coutdown 555 timer values?

[John Fields]

On Sun, 24 Jul 2005 13:39:29 -0400, John Popelish <jpopelish@rica.net>
wrote:

John Fields wrote:
On 22 Jul 2005 14:01:01 -0700, "aman" <aman.bindra@gmail.com> wrote:


You might want to consider a simple 8 bit microcontroller like PIC. All
micro-controllers have timer interrupts. If not using a CD4017 as a
counter and using 555 to generate a 30 sec pulse seems like a good idea.


---
It may seem like it is, but in reality it isn't.

He would have to add some logic, like some SR flip flops and a bit
more. I just get worried about the stability of more than a minute
versions of 555s.

Note that the OP wanted six timeout periods all starting at once, but
as they timed out, each one lasting 30 seconds longer than the one
which timed out before it. Like this:

___
T1___| |__________________________________________

______
T2___| |_______________________________________

_________
T3___| |____________________________________

____________
T4___| |_________________________________

_______________
T5___| |______________________________

__________________
T6___| |___________________________


You can't do that with a 4017.

I'll bet you could come up with something neat based on the 4015 shift
register.

---


The tricky part is shortening up that first 555 pulse in astable mode
(which is longer than the rest because of the need to get the timing
cap up to 2/3Vcc from 0V (instead of 1/3Vcc) on startup) and getting
all the shift register outputs to go high on startup.

Using something like three HC74's would solve the "start high" problem
by using the Qbars as outputs, using their SET inputs for POR, and
using their RESET inputs to start the timing sequence. Getting rid of
the extended first pulse problem could be done by using a couple of
555 as monostables in a ring counter.

Schematic on abse as "555 countdown timer (from seb), or:

newsoece11tgcals43avpikokbd73ubinh5kj@4ax.com


Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.


--
John Fields
Professional Circuit Designer

 

[John Fields]

On 25 Jul 2005 16:04:44 -0700, "Chris" <cfoley1064@yahoo.com> wrote:

Actually, one of the major problems here that nobody is talking about
is the power supply. The 555 changes state on 1/3Vcc and 2/3Vcc. If
the supply is unregulated or poorly regulated, the power supply sags
are going to change things quite a bit, too.

---
Not really, since the trigger and threshold volatges are derived
ratiometrically from Vcc through a three-resistor voltage divider.

--
John Fields
Professional Circuit Designer

 

[John Fields]

On 26 Jul 2005 06:26:42 -0700, "Tristar500"
<robot@innovationrobotics.com> wrote:

Hi Chris, Thanks for taking the time to reply. I hadn't thought of
using cmos 555's . I understand they and other cmos ic's are popular
with very low current draw is needed. I was thinking of using a
computer power supply or a cell phone charger as a regulated power
supply. I know the computer supplies are well regulated, not so sure
about those little black cubes we all have at the end or our cell phone
chargers, answering machine power supplies etc. Here is a link to a
jpg of the schematic I put together from the earlier post from John
Fields along with some resister values added by other members.

http://www.rcgroups.com/gallery/zips/27524/biketimer.jpg

Anybody care to comment? I'm ready to etch a board if all looks good.

---
Go the shift register route and save yourself some grief.

Check out the schematic on abse as "555 countdown timer (from seb)",
or:

newsoece11tgcals43avpikokbd73ubinh5kj@4ax.com

--
John Fields
Professional Circuit Designer

 

[Chris]

Tristar500 wrote:

Hi Chris, Thanks for taking the time to reply. I hadn't thought of
using cmos 555's . I understand they and other cmos ic's are popular
with very low current draw is needed. I was thinking of using a
computer power supply or a cell phone charger as a regulated power
supply. I know the computer supplies are well regulated, not so sure
about those little black cubes we all have at the end or our cell phone
chargers, answering machine power supplies etc. Here is a link to a
jpg of the schematic I put together from the earlier post from John
Fields along with some resister values added by other members.

http://www.rcgroups.com/gallery/zips/27524/biketimer.jpg

Anybody care to comment? I'm ready to etch a board if all looks good.

Thanks, Lawrence

That pretty much looks like John Fields' circuit, with one exception.
For long-duration timers, there _must_ be a cap from pin 5 (control) to
GND. Pin 5 is connected to the 2/3Vcc point of the three 5K resistors
that are used as a voltage divider to set threshold and trigger values.
The cap helps filter out power supply noise from messing up your
timing. An 0.01uF cap is usually used here.

Also, you should have power supply bypass caps from pin 8 of each 555
to GND (0.1uF ceramics should work here), as well as having a bigger
(100uF or more) cap on the board for filtering the power supply.

