Stretch 0.5sec pulse to 6sec

[amdx]

I have a 4 volt pulse that is 0.5 sec long.
I need a 6 sec pulse that will switch a 2n3904.
The 4 volts has about a 400 ohm source impedance.

I could use a 555, but first I need to invert the signal then
drive my 2n3904 switch, so I have 2 transistors and a 555. Oh yes I
could probably differentiate the 0.5 sec pulse to trigger the 555, so
maybe only 1 transistor and a 555.

Could someone help with a pulse stretcher circuit.
Maybe an RC, transistor, or, I have some FETs, J301 or 2N3819, probably
a diode. That will drive my final 2N3904 switch.
The 2N3904 switch is across the alert button on a Walkie Talkie and
shorts about 3.5v to ground.


The longer story;
This started with "Have 5v regulator need 9 volts"
That particular problem is solved. everything has good clean power.

I have a Harbor Freight *Driveway sensor.
The sensor triggers a receiver with RF.
The receiver in turn triggers the alert button on a **Walkie Talkie.
After a temporary install, I noted the Talkie alert signaled 10 times.
That is to much, three times is fine.
After trying several things to eliminate the 10 alerts, I finally
wrapped the walkie talkie in aluminum foil. Problem solved. I'll deal
with this later by putting the Harbor Freight receiver inside my boat
and the walkie talkie outside where it's RF can get out. (it's an all
Aluminum boat)
The 0.5 sec pulse causes one alert signal. A six second pulse causes
three alert signals. Perfect!


*
http://www.harborfreight.com/wireless-driveway-alert-system-93068.html

**
http://www.motorola.com/Business/US...umers/Talkabout+FV300+Two-Way+Radio_B2B_US_EN

Mikek

 

[amdx]

I presume, then, also positive-going?

You presume correctly.
But you already new that. ;-)
Mikek

 

[amdx]

I have a 4 volt pulse that is 0.5 sec long.
I need a 6 sec pulse that will switch a 2n3904.
The 4 volts has about a 400 ohm source impedance.

I could use a 555, but first I need to invert the signal then
drive my 2n3904 switch, so I have 2 transistors and a 555. Oh yes I
could probably differentiate the 0.5 sec pulse to trigger the 555, so
maybe only 1 transistor and a 555.

Could someone help with a pulse stretcher circuit.
Maybe an RC, transistor, or, I have some FETs, J301 or 2N3819, probably
a diode. That will drive my final 2N3904 switch.
The 2N3904 switch is across the alert button on a Walkie Talkie and
shorts about 3.5v to ground.

I have breadboarded the two transistor, 555 circuit.
Three alert signals as wanted. I can wire it on a pcb
in the morning, if no one comes up with a simple RC one transistor
pulse stretcher. I couldn't make it work, maybe it can't be done.
Seems the timing would be difficult, but for my circuit 20% fluctuation
would be ok.
Mikek

 

[Jasen Betts]

I have a 4 volt pulse that is 0.5 sec long.
I need a 6 sec pulse that will switch a 2n3904.
The 4 volts has about a 400 ohm source impedance.

I could use a 555, but first I need to invert the signal then
drive my 2n3904 switch, so I have 2 transistors and a 555. Oh yes I
could probably differentiate the 0.5 sec pulse to trigger the 555, so
maybe only 1 transistor and a 555.

Could someone help with a pulse stretcher circuit.
Maybe an RC, transistor, or, I have some FETs, J301 or 2N3819, probably
a diode. That will drive my final 2N3904 switch.
The 2N3904 switch is across the alert button on a Walkie Talkie and
shorts about 3.5v to ground.

you can't just use the discharge pin of the 555 instead of the collector of the
2n2904? is 200mA not enough?

 

[Jamie]

I have a 4 volt pulse that is 0.5 sec long.
I need a 6 sec pulse that will switch a 2n3904.
The 4 volts has about a 400 ohm source impedance.

I could use a 555, but first I need to invert the signal then
drive my 2n3904 switch, so I have 2 transistors and a 555. Oh yes I
could probably differentiate the 0.5 sec pulse to trigger the 555, so
maybe only 1 transistor and a 555.

Could someone help with a pulse stretcher circuit.
Maybe an RC, transistor, or, I have some FETs, J301 or 2N3819, probably
a diode. That will drive my final 2N3904 switch.
The 2N3904 switch is across the alert button on a Walkie Talkie and
shorts about 3.5v to ground.


The longer story;
This started with "Have 5v regulator need 9 volts"
That particular problem is solved. everything has good clean power.

