One shots, timers, etc

Guest
I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am a somewhat familiar with a number of one-shots and timer such as
the hc123 and hc4538 for short pulses, and timing, but I am in need of
something that can give me a timeout pulse for up to five minutes. When
I look at the data sheets for these devices, I am not sure where I need
to check to learn if the device is stable enough for a pulse that long,
and if a simple R/C network can accomplish what I am hoping to provide.


The question will be, how accurate? and I can only say that it must be
as accurate as I can make it, and still make it affordable, so that is
why I am wondering where on the datasheet I can find such information
so I can understand my options.

Are these kinds of chips going to be the right approach, or is there
another way I should be accomplishing this long of a timeout with some
degree of accuracy?

I will need to select five timeouts, from one minute, through five
minutes. I thought about using a dipswitch to choose between five
different RC combinations. An idea for something more elegant would
certainly be appreciated.
Thanks for the space and the consideration.

John
 
Hello John,

For such long intervals I would use a CD4060. That chip contains an
oscillator and a 14-stage divider. You can even crystal-control it for
really accurate timing. This will be much more reliable than messing
around with microamps and some one-shot chip.

Regards, Joerg

http://www.analogconsultants.com
 
IIRR the classic bipolar 555 timer is not a good choice for long time
intervals - the bias current on the input that senses the voltage
across the capacitor is way too high.

There are CMOS versions of the 555 around which probably can be
persuaded to give a 5 minute delay, but the worst case leakage current
through the protection diodes may be too high to allow you to guarantee
this.

A digital counter based approach is almost always cheaper, more
reliable, and more compact. The Xilinx (formerly Philips) Coolrunner
CMOS PLDs are probably the simplest single active device you can use to
do the job, but a minimal single chip microprocessor is probably going
to be cheaper.

-------------
Bill Sloman, Nijmegen
 
vanagonvw@gmail.com wrote:
I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am a somewhat familiar with a number of one-shots and timer such as
the hc123 and hc4538 for short pulses, and timing, but I am in need of
something that can give me a timeout pulse for up to five minutes. When
I look at the data sheets for these devices, I am not sure where I need
to check to learn if the device is stable enough for a pulse that long,
and if a simple R/C network can accomplish what I am hoping to provide.


The question will be, how accurate? and I can only say that it must be
as accurate as I can make it, and still make it affordable, so that is
why I am wondering where on the datasheet I can find such information
so I can understand my options.

Are these kinds of chips going to be the right approach, or is there
another way I should be accomplishing this long of a timeout with some
degree of accuracy?

I will need to select five timeouts, from one minute, through five
minutes. I thought about using a dipswitch to choose between five
different RC combinations. An idea for something more elegant would
certainly be appreciated.
Thanks for the space and the consideration.

John
Click to expand...
The simplest would be a START/STOP 4060 clocking a 4017 and then using
the DIP switch to select outputs 1 through 5 of the 4017. This implies
that the 4060 produces the 4017 input clock at 60 second intervals. The
60 seconds is 1/2 the period of 4060 2^14 divide register making it 120
second=Tosc x 2^14 or 4060 oscillation frequency=7.3msec which is quite
high. A cascade of two 4060's would make this 1.8usec which can be made
very accurate- something like RC*2.2 nominally or a 220pF NPO and 5K
adjustable pot set for 3.7K or so for the timing. The tricky part is
getting the 4017 to clock off the positive and negative transitions of
the 4060 output clock- and this is easily done with an EXOR. Then your
control is simply to hold all the IC's in RESET, lift RESET to initiate
timing, and terminate when selected 4017 goes high. Something like this,
press START to initiate timing- use the MR signal as an output- and
spare NAND if you want output HIGH during timing interval:

View in a fixed-width font such as Courier.