With those exceptions, if you do the layout right you should be able to
get something that works for your +/-20% timing spec. As I said, it
might be better to use a CMOS 555, and use the 1000uF low leakage caps
and lower resistor values specified in the post above, but that
shouldn't affect layout.

If you want better tolerance on the timing, you can compensate for the
tolerance of the caps by tweaking resistor values up or down. But,
like I said, John Fields' basic circuit will work.

You didn't mention what you're doing with the outputs here. If you're
driving a non-resistive load, be careful. For relays, it might be
better to drive them with an NPN transistor, and use a diode across the
coil. And if the relays are driving an inductive load, use snubbers on
the load or at least try to physically separate the relays as much as
possible from the timer ICs.

If you're sourcing a lot of current (more than a couple of mA) from the
output, you might want to consider using transistors and having the
power wiring for the output wired separately from the power supply for
the timer board.

Make your Vcc and GND traces as beefy as possible to minimize resistive
voltage drops from point to point.

I'd be a little concerned about the load for these 555s. Please post
again, and let us know what you're doing with the outputs.

Thanks
Chris

 

[Tristar500]

Link doesn't work John?

 

[John Popelish]

John Fields wrote:

The tricky part is shortening up that first 555 pulse in astable mode
(which is longer than the rest because of the need to get the timing
cap up to 2/3Vcc from 0V (instead of 1/3Vcc) on startup) and getting
all the shift register outputs to go high on startup.

I think this is easily solvable by using a pair of capacitors in
series across the supply as the timing capacitor. If you use one
capacitor about twice as large as the outer, the mid point voltage
will begin the first timing cycle at about 1/3rd of the total voltage,
just like it will be at the end of the next cycle.

Using something like three HC74's would solve the "start high" problem
by using the Qbars as outputs, using their SET inputs for POR, and
using their RESET inputs to start the timing sequence. Getting rid of
the extended first pulse problem could be done by using a couple of
555 as monostables in a ring counter.

Schematic on abse as "555 countdown timer (from seb), or:

newsoece11tgcals43avpikokbd73ubinh5kj@4ax.com


Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.

This is a lot simpler than a whole batch of 555s running in parallel.

 

[John Popelish]

John Fields wrote:

On 25 Jul 2005 16:04:44 -0700, "Chris" <cfoley1064@yahoo.com> wrote:


Actually, one of the major problems here that nobody is talking about
is the power supply. The 555 changes state on 1/3Vcc and 2/3Vcc. If
the supply is unregulated or poorly regulated, the power supply sags
are going to change things quite a bit, too.


---
Not really, since the trigger and threshold volatges are derived
ratiometrically from Vcc through a three-resistor voltage divider.

That only helps if the supply variations happen on a much longer time

scale than the timing cycle. If there is 120 Hz ripple during a 30
second time phase, the ripple will shorten the period.

 

[Tristar500]

Hi All, the outputs are going to be used to trigger 6 separate digital
voice recorders (radio shack Model: 276-1323) I was thinking something
like (Mouser part number 433-812H ) PCB relay used to close the switch
on the voice recorder.

The voice recorders will have various recorded messages encouraging the
children to keep pedeling.

Thoughts?

Thanks again guys!

 

[John Fields]

On 26 Jul 2005 08:17:53 -0700, "Tristar500"
<robot@innovationrobotics.com> wrote:

Link doesn't work John?

---
Yeah, it does, but you can't get there from Google.

I'll email it to you.

--
John Fields
Professional Circuit Designer

 

[Chris]

Tristar500 wrote:

Hi All, the outputs are going to be used to trigger 6 separate digital
voice recorders (radio shack Model: 276-1323) I was thinking something
like (Mouser part number 433-812H ) PCB relay used to close the switch
on the voice recorder.

The voice recorders will have various recorded messages encouraging the
children to keep pedeling.

Thoughts?

Thanks again guys!

One of the great things about electronics is that "contrivers of
contrivances" aren't limited to one and only one solution. Usually,
there are many different solutions, some less expensive, some more
elegant or smaller, and you can have many acceptable answers to your
problem. It's not like a jigsaw puzzle.

I'm not sure about this, but I would think you actually want the first
output to go active after 30 seconds, remain active for enough time to
go through the voice recording, then go off. You want the second
output to go active after 60 seconds, remain active for a few seconds,
then go off. The third output starts at 90 seconds, and so on.

Either that, or you'd want the first output to go on as soon as the
bike starts, and remain on for a few seconds. It goes off, and then at
30 seconds, the second one goes on for a few seconds. At 90 seconds,
the third goes on for a few seconds, and so on.