I have a Harbor Freight *Driveway sensor.
The sensor triggers a receiver with RF.
The receiver in turn triggers the alert button on a **Walkie Talkie.
After a temporary install, I noted the Talkie alert signaled 10 times.
That is to much, three times is fine.
After trying several things to eliminate the 10 alerts, I finally
wrapped the walkie talkie in aluminum foil. Problem solved. I'll deal
with this later by putting the Harbor Freight receiver inside my boat
and the walkie talkie outside where it's RF can get out. (it's an all
Aluminum boat)
The 0.5 sec pulse causes one alert signal. A six second pulse causes
three alert signals. Perfect!


*
http://www.harborfreight.com/wireless-driveway-alert-system-93068.html

**
http://www.motorola.com/Business/US...umers/Talkabout+FV300+Two-Way+Radio_B2B_US_EN


Mikek

|
===
GND
.-.
| |1MEG PTT load (Radio)
| | |
0.5s Pulse 1k '-' ||-+
___ | ||<-
-|___|-->|+-------+-+-----------+-||-+
+ |
1N914 + |
--- 10uf ===
--- GND
+
| Power NMOS or what ever
|
===
GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

Depending on your Vgs(th) spec you use for the NMOS will determine
the time off delay, which is what I think you're after.

This is one should be ~ 5 to 6 secs.

I didn't do a sim on this, just used a calculator but it should be
close.
THe pulse should pull the output down fast and the off delay
will give you a slight ramp due to the way fets are biased and behave
when the drain starts to get higher in voltage than the gate. This
should work out.

The diode is there to prevent back drain from the pulse source.

Jamie

 

[amdx]

you can't just use the discharge pin of the 555 instead of the collector of the
2n2904? is 200mA not enough?

Do you mean the discharge (pin 2) or output (pin 3).
The pin from the walkie talkie already has a voltage on
it, pin 3 will source current to the walkie talkie before
it sinks current. I need an open collector to pull the Walkie talkie
low. That's my understanding for now.
Mikek

 

[Jamie]

Do you mean the discharge (pin 2) or output (pin 3).
The pin from the walkie talkie already has a voltage on
it, pin 3 will source current to the walkie talkie before
it sinks current. I need an open collector to pull the Walkie talkie
low. That's my understanding for now.
Mikek

if you drive the trigger input via the output of the timer through a
R of lets say 100k for example. Connect a 50uf cap from the trigger
input pin to common and then use a diode in parallel with the 100k R
to quickly recharge the cap when the output is high. The anode of the
diode will be on the output while the cathode is on the trigger input.

WIth this configuration, you can drive the threshold pin from your 4
volt source for the TX PTT signal. This will force both the output pin
to go low along with the discharge pin also going low. Take your pick
which one you want to use. I would use the discharge pin since it is a
open collector and not yet in use. The output pin will be busy
discharging the cap so the trigger input will reset the output when it
reaches 1/3 of the Vcc voltage.

And that is another issue. Since you only have 4 volts to drive the
threshold pin for the TX signal, your Vcc can not exceed 6 volts abs which
is on the edge, I would use a common 5 volt post reg to be sure for the
vcc.

The diode that is placed there for quick reset can be tailored with
adding a R in series with it if you don't want it respond to another
input pulse too early.. THat is up to you.

THis is just my opinion of course... others most likely have other ideas.

Jamie

 

[amdx]

|
===
GND
.-.
| |1MEG PTT load (Radio)
| | |
0.5s Pulse 1k '-' ||-+
___ | ||<-
-|___|-->|+-------+-+-----------+-||-+
+ |
1N914 + |
--- 10uf ===
--- GND
+
| Power NMOS or what ever
|
===
GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

Depending on your Vgs(th) spec you use for the NMOS will determine
the time off delay, which is what I think you're after.

This is one should be ~ 5 to 6 secs.

I didn't do a sim on this, just used a calculator but it should be
close.
THe pulse should pull the output down fast and the off delay
will give you a slight ramp due to the way fets are biased and behave
when the drain starts to get higher in voltage than the gate. This
should work out.

The diode is there to prevent back drain from the pulse source.

Jamie

Thanks Jamie,
That seems to be working well. I have it breadboarded using 500k and
47uf with an IRFBE30* MOSFET, because that's what I have.
I'll work with it a bit more to get me centered between 2 alerts and 4
alerts. Right now it is alerting 3 times but I don't know if it is
centered between 2 and 4 alerts.
This will fit better in the case.
Thanks again, Mikek

Went back to the bench with a 1 meg pot, Found two alerts at 344k and 4
at 513k, centered at 428k and it consistently alerts 3 times. (47uf)


* IRFBE30 800V 30amp 3 ohm on resistance. Overkill but I have it.

 

[amdx]

74<your choice>123

Nice chip. You choose between positive- and negative-going pulses in and
out.

Jamie's RC, power FET circuit worked good and fit easily, no power
supply needed, only 5 parts.

Bonus, I had all the parts.
I'm in Panama City Fl. The highway doesn't make it here. We are at
the end of the line. Nothing goes past us to get somewhere. I have a
radio Shack and a one man audio repair shop, that sells some parts.
Thanks, Mikek

 

[[email protected]]

Not so great with that kind of delay.