..
..
..
.. +-----------------------------------------+--------------------+
.. +-| \ 10K _ | |
.. | | O--/\/\-+----| \ | |
.. +-|__/ | | O----+ / |
.. | +|__/ | 10K |
.. | | | / |
.. _ | +-------+ | \ |
.. GND---o o------+ | | | |
.. RESET +-------|-+---+ GND |
.. | __ | | SELECTED -----> |
.. +| \ | | OUTPUT |
.. START | O--+ | MINUTES |
.. GND--o o-+----------|__/ | (DIP SW ) |
.. | | |
.. VCC-/\/\-+ | |
.. 10K +----------------+-------------------+ |
.. | | | |
.. NPO +------4060 +------4060 +------4017 |
.. 220p | MR | | MR | | MR | |
.. +--||--+CT | |CT | | | |
.. | | Q12|--+ | Q14|--+ +--+CP Q1| |
.. |5K pot| | | | | | | | Q2| |
.. +-/\/\-+RT | | |RT | | | | Q3|>--+
.. | | | | | | | | | Q4|
.. | 10K | | | | | | | | __ Q5|
.. +-/\/\-|RS | +---|RS(phi1) | | | | CE |
.. +---------+ +---------+ | | +---------+
.. | | |
.. | | GND
.. | |
.. +-------------------+ |
.. | |
.. | |
.. | |
.. | .--------. |
.. | |74HCT132| |
.. | .--------. |
.. +----------| \ |
.. | | O-+ |
.. | +-|__/ | |
.. 10K | __ | | |
.. +--/\/\---+-| \ | | __ |
.. | | O--+ +-| \ |
.. +---------+-|__/ | | O--+
.. | | | +-|__/
.. ===470p | | __ |
.. | | +-| \ |
.. GND | | O-+
.. +----------|__/
..
..
..
 
Fred Bloggs wrote:
vanagonvw@gmail.com wrote:

I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am a somewhat familiar with a number of one-shots and timer such as
the hc123 and hc4538 for short pulses, and timing, but I am in need of
something that can give me a timeout pulse for up to five minutes. When
I look at the data sheets for these devices, I am not sure where I need
to check to learn if the device is stable enough for a pulse that long,
and if a simple R/C network can accomplish what I am hoping to provide.


The question will be, how accurate? and I can only say that it must be
as accurate as I can make it, and still make it affordable, so that is
why I am wondering where on the datasheet I can find such information
so I can understand my options.

Are these kinds of chips going to be the right approach, or is there
another way I should be accomplishing this long of a timeout with some
degree of accuracy?

I will need to select five timeouts, from one minute, through five
minutes. I thought about using a dipswitch to choose between five
different RC combinations. An idea for something more elegant would
certainly be appreciated.
Thanks for the space and the consideration.

John


The simplest would be a START/STOP 4060 clocking a 4017 and then using
the DIP switch to select outputs 1 through 5 of the 4017. This implies
that the 4060 produces the 4017 input clock at 60 second intervals. The
60 seconds is 1/2 the period of 4060 2^14 divide register making it 120
second=Tosc x 2^14 or 4060 oscillation frequency=7.3msec which is quite
high. A cascade of two 4060's would make this 1.8usec which can be made
very accurate- something like RC*2.2 nominally or a 220pF NPO and 5K
adjustable pot set for 3.7K or so for the timing. The tricky part is
getting the 4017 to clock off the positive and negative transitions of
the 4060 output clock- and this is easily done with an EXOR. Then your
control is simply to hold all the IC's in RESET, lift RESET to initiate
timing, and terminate when selected 4017 goes high. Something like this,
press START to initiate timing- use the MR signal as an output- and
spare NAND if you want output HIGH during timing interval:

View in a fixed-width font such as Courier.