All of the above responses have been based on the wording of your
original post. With the 6 X CMOS 555 circuit, all of the outputs (and
the relays) will turn on at the same time. The first relay would turn
off after 30 seconds, the second after 60 seconds, and so on. You will
have six module voices chirping different things continuously, and
every 30 seconds, one of them will go silent until they are all
stilled. Is that what you want? Carefully examine the wording of your
first post. That is exactly what you asked for, and that's what Mr.
Popelish, Mr. Fields and the rest have been discussing.

Please take a minute, and carefully think about what you want the
electronics to actually do for you. Then write something down which
really describes that action. Include how long you need each relay to
be on, please. From what you've said, you probably actually need
something completely different from what we've talked about.

By the way, don't feel too discouraged about this. Inadequate project
description has snagged more engineers and electronics people than
practically any other problem. It's a real gumption trap, because
actually, you've made something that works -- it just doesn't work the
way you want. Fortunately, you caught it before you wasted time making
something that would not do what you need.

Thanks
Chris

 

[John Fields]

276-1323On 26 Jul 2005 08:52:49 -0700, "Tristar500"
<robot@innovationrobotics.com> wrote:

Hi All, the outputs are going to be used to trigger 6 separate digital
voice recorders (radio shack Model: 276-1323) I was thinking something
like (Mouser part number 433-812H ) PCB relay used to close the switch
on the voice recorder.

The voice recorders will have various recorded messages encouraging the
children to keep pedeling.

Thoughts?

---
Aaarghhhh!!!

This is what you said you needed:
___
T1____| |______________________________
_______
T2____| |__________________________
___________
T3____| |______________________
_______________
T4____| |__________________
___________________
T5____| |______________
_______________________
T6____| |__________


But, according to what you posted above, you need the following
instead:


-->| |<---Tw
-->| |<---Tp
_ |
T1____| |__________________________________
_
T2_________| |_____________________________
_
T3______________| |_________________________
_
T4___________________| |____________________
_
T5________________________| |_______________
_
T6_____________________________| |__________

Where Tw is the width of the pulse necessary to turn on the recorder,
and Tp is the time between pulses, ostensibly 30 seconds.

What is it exactly, that you want to accomplish?

--
John Fields
Professional Circuit Designer

 

[Tristar500]

Hi all,

OK, Here is EXACTLY what the circuit needs to do. (Thanks Chris for
making sure we are all on the same page and we don't all waste a lot of
time due to misunderstanding.

Kid gets on exercise bike. Instructions instruct him to start pedaling
and then push the start button. (I thought this would be necessary so
that the missing pulse detector doesn't immediately restart the routine
because it thinks the kid has stopped pedaling)

After 30 seconds a prerecorded voice tells the kid he's doing good,
keep going, yada, yada..

At the 60 second mark another a prerecorded voice tells the kid he's
doing good, keep going, yada, yada..

At the 90 sec... and so forth

At the 3 minute mark the final prerecorded voice unit sings him praise
(I'll probably add a relay here so flashing lights or whatever can
also add to the grand finally) and then the whole thing shuts down and
waits for the next victim, er, I mean rider.

If the child stops pedaling or pedals so slow the entire circuit should
reset forcing the rider to start pedaling, hit the start button etc.

The digital voice recorders have two buttons on them (radio shack
Model: 276-1323), one you press to record your message, the other to
play it back. It's a momentary contact switch so I don't think the
trigger out to the voice recorder needs to stay on for more then a few
milliseconds or so.

I have a proximity sensor and amplifier of the type commonly used on
automation equipment to sense if the wheel is turning. Alternately I
could use a reed relay as suggested earlier.

I really can't think of anything else to add.

Thanks guys, someday this project will make a lot of kids happier and
in better physical shape!

Lawrence

 

[Bob Monsen]

Tristar500 wrote:

Hi all,

OK, Here is EXACTLY what the circuit needs to do. (Thanks Chris for
making sure we are all on the same page and we don't all waste a lot of
time due to misunderstanding.

Kid gets on exercise bike. Instructions instruct him to start pedaling
and then push the start button. (I thought this would be necessary so
that the missing pulse detector doesn't immediately restart the routine
because it thinks the kid has stopped pedaling)

After 30 seconds a prerecorded voice tells the kid he's doing good,
keep going, yada, yada..

At the 60 second mark another a prerecorded voice tells the kid he's
doing good, keep going, yada, yada..

