor that generates
two switch closures per revolution, 180 degrees apart, that drive
the electronic control unit (ECU). I intend to plug the tester
into the harness in place of the trigger board, thus allowing me
to drive the ECU, amplifier, and fuel injector solenoids
without the engine running. When the engine is running there are variations
in speed etc.
that make interpretation of scope traces difficult. My hope is this will
allow me to
test the various components under more controlled conditions.

So, with Chris' great design the a-stable 555 runs at 10 times the
rotor speed, so to speak, and the decade counter outputs are
ad one 10th that frequency, thus emulating the spinning distributor.
By picking up two the the decade counter outputs, e.g. 1 and 6,
I get a very good representation of what actually happens. Importantly,
the speed can easily be varied while always having the two outputs
exactly 180 degrees apart.

Chris' design meets my original description of the problem, but I then got
ambitious. I noticed that by putting another 555 IC (actually 2, one for
each
output) between the decade counter and the output stage (optoisolators)
I could have control over both speed and pulse width, allowing the
same tester to be used to emulate the trigger board and drive the ECU
OR to emulate the trigger board and ECU.

Meanwhile, the big cat lumbers in the garage!

Ed

Other respondents seem to have no trouble with your spec, so the odds
are that I've missed something - but I'm darned if I can make sense of
your description. Perhaps if you could describe your end objective?

Meanwhile, as I understand it, you have a 555 astable clocking a 4017,
yes? You want the 4017's outputs to have a fixed duration? Then what's
the problem with just fixing the 555 astable period accordingly, which
then fixes each 4017 high output to the same duration?

In which case, where does the need for a 555 (or any other) monostable
arise?

You don't mention any specific values, but FWIW here's a 5V circuit
with a 555 clocking a 4017 so that each of its outputs flashes an LED
for about 4 ms.

http://www.terrypin.dial.pipex.com/Images/555+4017.PDF

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[John Fields]

On Mon, 15 Nov 2004 22:14:04 GMT, "Jag Man"
<Jag_Man653R-E-MOVE@hotmail.com> wrote:

John,

I now have the input to the 555 going to BOTH TRIGGER
and RESET, without differentiation (i.e., no capacitor in series),
and it seems to work. I can definitely get pulses out of the 555
that are shorter than the inputs. As I see it, I am resetting and triggering
at the same time. Is that antithetical?

---
No, it's actually pretty slick!

What's happening is that when the 4017 output goes low it
unconditionally resets the 555, discharging the timing cap, then when
it goes high and it gets to about 0.7V it releases the reset and drags
the trigger pin high along with it. Now, since the trigger pin will be
at less than about 1/3 Vcc when the 555 comes out of reset, the output
will go high and the 555 will start timing out. What's slick is that
it's only going to take ever how long it takes for the counter's
output to get to higher than 1/3 Vcc to not be triggering the 555 any
more, and then the 555 will time out during the time the counter's
output is high.

__________ __________ ______
4017 OUT __________| |__________| |__________|
_____ __________ __________ ______
RESET __________| |__________| |__________|
____ __________ __________ ______
TRIG __________| |__________| |__________|
_ _ _
555 OUT ___________| |___________________| |___________________| |___

-->||<--- effective trigger pulse width


--
John Fields

 

[John Fields]

On Tue, 16 Nov 2004 16:42:37 GMT, "Jag Man"
<Jag_Man653R-E-MOVE@hotmail.com> wrote:

Thanks, John. I may have misled you though, as I actually have an NPN (NTE
123AP) transistor
between the 4017 and the 555. E.g., pin 2 (a decoded output) of the 4107
goes a 10K resistor, thence
to the base of the transistor. Another 10k resistor connects the base to
ground. The collector
goes to pin 2 of the 555, and the emitter is grounded. Pin 2 is also
connected through a 10k
resistor to +12. This is to present pin 2 with a negative pulse, which is
the way
I've triggered 555 in other projects.

So, does the analysis still hold?

