Multiple Timer Request
Neil
no picchips/ microcontrollers.
All times are approximate.
Hopefully everything is clear, but if not, I'll elaborate upon request. The
operation of timer 3 is the most difficult to explain, but Ill try my best.
1) Button closure activates TIMER1 (operator variable from 1 second to 60
seconds), and stays closed until TIMER1 times out.
2) TIMER2 activates 3 seconds before TIMER3, and times out when TIMER3 times
out.
3) TIMER3 (operator variable from 1 to 15 seconds) activates for preset time
during TIMER1's cycle, and times out 5 seconds before TIMER1 times out.
4)TIMER1 times out at preset time, resetting circuit,opening the switch.
EXAMPLE
If TIMER1 is set for 30 seconds,
and TIMER3 is set for 10 seconds,
Here is a time-line:
{TIMER2 ON
............................... OFF}
{TIMER 3 ON.................OFF}
{TIMER1
ON..........................................................................
........................OFF}{Button open...........
Button
closed-1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,
27,28,29,30,31,32.... seconds
Thanks in advance,
Kim
Chuck Simmons
OK, let's set timer 1 to 1 second. Timer 3 times out 5 seconds before
timer 1. Does this mean that timer 3 activates before the button is
pressed? This is a job for resublimated thiotimoline. You definitely
need some sort of endochronic timer here.
Chuck
--
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons chr...@webaccess.net
Jim Thompson
Chuck, ROTFLMAO! I like that phrase "resublimated thiotimoline", even
if it doesn't pass my spell checker !-)
...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 | |
| Jim-T@analog_innovations.com Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
For proper E-mail replies SWAP "-" and "_"
Democrats, The Axis of the Evil Empire
Chuck Simmons
This is not original. Back in the early 1940s, a grduate student, I
think at Cornell, wrote a paper called "The Endochronic Properties of
Resublimated Thiotimoline." To his horror, the paper appeared in print a
week or two before his thesis defence. He got his PhD anyway and became
quite famous.
Chuck Simmons
I could have suggested that the timers be based on z^n/(z-1), z>1. Sort
of future integrators or "flux capacitors." You got an old Delorean we
can fix up?
Jim Thompson
<chr...@webaccess.net> wrote:
>Jim Thompson wrote:
>>
>> On Fri, 25 Apr 2003 02:00:28 GMT, Chuck Simmons
>> <chr...@webaccess.net> wrote:
[snip]
>> > This is a job for resublimated thiotimoline. You definitely
>> >need some sort of endochronic timer here.
>> >
>> >Chuck
>>
>> Chuck, ROTFLMAO! I like that phrase "resublimated thiotimoline", even
>> if it doesn't pass my spell checker !-)
>
>I could have suggested that the timers be based on z^n/(z-1), z>1. Sort
>of future integrators or "flux capacitors." You got an old Delorean we
>can fix up?
>
>Chuck
I used to see several DeLoreans on the streets around here, but none
in recent years.
EEng
button activating the Timer1, which in turn activates the other two
timers. That's as simple and as cheap as its going to get, although a
PIC16C54C is even cheaper but you already said you don't want any
micro's
On Thu, 24 Apr 2003 21:22:48 -0400, "Neil" <ksl...@sympatico.ca>
wrote:
A E
A E
Chuck Simmons wrote:
If you reverse the polaron field, you can get a sub-space tachyon burst to open a
wormhole to phase space and reverse the time flow locally. Don't forget to
re-route engine plasma through the main deflector using Geordi's visor as a
sensitive phase discriminator.
Ask Wesley for help.
Jonathan Kirwan
It was Isaac Asimov. And it appeared in print just a week or so
prior to having to do his Ph.D. oral exam. The professors had
already read it and were ready for him.
Jon
Samuel Tardieu
A> How much does this pay?
It looks like homework, so I guess you can get a fraction of a diploma
if you solve it :)
Sam
--
Samuel Tardieu -- s...@rfc1149.net -- http://www.rfc1149.net/sam
Bill Sloman
This is poorly specified.
If you can settle for two operator programmable intervals - "start
delay" and "active period" you could do it with four monostables (two
74HCT221 dual monstables), and a bunch of logic gates.