At the 90 sec... and so forth

At the 3 minute mark the final prerecorded voice unit sings him praise
(I'll probably add a relay here so flashing lights or whatever can
also add to the grand finally) and then the whole thing shuts down and
waits for the next victim, er, I mean rider.

If the child stops pedaling or pedals so slow the entire circuit should
reset forcing the rider to start pedaling, hit the start button etc.

The digital voice recorders have two buttons on them (radio shack
Model: 276-1323), one you press to record your message, the other to
play it back. It's a momentary contact switch so I don't think the
trigger out to the voice recorder needs to stay on for more then a few
milliseconds or so.

I have a proximity sensor and amplifier of the type commonly used on
automation equipment to sense if the wheel is turning. Alternately I
could use a reed relay as suggested earlier.

I really can't think of anything else to add.

Thanks guys, someday this project will make a lot of kids happier and
in better physical shape!

Lawrence

You need a C555 and a cmos 4017, along with some transistors or relays
to act as momentary switches.

On the other hand, you could use a "winbond ISD4002", along with a bit
of glue (or a microcontroller) to do the entire thing, replacing the
radio shack digital recorders. (they probably contain this chip, a mic,
a speaker, and a battery.)

http://www.futurlec.com/ICSFOthers.shtml

Or, you could use an eprom and some glue logic to replace the pricy
ISD4002 chip. This would probably be the cheapest option, although
getting the audio right might be a chore.

--
Regards,
Bob Monsen

If a little knowledge is dangerous, where is the man who has
so much as to be out of danger?
Thomas Henry Huxley, 1877

 

[John Fields]

On Tue, 26 Jul 2005 11:36:27 -0400, John Popelish <jpopelish@rica.net>
wrote:

John Fields wrote:

The tricky part is shortening up that first 555 pulse in astable mode
(which is longer than the rest because of the need to get the timing
cap up to 2/3Vcc from 0V (instead of 1/3Vcc) on startup) and getting
all the shift register outputs to go high on startup.

I think this is easily solvable by using a pair of capacitors in
series across the supply as the timing capacitor. If you use one
capacitor about twice as large as the outer, the mid point voltage
will begin the first timing cycle at about 1/3rd of the total voltage,
just like it will be at the end of the next cycle.

---
Yeah, that's clever!
---

Using something like three HC74's would solve the "start high" problem
by using the Qbars as outputs, using their SET inputs for POR, and
using their RESET inputs to start the timing sequence. Getting rid of
the extended first pulse problem could be done by using a couple of
555 as monostables in a ring counter.

Schematic on abse as "555 countdown timer (from seb), or:

newsoece11tgcals43avpikokbd73ubinh5kj@4ax.com


Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.

This is a lot simpler than a whole batch of 555s running in parallel.

---
Yup, and here are the values for the passives:

VCC>------+----+
| | V+
| [787K] |
| | [10K]
[499] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[232] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[340] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[511] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[750] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[1050] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[1620] |+ G--+
| [100ľF] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

I haven't shown any hysteresis, but a megohm or so from each of the
comparators' outputs back to their non-inverting inputs ought to take
care of it without upsetting the thresholds too much, considering that
that 100ľF is +/- 10% at best.

Unfortunately, it turns out that the OP needed something other than
what he asked for, so he won't be able to use the circuit. :-(

Fortunately, it was a fun exercise to linearize that cap's response
without having to resort to an active current source!

--
John Fields
Professional Circuit Designer

 

[ehsjr]

John Fields wrote:

Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.

Elegant solution. "Mikey likes it!"

Ed

 

[Bob Monsen]

John Fields wrote:

Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.

I know the thread took a left turn, but I'll just point out that one may
want to use a constant current source to charge CT. Then, all the
resistors can be identical. If I is constant, dV/dt is also constant.
One can get quad packages of comparators...

--
Regards,
Bob Monsen

If a little knowledge is dangerous, where is the man who has
so much as to be out of danger?
Thomas Henry Huxley, 1877

 

[John Fields]

On Wed, 27 Jul 2005 04:13:34 GMT, ehsjr <ehsjr@bellatlantic.net>
wrote:

John Fields wrote:


Then, of course, there's always:

VCC>------+----+
| | V+
| [RT] |
| | [10K]
[R7] +----|-\ |
| | | >--+-->180s
+----|----|+/
| | V+
| | |
[R6] | [10K]
| +----|-\ |
| | | >--+-->150s
+----|----|+/
| | V+
| | |
[R5] | [10K]
| +----|-\ |
| | | >--+-->120s
+----|----|+/
| | V+
| | |
[R4] | [10K]
| +----|-\ |
| | | >--+-->90s
+----|----|+/
| | V+
| | |
[R3] | [10K]
| +----|-\ |
| | | >--+-->60s
+----|----|+/
| | V+
| | |
[R2] | [10K]
| +----|-\ |
| | | >--+-->30s
+----|----|+/
| |
| +----D
[R1] | G--+
| [CT] S |
| | | |
GND>------+----+----+ |
_ |
R/T>---------------------+