---

Yeah, but now your timing will look like this:

__________ __________
4017 OUT _______________| |____||___| |______
||
_____ _______________ ____||___ ______
RESET |__________| || |__________|
____ _______________ ____||___ _____
TRIG |__________| || |__________|
_ || _
555 OUT __________________________| |__||______________| |___
||

That is, the 555 will trigger on the trailing edge of the 4017 output
instead of on the leading edge.

--
John Fields

 

[Glenn Gundlach]

"Brian" <bellis350@comcast.net> wrote in message news:<w4OdnSkADt3YXAXcRVn-uA@comcast.com>...

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote in message
news:9gSld.8844$zx1.6054@newssvr13.news.prodigy.com...
I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses
are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the
decade
counter.

My guess is I have to use the decade counter output to both reset the
555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by
dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the
reset
want,
high or ground?

TIA

Ed



Why not use a one-shot, like a 74123. Then you can use either a positive
or negative going edge triggering.
Brian

Because that would make sense. You're confusing logic with your
thinking. (a line by Gene Burrows from ABC TV Hollywood.)
gg

 

[Jag Man]

Thanks for your interest, Terry.

I just uploaded a scan of my circuit (thanks to Chris Foley) to:

http://sowell.ecs.fullerton.edu/jag/fuelingsys/tester111704.jpg

The diagram shows a 400/500 ohm resistor and an LED in the 2N3094 collector
circuit, in the path from pin 2 of the 4N32. At the moment the resistor is
a 5% 400 ohm, measuring about 380 ohms, and the LED has been removed.

BTW, what version of CircuitMaker are you using, student of full? The full
version is kind of pricy for someone not making their living in this stuff.


Ed

OK, thanks, understood. A more complex spec than I'd thought!

From your separate thread 'Special "dual" pulse generator circuit
(follow up)', I see you're sorted, apart from one residual issue.

FWIW, simulating an opto-isolator fed with a 0-12V fast rise time
pulse via a 10k base resistor gave me much faster output rise times
than those you observed. With both models in its library, a 'generic'
and the OP4N25, CircuitMaker gave the following results (which square
with Chris's point about lowering the resistor value).

Rc Rise time
---- ---------
10k 300 us
1k 50 us

So I'd suspect some other cause. Is your circuit, or at least that
section, posted anywhere?

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Jag Man]

Robert,

Would a small capacitor in the input to the 4017 serfe as an "edge
detecter?"


TIA

Ed

"Robert Monsen" <rcsurname@comcast.net> wrote in message
news:1q6md.43208$5K2.39526@attbi_s03...

I think what he wants is to shorten the pulse coming out of the 4017.
Thus, I believe he is asking how to set up a 555 as a one-shot, given a
long pulse.

Now that I think about it, he probably wants an edge detector to drive
the 555 trigger input.

--
Regards,
Robert Monsen

 

[Robert Monsen]

Jag Man wrote:

Robert,

Would a small capacitor in the input to the 4017 serfe as an "edge
detecter?"


TIA

Ed

"Robert Monsen" <rcsurname@comcast.net> wrote in message
news:1q6md.43208$5K2.39526@attbi_s03...

I think what he wants is to shorten the pulse coming out of the 4017.
Thus, I believe he is asking how to set up a 555 as a one-shot, given a
long pulse.

Now that I think about it, he probably wants an edge detector to drive
the 555 trigger input.

--
Regards,
Robert Monsen

Jag:

If you are worried about the length of the pulse coming out of the 4017,
you can use a capacitor and resistor to detect either a rising or
falling edge of the output.

To detect the rising edge, connect an output to the input of the next
stage through the cap, and use a resistor to ground on the far side of
the cap, as so:

.---------.
| |
| | C
| |
| | ||
| 4017 Qx|-----||--o---- Out
| | || |
| | |
| | |
| | \
| | / R
| | \
'---------' /
|
GND
created by Andy´s ASCII-Circuit v1.25.250804 www.tech-chat.de

Then, when without a transistion for a while, the output will be
grounded. When the Qx output goes from GND to 5V, however, the output
will momentarily go up to 5V, and then will decay back to ground
depending on the values of R, C, and the impedance of the output. The
voltage at Out will fall to about 37% of the input pulse in t = R*C.