Jamie's RC, power FET circuit worked good and fit easily, no power
supply needed, only 5 parts.

Bonus, I had all the parts.
I'm in Panama City Fl. The highway doesn't make it here. We are at
the end of the line. Nothing goes past us to get somewhere. I have a
radio Shack and a one man audio repair shop, that sells some parts.

I'll bet even DigiKey (via UPS) ships to that end of the line. ;-)

 

[amdx]

Not so great with that kind of delay.



I'll bet even DigiKey (via UPS) ships to that end of the line. ;-)

Sure.
But I want it now!
Mikek

 

[Jasen Betts]

Do you mean the discharge (pin 2) or output (pin 3).
The pin from the walkie talkie already has a voltage on
it, pin 3 will source current to the walkie talkie before
it sinks current. I need an open collector to pull the Walkie talkie
low. That's my understanding for now.
Mikek

pin 2 is trigger one of the main inputs.

Discharge is an open collector output it's on pin 7

 

[[email protected]]

Six seconds. Hmm. Good point. Of course, nothing that's RC-based is
going to really be solid at that delay. For what he wants, though, any
halfway decent approximation to six seconds ought to be OK.

It's way off (three orders of magnitude off) the chart for the HC
types, at least according to TI. This sort of thing is right up the
alley for a 555, though.

 

[[email protected]]

So extend the charts!!

I don't believe what's printed. Drawing lines outside what's printed
is suicidal.

The delay time of a 74HC123 is roughly one RC time constant, you don't
want the resistance to be too low and if it gets too high then you have
to contend with board or capacitor leakage. The delay time of a 555 is
roughly the same, but you don't have the flexibility of input and output.

555s are designed to work with much higher resistors. The '123 really
isn't very good.

Either chip demands that you select a big, low-leakage cap, and a high
resistance, and accept the fact that your leakage is going to vary with
temperature and humidity. Understand that you'll wish that you had done
something with a counter from the git-go, and build your circuit.

I _think_ you could do this more reliably with a CD4060 and a dual J-K
flip-flop, or a 555, a CD4040, and that same J-K flip-flop, but I'd have
to toy with it to be sure and I don't think I care enough to try.

Use a PIC. ;-)

 

[Fred Bartoli]

Tim Wescott a écrit :

So extend the charts!!

The delay time of a 74HC123 is roughly one RC time constant, you don't
want the resistance to be too low and if it gets too high then you have
to contend with board or capacitor leakage. The delay time of a 555 is
roughly the same, but you don't have the flexibility of input and output.

Either chip demands that you select a big, low-leakage cap, and a high
resistance, and accept the fact that your leakage is going to vary with
temperature and humidity. Understand that you'll wish that you had done
something with a counter from the git-go, and build your circuit.

I _think_ you could do this more reliably with a CD4060 and a dual J-K
flip-flop, or a 555, a CD4040, and that same J-K flip-flop, but I'd have
to toy with it to be sure and I don't think I care enough to try.

I just saw a nice CSS555 (available at Jameco)

 

[George Herold]

Tim Wescott a écrit :










I just saw a nice CSS555 (available at Jameco)

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

I stopped buying stuff from jameco after a bad batch of 79L12 voltage
regulators.... cost me too much time.

George H.

 

[amdx]

Jamie's RC, power FET circuit worked good and fit easily, no power
supply needed, only 5 parts.
Bonus, I had all the parts.

I had the 555 circuit breadboarded, it worked.
I'm sure a 74123 could work.
A CD4060 and a dual J-K flip-flop could work.
A pic would work.

But Jamie's circuit worked with 5 parts, I had on hand.

Check with my 6 months from now, if it still works, ya'll can call it
a success.
I'm calling it now.

For Jim,
Alert, Alert, Alert.
That's what it does now.
Alert, Alert, Alert.

Thanks, Mikek

 

[Fred Bartoli]

George Herold a écrit :

I stopped buying stuff from jameco after a bad batch of 79L12 voltage
regulators.... cost me too much time.

I just mentioned jameco because the OP seems to be an hobbyist, but have
a look at the datasheet. It's a pretty interesting son of a ...555

 

[George Herold]

George Herold a écrit :







I just mentioned jameco because the OP seems to be an hobbyist, but have
a look at the datasheet. It's a pretty interesting son of a ...555

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

No problem Fred, I use to buy lots of things from Jameco.
That is an interesting part. low voltage and an internal eeprom
controlled divider.

George H.

 

[Jamie]

I stopped buying stuff from jameco after a bad batch of 79L12 voltage
regulators.... cost me too much time.

George H.

I remember having a few of those too, and the voltages were all over
the place. I don't remember where I got them from but it would appear to
me they may have been seconds that I got from some surplus dealer.

Sometimes you never know.


Jamie