In this circuit we use six comparators with references derived
ratiometrically from Vcc, which will make the circuit insensitive to
longish-term variations in Vcc. The circuit works by applying a short
positive pulse to the gate of the MOSFET, which discharges CT and
forces all of the inverting (-) inputs to the comparators lower than
the reference voltages on their non-inverting (+) inputs. This will
cause all of the outputs to go high. Then, when CT charges it will
rise through each of the reference voltages on the + inputs
sequentially, turning each of the comparators ON sequentially,
forcing, and holding, their outputs low, one at a time.

In order to determine the widths of the various outputs we can use

Vcc
T = RC ln ----------- = kRC
Vcc - Vth

to determine what the voltages will be on the - inputs of the
comparators at the times we're interested in.



Elegant solution. "Mikey likes it!"

---
Thanks!

--
John Fields
Professional Circuit Designer

 

[John Fields]

On 26 Jul 2005 10:57:25 -0700, "Tristar500"
<robot@innovationrobotics.com> wrote:

Hi all,

OK, Here is EXACTLY what the circuit needs to do. (Thanks Chris for
making sure we are all on the same page and we don't all waste a lot of
time due to misunderstanding.

Kid gets on exercise bike. Instructions instruct him to start pedaling
and then push the start button. (I thought this would be necessary so
that the missing pulse detector doesn't immediately restart the routine
because it thinks the kid has stopped pedaling)

After 30 seconds a prerecorded voice tells the kid he's doing good,
keep going, yada, yada..

At the 60 second mark another a prerecorded voice tells the kid he's
doing good, keep going, yada, yada..

At the 90 sec... and so forth

At the 3 minute mark the final prerecorded voice unit sings him praise
(I'll probably add a relay here so flashing lights or whatever can
also add to the grand finally) and then the whole thing shuts down and
waits for the next victim, er, I mean rider.

If the child stops pedaling or pedals so slow the entire circuit should
reset forcing the rider to start pedaling, hit the start button etc.

The digital voice recorders have two buttons on them (radio shack
Model: 276-1323), one you press to record your message, the other to
play it back. It's a momentary contact switch so I don't think the
trigger out to the voice recorder needs to stay on for more then a few
milliseconds or so.

I have a proximity sensor and amplifier of the type commonly used on
automation equipment to sense if the wheel is turning. Alternately I
could use a reed relay as suggested earlier.

I really can't think of anything else to add.

---
OK, if that's what you need, then the first schematic I posted for you
won't work, since it's a missing pulse detector _and_ a three minute
timer, which is what you said you wanted at the time.

What you want now is a missing pulse detector which will enable a
timer with six separate momentary outputs which will go true
sequentially at intervals of 30 seconds, the first output to go true
30 seconds after the timer is enabled and the last output to activate
a relay for some unspecified amount of time as well as provide the
momentary output, right?

Also, if the missing pulse detector detects that pedaling stopped it
will reset everything to the beginning and the sequence will start up
again once pedaling resumes.

One thing you could do which would be helpful would be to find out
whether the playback switch on the recorder provides a simple closure
to ground or to Vcc, or whether it needs to be an isolated contact
closure. That way, if it needs to be isolated it'll probably need to
be a relay. Otherwise, it can be just a transistor.

--
John Fields
Professional Circuit Designer

 

[Tristar500]

Yes, John, this is correct. I sent you an email with a small photo of
the voice recorder circuit board.

It appears the playback button pulls a trace to ground when the button
is pushed.

Thanks again,

Lawrence

 

[John Fields]

On 27 Jul 2005 11:28:19 -0700, "Tristar500"
<robot@innovationrobotics.com> wrote:

Yes, John, this is correct. I sent you an email with a small photo of
the voice recorder circuit board.

It appears the playback button pulls a trace to ground when the button
is pushed.

---
OK.

I got your email, but it was empty.

I've posted a schematic of a circuit which should do what you want,
and a fairly loose circuit description of it to abse and, since you
don't have access to binaries through google, emailed it to you as
well.

If you have any questions it would be nice if you posed them to the
group so that we can all follow your progress.

--
John Fields
Professional Circuit Designer