.
.
.
. +-----------------------------------------+--------------------+
. +-| \ 10K _ | |
. | | O--/\/\-+----| \ | |
. +-|__/ | | O----+ / |
. | +|__/ | 10K |
. | | | / |
. _ | +-------+ | \ |
. GND---o o------+ | | | |
. RESET +-------|-+---+ GND |
. | __ | | SELECTED -----> |
. +| \ | | OUTPUT |
. START | O--+ | MINUTES |
. GND--o o-+----------|__/ | (DIP SW ) |
. | | |
. VCC-/\/\-+ | |
. 10K +----------------+-------------------+ |
. | | | |
. NPO +------4060 +------4060 +------4017 |
. 220p | MR | | MR | | MR | |
. +--||--+CT | |CT | | | |
. | | Q12|--+ | Q14|--+ +--+CP Q1| |
. |5K pot| | | | | | | | Q2| |
. +-/\/\-+RT | | |RT | | | | Q3|>--+
. | | | | | | | | | Q4|
. | 10K | | | | | | | | __ Q5|
. +-/\/\-|RS | +---|RS(phi1) | | | | CE |
. +---------+ +---------+ | | +---------+
. | | |
. | | GND
. | |
. +-------------------+ |
. | |
. | |
. | |
. | .--------. |
. | |74HCT132| |
. | .--------. |
. +----------| \ |
. | | O-+ |
. | +-|__/ | |
. 10K | __ | | |
. +--/\/\---+-| \ | | __ |
. | | O--+ +-| \ |
. +---------+-|__/ | | O--+
. | | | +-|__/
. ===470p | | __ |
. | | +-| \ |
. GND | | O-+
. +----------|__/
.
.
.
Click to expand...
I suppose you can eliminate a chip by doing this- 4 cheap chips and a
DIP sw:
View in a fixed-width font such as Courier.


..
..
..
..
.. +---------------------------------------------------------+----+
.. +-| \ 10K _ | |
.. | | O--/\/\-+----| \ / |
.. +-|__/ | | O----+ 10K |
.. | +|__/ | / |
.. | | | \ |
.. _ | +-------+ | | |
.. GND---o o------+ | | GND |
.. RESET +-------|-+---+ |
.. | __ | | SELECTED -----> |
.. +| \ | | OUTPUT |
.. START | O--+-----|--->OUT MINUTES |
.. GND--o o-+----------|__/ | (DIP SW ) |
.. | | |
.. VCC-/\/\-+ | |
.. 10K +----------------+-------------------+ |
.. | | | |
.. NPO +------4060 +------4060 +------4017 |
.. 220p | MR | | MR | | MR | |
.. +--||--+CT | |CT | | | |
.. | | Q12|--+ | Q13|--+ +--+CP Q1| |
.. |5K pot| | | | | | | | Q2| |
.. +-/\/\-+RT | | |RT | | | | Q3|>--+
.. | | | | | | | | | Q4|
.. | 10K | | | | | | | | __ Q5|
.. +-/\/\-|RS | +---|RS(phi1) | | | | CE |
.. +---------+ +---------+ | | +---------+
.. | | |
.. | | GND
.. | |
.. +-------------------+ |
.. | __ |
.. +---+-| \ |
.. | | O---------------+
.. +-|__/
..
..
..
..
.. 1-5 MINS
.. OUT +------------------+
.. | | | | | |
.. | | | | | |
.. | | | | | |
.. ----+ +------
..
.. ^
.. |
.. press
.. start
..
..
 
Fred Bloggs wrote:
vanagonvw@gmail.com wrote:
I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.



The simplest would be a START/STOP 4060 clocking a 4017 and then
using
the DIP switch to select outputs 1 through 5 of the 4017.
Click to expand...
<schematic diagram snipped>

Fred,

I am much obliged for you to take the time to sketch that circuit out
for me to see. Helps a lot, and once I figure the starting and stopping
sequences, it looks like something I can use, and learn from, so please
pat yourself on the back for me, I appreciate your time and efforts.
:)

Thanks,

John
 
vanagonvw@gmail.com wrote:

I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am a somewhat familiar with a number of one-shots and timer such as
the hc123 and hc4538 for short pulses, and timing, but I am in need of
something that can give me a timeout pulse for up to five minutes. When
I look at the data sheets for these devices, I am not sure where I need
to check to learn if the device is stable enough for a pulse that long,
and if a simple R/C network can accomplish what I am hoping to provide.