Monostable A is triggered by the push-button, and generates the
programmable start delay.
The trailing edge of the "A" output triggers monostable B, which has a
fixed three second period.
The trailing edge of the "B" output triggers monostable C, which
generates the programmable "active period" output - your Timer 3.
The trailing edge of the "C" output triggers monostable D, which has a
fixed five second period.
Your Timer 1 output is then the OR of the A,B,C and D. There will be
glitches of tens of nanoseconds between each of A, B, C and D - you
might need an analog low-pass filter on the output to get rid of them,
in which case you'd want to buffer the output with a Schmitt trigger
inverter.
Your "Timer 1" period ceases to be independently programmable - it is
just the sum of the "start delay", the "active period" and the 8
seconds of fixed delays.
You can stretch it by adding a fifth monostable E whose pulse width
would a third programmable input "recovery time".
Monostable E would be triggered by the trailing edge of the "D"
output, and your Timer 1 output would be the logical OR of A,B,C,D and
E.
This would be messy. Periods up to 37 sec can be generated by 74HCT221
monostables, but the capacitors and resistors get big, and the whole
system is a bit vulnerable to noise. You'd presumably be programming
the delays with multi-turn potentiometers on a front panel, which
shouldn't be the timing resistors in the standard 74HCT221 timing
circuit - using them to set a current level from a constant current
source (probably a PNP current mirror) works a lot better.
You could do a version of this scheme as a fully digital system in a
decent sized programmable logic device (PLD), but then your
programmable panel consists of a bunch of binary-coded decimal
switches, with four leads per digit, which could use up a lot of pins
on your PLD. It would be more reliable than the monostable-based
solution, but would take some programming - programming a PLD is even
more hands-on than machine language.
A PIC is a lot neater.
Hope this helps.
--------------
Bill Sloman, Nijmegen
Fred Bloggs
Neil wrote:
> Ok, I need a simple, low parts count circuit for a industrial usage. Please
> no picchips/ microcontrollers.
> All times are approximate.
> Hopefully everything is clear, but if not, I'll elaborate upon request. The
> operation of timer 3 is the most difficult to explain, but Ill try my best.
>
> 1) Button closure activates TIMER1 (operator variable from 1 second to 60
> seconds), and stays closed until TIMER1 times out.
>
> 2) TIMER2 activates 3 seconds before TIMER3, and times out when TIMER3 times
> out.
>
> 3) TIMER3 (operator variable from 1 to 15 seconds) activates for preset time
> during TIMER1's cycle, and times out 5 seconds before TIMER1 times out.
Really? How is Timer3 going to time out 5 seconds before Timer1 when
Timer1 has been adjusted to a 1 second duration?
onestone
unless you seriously think you can create a time machine without a PIC
;@}.
In reality you need a variable timer from 0 to 51 seconds that starts
with the button push, this is really TIMER1, when that timer expires it
starts the next timer, which is a fixed 3 second delay (TIMER2), the
third timer is your TIMER3, which is programmable from 1 to 15 seconds,
finally the last stage is a fixed 5 second timer, four cascaded stages.
You will also perhaps need a little logic, like an SR latch or similar
that is SET on the button press and reset by the expiry of the fixed 5
second timer. A sane engineer would probably use a cheap micro, as it's
cheaper and more flexible than the discrete solution, it's also less
prone to drift. You could use 555's here (or 556's), or one shots. IIRC
there is a long period version of the 74123 but I don't remember it's
part number, since I'd use a micro in the circumstances. The problem
with the cascaded approach, and with your description in general, is
that TIMER1 isn't really programmable, it is a function of TIMER3. For
example it's minimum period is NOT 1 second given the above description,
its the 3 seconds of TIMER2, plus the minimum 2 seconds of TIMER3 plus
the fixed delay of 5 seconds, hence 9 seconds. However it's minimum
period depends upon TIMER3, hence it cannot be independently set. This
would require the operator to know that he must account for TIMER3 when
setting TIMER1. If this is a real world problem, and not a homework
issue, I would NEVER rely on the operator to figure this out, or take
the time to do it right, hence have a micro to figure things out for
them. It is cheaper, it is more reliable, it is more flexible, it can
deal with anomalous input, like TIMER2 firing before the button is
pressed, not considering it is senseless. Why would any engineer
deliberately restrict their tool set?