When the 4017 output goes from 5V to 0V, the output will drop below GND.
However, you can add a diode from GND to Out to prevent this.

For a falling edge detector, you can use this (with the diode drawn in)


VCC o--o
| |
.---------. \ |
| | R / - Diode
| | \ ^
| | / |
| | || | |
| 4017 Qx|-----||--o--o- Out
| | ||
| | C
| |
| |
| | Pulse length is ~RC
| |
'---------'
created by Andy´s ASCII-Circuit v1.25.250804 www.tech-chat.de

For your problem, I'd say you can use the second circuit between your
4017, and the trigger input of a 555. It will generate a nice negative
pulse. Then, you can set the output pulse width using the forumlas given
in the 555 datasheet for the monostable configuration.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

 

[Robert Monsen]

Terry Pinnell wrote:

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote:


I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the decade
counter.

My guess is I have to use the decade counter output to both reset the 555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the reset
want,
high or ground?

TIA

Ed


Other respondents seem to have no trouble with your spec, so the odds
are that I've missed something - but I'm darned if I can make sense of
your description. Perhaps if you could describe your end objective?

Meanwhile, as I understand it, you have a 555 astable clocking a 4017,
yes? You want the 4017's outputs to have a fixed duration? Then what's
the problem with just fixing the 555 astable period accordingly, which
then fixes each 4017 high output to the same duration?

In which case, where does the need for a 555 (or any other) monostable
arise?

You don't mention any specific values, but FWIW here's a 5V circuit
with a 555 clocking a 4017 so that each of its outputs flashes an LED
for about 4 ms.

http://www.terrypin.dial.pipex.com/Images/555+4017.PDF

Hey, Terry.

I think what he wants is to shorten the pulse coming out of the 4017.
Thus, I believe he is asking how to set up a 555 as a one-shot, given a
long pulse.

Now that I think about it, he probably wants an edge detector to drive
the 555 trigger input.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

 

[Terry Pinnell]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote:

Thanks for coming into the discussion, Terry.

<snip full description>

Chris' design meets my original description of the problem, but I then got
ambitious. I noticed that by putting another 555 IC (actually 2, one for
each
output) between the decade counter and the output stage (optoisolators)
I could have control over both speed and pulse width, allowing the
same tester to be used to emulate the trigger board and drive the ECU
OR to emulate the trigger board and ECU.

Meanwhile, the big cat lumbers in the garage!

Ed

OK, thanks, understood. A more complex spec than I'd thought!

From your separate thread 'Special "dual" pulse generator circuit
(follow up)', I see you're sorted, apart from one residual issue.

FWIW, simulating an opto-isolator fed with a 0-12V fast rise time
pulse via a 10k base resistor gave me much faster output rise times
than those you observed. With both models in its library, a 'generic'
and the OP4N25, CircuitMaker gave the following results (which square
with Chris's point about lowering the resistor value).

Rc Rise time
---- ---------
10k 300 us
1k 50 us

So I'd suspect some other cause. Is your circuit, or at least that
section, posted anywhere?

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Jag Man]

Thanks, Bill.

Actually, what I have been trying is close to what you suggest.
I'm using a 2N3094 and 10k resistors instead of 5.1k,
but no capacitor. In that configuration, with the RESET at 12 v,
I can't seem to get the output pulse any shorter than the incoming pulse.
However, I've now learned that by connecting the RESET pin
to the triggering pulse (i.e., same as pin 2) it seems to work.

Not sure what the capicitor would do.

Ed

The reset pin is active high but it doesn't need to
be used. Just connect it to ground.

The 4017 pulse is probably positive going, and the
555 needs a negative going edge, so you will need
a NPN transistor to invert the pulse and a cap
to trigger the 555 on the leading edge. Something
like the drawing, but you may also need a diode from
the transistor base to GND if the input is more than
6 volts.