The question will be, how accurate? and I can only say that it must be
as accurate as I can make it, and still make it affordable, so that is
why I am wondering where on the datasheet I can find such information
so I can understand my options.

Are these kinds of chips going to be the right approach, or is there
another way I should be accomplishing this long of a timeout with some
degree of accuracy?

I will need to select five timeouts, from one minute, through five
minutes. I thought about using a dipswitch to choose between five
different RC combinations. An idea for something more elegant would
certainly be appreciated.
Thanks for the space and the consideration.

John
Click to expand...
I'm afraid I'm going against the tide here. Unlike the professionals,
I think a 555 circuit with a 5-way rotary switch sounds fine!

You haven't been explicit about the accuracy, but the 555 itself is
capable of 1%. With a low-leakage electrolytic as timing capacitor,
and reasonably careful construction, you could expect close to that in
practice. Here's a circuit to do what you specified:
http://www.terrypin.dial.pipex.com/Images/555Mono+Reset.gif

--
Terry Pinnell
Hobbyist, West Sussex, UK
 
In article <bgmkv0tj8echn497a2qk9romgvtgqtuk28@4ax.com>,
thegreatone@example.com says...
On Fri, 28 Jan 2005 15:21:14 +0000, Terry Pinnell
terrypinDELETE@THESEdial.pipex.com> wrote:
snip
I'm afraid I'm going against the tide here. Unlike the professionals,
I think a 555 circuit with a 5-way rotary switch sounds fine!

You haven't been explicit about the accuracy, but the 555 itself is
capable of 1%. With a low-leakage electrolytic as timing capacitor,
and reasonably careful construction, you could expect close to that in
practice. Here's a circuit to do what you specified:
http://www.terrypin.dial.pipex.com/Images/555Mono+Reset.gif

Naaah! Use a timer running at a reasonable speed where leakage can't
get you, then divide down to the timing you want.
Click to expand...
Use a PIC. <rdh>

--
Keith
 
I'm afraid I'm going against the tide here. Unlike the professionals,
I think a 555 circuit with a 5-way rotary switch sounds fine!

You haven't been explicit about the accuracy, but the 555 itself is
capable of 1%. With a low-leakage electrolytic as timing capacitor,
and reasonably careful construction, you could expect close to that in
practice. Here's a circuit to do what you specified:
http://www.terrypin.dial.pipex.com/Images/555Mono+Reset.gif
Click to expand...
You should be able to use a single pot on the CTL threshold setting:

View in a fixed-width font such as Courier.




V+
|
+------------+-----------+---+---+---+----+
| | | | | | |
| | | | | | |
| | | | | | |
| | +-------------------+
| +-----+-----+ |o o o o o |
| | | | / / / / /|
| | | |o o o o o |
| | | +-------------------+
/ | | | | | | |
\ | | [R] [R] [R] [R] [R]
/<-----|CTL | | | | | |
\ | DIS|-----+---+---+---+----+
| | | |
| | THRESH|-----+
| | | |
| | | |
| +-----+-----+ |
| | ===
| | |CT
| | |
+------------+-----------+-------
|
GND
 
Fred Bloggs wrote:
vanagonvw@gmail.com wrote:

I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am going to amend that circuit to improve the margin against race
on
the reset latch, clocking the 4017 through its CLOCK ENABLE /CE
input,
which is a negative active edge, and this frees the NAND inverter to
drive the MASTER RESET MR of the counters and provide a 3:1 Tpd
margin
on the NAND latch hold time.
Click to expand...
Fred,

I am loathe to admit my ignorance in a public forum, but I have gone
over this change and do not understand how it will work when you switch
the enable and clock pins. If I had the parts in-hand, I would
breadboard it and learn it that way, but they won't be here till the
end of the week. I am staring at the data sheet and not seeing how the
4017 will get clocked at all using this method.