Cheers
Al
Fred Bloggs
An analog approach will be roughly 2x quad comparators like the the
LM339 and a single quad single supply opamp such as the LM324. The
scheme is diagrammed below. You use two front panel potentiometers to
adust A*time1, the TIMER1 duration threshold voltage, and A*time3, the
TIMER3 duration threshold voltage. These are applied to difference
amplifiers to produce Vth1, Vth2L, Vth3L, and Vth3U. You can use the
remaining comparators to create a debounced input latch triggered by the
button press and cleared by the negative going edge of timer1 output-
same latch used to "enable" all comparator outputs. The input ramp can
be a simple transistor current source driving a grounded capacitor-
capacitor shorted when input latch is cleared. There maybe additional
requirements depending upon more specifics.
View in a fixed-width font such as Courier.
--Input timing ramp
| --
| -- Vin=A*t A=Volts/sec
| --
| -- Applied to comparators
| -- in parallel
-+------------------
|
| ON
|-----------------+
TIMER1 | |
COMPRTR | |
| |
| | OFF
-+------------------------------
Vth1=A*time1
|
| ON
| +----------+
TIMER2 | | |
WINDOW | | |
CMPRTR | | |
| | | OFF
-+------------------------------
Vth2L Vth2U=Vth3U
=
Vth3L-3*A
|
| ON
| +------+
TIMER3 | | |
WINDOW | | |
CMPRTR | | |
| | | OFF
-+------------------------------
Vth3L Vth3U=Vth1-A*5
=
Vth3u-A*time3
Klaus Vestergaard Kragelund
news:3EA8CCF8...@videotron.ca...
That last sentence cracked me. Good one :-)
Cheers
Klaus
Chuck Simmons
Yep. I think it was John W. Campbell that jumped the gun. John had not
understood Asimov's timing and quite innocently published the piece.
After Asimov got his PhD, it was all a bit of a joke but during the week
between publication and the exam, I'm sure Asimov was quite nervous.
In the late 1950s, Asimov caused me some embarrassment. My school
library had Asimov's "The World of Carbon." I was familiar with Asimov
science fiction then but I was even more fascinated by his popular
science writing. My mother dropped me at a book store an Saturday as was
custom and I asked for books on science by Asimov. The clerk responded
loudly so all in the store could hear, "But Asimov is a science fiction
writer." If I ever had an urge to kill, it was then.
Years later it was quite respectable to ask for science by Asimov. I
consider Asimov works in science fiction and popular science of equal
value. The man was an entertainer and a professor in the truest sense.
Bill Sloman
Depends where you were. It Tasmania I got his "World of Carbon" as an
end of year prize at school - which must have been before 1960- and I
was reading his science fiction at the same time (including "The
Endochronic Properties of
Resublimated Thiotimoline." in ASF, though not when it was first
published.)
------
Bill Sloman, Nijmegen
Neil
making this for. I only wish to use good old fashoned non-pic circuits, only
because I have VERY limited resources at work, and I have yet to get into
understanding them.
As well, timer 2 is wrong, It should only activate 1 second prior to timer
3.
the machines that I need this circuit to control is called a vacuum packer,
used to vacuum package package food in a sealed plastic bag.
There are only 2 user-controlled variables. One is the length the vacuum
runs, depending on the size of the food item, and bag, and the other is the
time that the sealing heating element is on, depending on the thickness of
the bag.
Its operation is as follows.
1-Lid is closed, activating vacuum pump, sucking all the air out of the
chamber, and the bag.
2-Bag lip is pushed closed 1 second before sealing wire is activated
3-Sealing wire is activated, sealing the bag
4-Valve is opened, allowing atmospheric room pressure to enter the chamber.
5-Lid springs open, opening the switch
Tah Dah!.
I hope this answers any rogue questions,
and thanks for the help.
Kim
"Jim Thompson" <Jim-T@analog_innovations.com> wrote in message
news:g97havkk1ja5tdb5m...@4ax.com...