+V
|
\ 555
/ +------+
5.1K \ | |
| | 3|--> Out
+--------|2 |
C | |
0.01uF |/ +------+
In --/\/\/---| |--+--B| NPN
5.1K C | |\
\ E
5.1K / |
\ |
| |
GND GND

-Bill

 

[Brian]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote in message
news:9gSld.8844$zx1.6054@newssvr13.news.prodigy.com...

I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses
are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the
decade
counter.

My guess is I have to use the decade counter output to both reset the
555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by
dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the
reset
want,
high or ground?

TIA

Ed

Why not use a one-shot, like a 74123. Then you can use either a positive
or negative going edge triggering.
Brian

 

[Bill Bowden]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote in message news:<kU4md.21509$6q2.21047@newssvr14.news.prodigy.com>...

Thanks, Bill.

Actually, what I have been trying is close to what you suggest.
I'm using a 2N3094 and 10k resistors instead of 5.1k,
but no capacitor. In that configuration, with the RESET at 12 v,
I can't seem to get the output pulse any shorter than the incoming pulse.
However, I've now learned that by connecting the RESET pin
to the triggering pulse (i.e., same as pin 2) it seems to work.

Not sure what the capicitor would do.

Ed

No, you can't get it any shorter because the
transistor is holding the trigger (pin 2)
at a low level during the entire input time,
and the 555 output will not return low until the
trigger line goes back up.

If you use the capacitor, the transistor will
only turn on during the time it takes to charge
the capacitor, so you get a very short pulse at
pin 2 and the 555 output will end earlier determined
by the R and C parts at pin 6.

Try the capacitor to see what happens. You can
try different values of .1uF, .047uF, 0.01uF, etc.

There is also a mistake in the other post as John
Fields pointed out. The reset line (pin 4) will
reset the 555 output to a low state when it is
connected to ground. So it normally connects to
+V for the 555 to operate.

-Bill

The reset pin is active low but it doesn't need to
be used. Just connect it to +V. (corrected version)

The 4017 pulse is probably positive going, and the
555 needs a negative going edge, so you will need
a NPN transistor to invert the pulse and a cap
to trigger the 555 on the leading edge. Something
like the drawing, but you may also need a diode from
the transistor base to GND if the input is more than
6 volts.

+V
|
\ 555
/ +------+
5.1K \ | |
| | 3|--> Out
+--------|2 |
C | |
0.01uF |/ +------+
In --/\/\/---| |--+--B| NPN
5.1K C | |\
\ E
5.1K / |
\ |
| |
GND GND

-Bill

 

[Bill Bowden]

"Jag Man" <Jag_Man653R-E-MOVE@hotmail.com> wrote in message news:<6bQpd.27513$6q2.22864@newssvr14.news.prodigy.com>...

Thanks again, Bill. Just out of curiosity, though, why does it work
when I simply connect pin 4 to the same voltage seen by pin 2? That
is, connected this way the output of the 555 can indeed be set shorter
than the pulse width at 2/4. Curious.

Ed

As John Fields explained, it works because as the input at
pin 2 and 4 moves up, the reset ends while the trigger is
still active, so you get an output pulse starting at the
same time as the input. Problem is, when the input moves
from +V to ground, pin 2 may trigger another pulse before
pin 4 resets the output. So you may get a very short pulse
out at the end of the input pulse. But apparently, that's not
happening so it looks like you solved the problem.

-Bill

 

[Robert Monsen]

Jag Man wrote:

I have a decade counter (NTE4017B) that generates a pulse stream on each
decoded
output, a couple of which I will use. However, the width of these pulses are
too wide at lower frequencies. So, I'd like to use a 555 IC operating in
monostable mode to set the pulse width to what
I want, regardless of the frequency and pulse width coming from the decade
counter.

My guess is I have to use the decade counter output to both reset the 555
(pin 4)
and trigger it (pin 2). Is that right? I know the triggering is by dropping
pin 2
to ground, perhaps by switching with a transistor, but what does the reset
want,
high or ground?

You can just use it in monostable mode. Tie reset to high. Look at the
datasheet (google 555 datasheet) for more information.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.