Would you mind taking a minute and explaining how that last stage, the
4017, will get clocked correctly with these changes?
Thanks for your consideration, and your patience.

John
 
On Fri, 28 Jan 2005 15:21:14 +0000, Terry Pinnell
<terrypinDELETE@THESEdial.pipex.com> wrote:

vanagonvw@gmail.com wrote:

I have a question about one-shots, multivibrators, and other like
devices, if I may be permitted.

I am a somewhat familiar with a number of one-shots and timer such as
the hc123 and hc4538 for short pulses, and timing, but I am in need of
something that can give me a timeout pulse for up to five minutes. When
I look at the data sheets for these devices, I am not sure where I need
to check to learn if the device is stable enough for a pulse that long,
and if a simple R/C network can accomplish what I am hoping to provide.


The question will be, how accurate? and I can only say that it must be
as accurate as I can make it, and still make it affordable, so that is
why I am wondering where on the datasheet I can find such information
so I can understand my options.

Are these kinds of chips going to be the right approach, or is there
another way I should be accomplishing this long of a timeout with some
degree of accuracy?

I will need to select five timeouts, from one minute, through five
minutes. I thought about using a dipswitch to choose between five
different RC combinations. An idea for something more elegant would
certainly be appreciated.
Thanks for the space and the consideration.

John

I'm afraid I'm going against the tide here. Unlike the professionals,
I think a 555 circuit with a 5-way rotary switch sounds fine!

You haven't been explicit about the accuracy, but the 555 itself is
capable of 1%. With a low-leakage electrolytic as timing capacitor,
and reasonably careful construction, you could expect close to that in
practice. Here's a circuit to do what you specified:
http://www.terrypin.dial.pipex.com/Images/555Mono+Reset.gif
Click to expand...
Naaah! Use a timer running at a reasonable speed where leakage can't
get you, then divide down to the timing you want.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
 
Fred Bloggs wrote:
I'm afraid I'm going against the tide here. Unlike the professionals,
I think a 555 circuit with a 5-way rotary switch sounds fine!

You haven't been explicit about the accuracy, but the 555 itself is
capable of 1%. With a low-leakage electrolytic as timing capacitor,
and reasonably careful construction, you could expect close to that in
practice. Here's a circuit to do what you specified:
http://www.terrypin.dial.pipex.com/Images/555Mono+Reset.gif


You should be able to use a single pot on the CTL threshold setting:

View in a fixed-width font such as Courier.




V+
|
+------------+-----------+---+---+---+----+
| | | | | | |
| | | | | | |
| | | | | | |
| | +-------------------+
| +-----+-----+ |o o o o o |
| | | | / / / / /|
| | | |o o o o o |
| | | +-------------------+
/ | | | | | | |
\ | | [R] [R] [R] [R] [R]
/<-----|CTL | | | | | |
\ | DIS|-----+---+---+---+----+
| | | |
| | THRESH|-----+
| | | |
| | | |
| +-----+-----+ |
| | ===
| | |CT
| | |
+------------+-----------+-------
|
GND
Click to expand...
Hmm- maybe this would work better:

View in a fixed-width font such as Courier.




V+
|
+------------+-----------+---+---+---+---+---+
| | | | | | |
| | | | | | |
| | | | | | |
| | +------------------+
| +-----+-----+ |o o o o o |
| | | | / / / / /|
| | | |o o o o o |
| | | |1 2 3 4 5 | mins
/ | | +------------------+
\ | | | | | | |
/<-----|CTL | | | | | |
\ | DIS|-----+---+---+---+---+---+
| | | |[R] [R] [R] [R] [R]
| | THRESH|-----+
| | | |
| | | |
| +-----+-----+ |
| | ===
| | |CT
| | |
+------------+-----------+-------
|
GND
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top