Terry Pinnell
that you should expect to offer payment for this. After all, it
appears to be for commercial use, and it's a complete circuit design
you're seeking, not just a tricky or troublesome section or issue.
Relatively-complex too, if reliability is to be assured.
I've been looking forward to reading your feedback after the fairly
detailed suggestions from Bill Sloman and Fred Bloggs respectively.
--
Terry Pinnell
Hobbyist, West Sussex, UK
Tony Williams
Neil <ksl...@sympatico.ca> wrote:
> Its operation is as follows.
> 1-Lid is closed, activating vacuum pump, sucking all the air out of the
> chamber, and the bag.
> 2-Bag lip is pushed closed 1 second before sealing wire is activated
> 3-Sealing wire is activated, sealing the bag
> 4-Valve is opened, allowing atmospheric room pressure to enter the chamber.
> 5-Lid springs open, opening the switch
> Tah Dah!.
Try another way of thinking the sequence?
1. Lid is closed; Vacuum pump and rest of circuit energised.
2. A pressure switch senses when the required low pressure
has been reached, and activates bag-lip closure.
3. A microswitch senses when the bag-lip has closed, and
kicks the (one and only) TIMER to power up the sealing wire.
4. Valve is opened, etc....
--
Tony Williams.
John Crighton
wrote:
>It may actually help if I explain the functions of the machine that I am
>making this for. I only wish to use good old fashoned non-pic circuits, only
>because I have VERY limited resources at work, and I have yet to get into
>understanding them.
>As well, timer 2 is wrong, It should only activate 1 second prior to timer
>3.
>the machines that I need this circuit to control is called a vacuum packer,
>used to vacuum package package food in a sealed plastic bag.
>There are only 2 user-controlled variables. One is the length the vacuum
>runs, depending on the size of the food item, and bag, and the other is the
>time that the sealing heating element is on, depending on the thickness of
>the bag.
>
>Its operation is as follows.
>1-Lid is closed, activating vacuum pump, sucking all the air out of the
>chamber, and the bag.
>2-Bag lip is pushed closed 1 second before sealing wire is activated
>3-Sealing wire is activated, sealing the bag
>4-Valve is opened, allowing atmospheric room pressure to enter the chamber.
>5-Lid springs open, opening the switch
>Tah Dah!.
>
>I hope this answers any rogue questions,
>and thanks for the help.
>Kim
Hello Kim,
from your description above, a device called a
PLC (programable logic controller) came to mind.
There are a number of contact closure inputs and
contact closure outputs. The internal circuitry gives
you dozens/scores of relays, latches, counters and
timers to play with between your input and output.
I have an old Toshiba EX20. Here is a picture
http://www.tic.toshiba.com.au/web/plc/upgrades.htm
I paid about $500 (Australian) for the PLC and Programer unit
about 14 years ago so this model is probably very cheap now.
This EX20 unit has facilities for a plug in PROM module, so
if you had a couple of different procedures for your packaging
machine to perform then you just swap modules.
If you can pick up an old PLC cheaply it will make the control
of your machine very easy. Changes to timers and sequences
are easy to make.
I just typed PLC into the search box on ebay and lot of
stuff came up, expensive to very cheap, tens of dollars.
Anyway, it was just a thought.
Regards,
John Crighton
Hornsby
New South Wales
John Fields
>It may actually help if I explain the functions of the machine that I am
>making this for. I only wish to use good old fashoned non-pic circuits, only
>because I have VERY limited resources at work, and I have yet to get into
>understanding them.
>As well, timer 2 is wrong, It should only activate 1 second prior to timer
>3.
>the machines that I need this circuit to control is called a vacuum packer,
>used to vacuum package package food in a sealed plastic bag.
>There are only 2 user-controlled variables. One is the length the vacuum
>runs, depending on the size of the food item, and bag, and the other is the
>time that the sealing heating element is on, depending on the thickness of
>the bag.
>
>Its operation is as follows.
>1-Lid is closed, activating vacuum pump, sucking all the air out of the
>chamber, and the bag.
>2-Bag lip is pushed closed 1 second before sealing wire is activated
>3-Sealing wire is activated, sealing the bag
>4-Valve is opened, allowing atmospheric room pressure to enter the chamber.
>5-Lid springs open, opening the switch
>Tah Dah!.
It would help immeasurably if you could supply a timing diagram,
something like this:
_______________________________
LID____| |________________________
_________________________
VALVE____| |____________________________
____________
VACUUM_____| |_______________________________________
______
PUSHLIP__________________| |_______________________________
__
SEAL_______________________| |_________________________________
Also, you haven't mentioned anything about the valve, other than that
it's opened to bring the system down to air and open the lid. Is this
done automatically? How is the valve closed?
This is a simple application requiring a handful of one-shots and maybe
a little glue logic, but without being adequately described, up front,
it becomes a PITA to design by the successive elimination of assumptions
and errors. Do an adequate description and supply us with a timing
diagram annotated with times, and you'll probably get more help than you
can use!^)
So far, everything you've said is reflected in the timing diagram
following, and the relationship between the leading edge of LID and the
leading edge of VALVE isn't clear. How about editing it so it shows
what you want? Use a fixed pitch font like Courier or Courier New and
it ought to be fairly painless to read back. I used Courier New to
generate it, so it would probably work best if you edited it with
Courier New.
________________________________
LID____||||| |___________________
?-->| |<---
_______________________________
VALVE____| |______________________
var-->| |<--- |
____________ |
VACUUM_____| |______________ | ____________________
|
?-->||<-- -->| |<--?
___________
PUSHLIP__________________| |__________________________
1s --->| |<- ->| |<--?
_______
SEAL_______________________| |____________________________
var--->| |<---
Neil
ancient. The units are so old (sometimes more than 20 years) that their
electronic circuits are slowly giving up, and the circuits use ancient
technology that we have been told can no longer be repaired.
So far petrus b. has done the best job of simplifying and explaining the
functions yet, and I thank him for it.
Below is a copy of his posting.
Kim
As you see I rearanged your description, so now we have a more or less
correct specification of the problem. Nervertheless I recapitulate:
We have a lid with two functions.
1) A switch (output) that starts the proces
2) An input that blocks the switch or the lid for the time of the proces
Furthermore you have to serve a vacuumpump, a sealing wire and a valve.
The design proces is straight forward:
-On closing the lid, the switch has to set a flipflop
-The output of the flipflop blocks the lid
-The output of the flipflop also activates timer1, the 1-60s timer
-The output of timer1 serves the vacuumpump
-On timing out, timer1 activates timer2, a 1s timer
-(during this time the bag lip is pushed)
-On timing out, timer2 activates timer3, the 1-15s timer
-The output of timer3 serves the sealing wire
-On timing out, timer3 activates timer4, a 5s timer
-The output of timer4 serves the valve
-On timing out, timer4 resets the flipflop.
Succes,
pieter
"Fred Bloggs" <nos...@nospam.com> wrote in message
news:3EA9531...@nospam.com...
Neil
Succes,
pieter
ånønÿmøu§
<to...@ledelec.demon.co.uk> wrote:
>> Its operation is as follows.
>> 1-Lid is closed, activating vacuum pump, sucking all the air out of the
>> chamber, and the bag.
>> 2-Bag lip is pushed closed 1 second before sealing wire is activated
>> 3-Sealing wire is activated, sealing the bag
>> 4-Valve is opened, allowing atmospheric room pressure to enter the chamber.
>> 5-Lid springs open, opening the switch
>
> 1. Lid is closed; Vacuum pump and rest of circuit energised.
> 2. A pressure switch senses when the required low pressure
> has been reached, and activates bag-lip closure.
> 3. A microswitch senses when the bag-lip has closed, and
> kicks the (one and only) TIMER to power up the sealing wire.
> 4. Valve is opened, etc....
above functions takes too long...
Tony Williams
ånønÿmøu§ <res...@in.conference.com> wrote:
Could be.... The advantage of sequential interlocks
is that it guarantees that each process has completed
properly before the next process is allowed to start.
As you note, a failure of any one process will stop
the machine completely, so maybe a single 60 sec
overall alarm could be used.
With arbitrary timing it is possible for any one
process to fail and still complete (bad) product. If
it is food being vac-packed then that would not be
good news.
As remarked already, it is a job for a PLC (either way).
--
Tony Williams.