# Triggering 555 Monostable

Discussion in 'Electronic Basics' started by Animesh Maurya, Jan 21, 2004.

1. ### Animesh MauryaGuest

Hi everybody

I know that when a positive going pulse falls to zero, it triggers the
monostable.

It is very obvious that the time period for which the triggering pulse
goes zero should always be less than
time period of positive going pulse at the output (pin 3),

otherwise multivibrator will not return in its stable state after
(1.1* R*C) sec.

Which is causing problem to me. Please don't ask me why is this
happening so.

Till now I was performing simulation, but when I put my hands in
designing it's a problem to me.

I need your valuable suggestions.

Thanks for the time

Animesh Maurya

2. ### Tim DicusGuest

Hi Animesh,

You may want to consider AC coupling the trigger line to the 555. Use a cap
in series with pin 2 (trigger) and a pull-up resistor on the 555 side of the
cap from pin 2 (trigger) to pin 8 (power supply).

You might try a .1uf cap and a 4.7K resistor to start. I don't know what
your output pulse width is, or how fast the trigger source high-to-low
transition rate is, so you may need to play with the values of the cap and
resistor.

Hope that helps,

Tim

3. ### Tim DicusGuest

In addition to my earlier post about AC coupling the trigger input, I
usually include a diode in parallel with the resistor, anode to pin 2
(trigger) and cathode to pin 8 (power supply).

This will prevent the trigger input from overvoltage during the trigger
source low-to-high transition.

Tim

4. ### John PopelishGuest

The trigger input overrides the threshold input, so if the trigger is
held low while the threshold rises, the time-out function of the
threshold input is defeated, till the trigger input voltage goes above
1/3 of the Vcc voltage. If your trigger signal lasts longer than the
timing process of the 555, you will have to shorten its duration by
some means. Common means are to use a differentiator (series cap
between trigger signal and trigger input, with pull up resistor to set
the pulse duration, and diode to Vcc to prevent excessive positive
excursions at end of trigger signal pulse) or to add a non
retriggerable single shot chip before the trigger.

You might make one of those with CMOS gates. For example, you could
apply your long trigger signal to one input of an OR gate and also to
an inverter (or three in series). Then connect the output of the
inverter to the other input of the first OR. The output of the Or
gate will go low when the trigger signal goes low, but remain there
only till the transition propagates through the inverter. Then the Or
gate output will go high. On the low to high transition of the input,
the Or gate output remains high.

5. ### Terry PinnellGuest

John, Tim:

According to Jim Thompson on 5th Jan, that limiting diode (which I'd
been using for as long as I can remember) is redundant. I've therefore
ommitted it subsequently from a couple of projects. Do you agree this
is safe?

From: Jim Thompson <>
Newsgroups: alt.binaries.schematics.electronic
Subject: Re: Need to mod door bell
Message-ID: <>

6. ### Tim DicusGuest

Hi Terry,

The trigger input pin 2 on the LM555 is connected directly to the base of a
PNP transistor with no other connections or devices. See the schematic
diagram here:
http://www.national.com/ds/LM/LM555.pdf

There was a discussion last year (July 2003) on the sci.electronics.design
group about reverse voltage on the base-emitter junction of bipolar
transistors, in particular a 2N3906(PNP).

I agree with the following people. I use the diode, and will continue to do
so.

These are excerpts from that discussion:
------
Jim Thompson said:

It will fail. Devices subjected to this sort of repetitive reverse bias of
the B-E junction gradually lose Beta.

It can take awhile, although high temperatures will accelerate the effect.
For example, under the hood of a car, failure occurs within 1-2 days.
------
Jim Thompson later responded to this question:
I don't know if there's a lower limit where damage doesn't occur. My
experience was in an automotive application at ~1mA.

I just make it a rule *not* to breakdown B-E junctions.

...Jim Thompson
------
Watson A Name said:

I told a guy that he was full of bs when he told us this in the ngs. He said
that putting reverse current thru the E-B junction of a transistor will
degrade the transistor's gain. I didn't believe him, so I tried it with a
few mA, and found that the transistor went from a gain of about 100 before
to a gain of less than 30 after just a half hour of reverse current of a few
mA. I was dumbfounded. It was
really true. Take a transistor and try it if you don't believe me.

If you're going to subject the transistor's E-B to reverse voltage greater
than its max rating, put a diode in series with the base or use a voltage
divider. But by all means, protect it, or else you _will_ find that the
transistor becomes degraded.

7. ### Terry PinnellGuest

Thanks Tim, but did you read the more recent thread I referenced? In
that, I concluded Jim was saying the diode was redundant. Perhaps he
could comment? (I'd email him to be sure he sees this, but don't
appear to have a valid email address.)

8. ### Tim DicusGuest

Hi Terry,

I take the word "redundant" to mean there is already this kind of device or
protection provided internally in the device, and that adding the diode will
duplicate its function. In the case of the LM555 from National
Semiconductor, this is not the case.

You can see in the schematic diagram there is no other device or connection
from pin 2 to the base of the PNP transistor.

I could not find the thread from January. If you can post the relative part
of that message, I would appreciate it. If I am duplicating the protection,
I would certainly like to know!

Thanks,

Tim

9. ### Terry PinnellGuest

OK, here's the thread in its virtual entirety.
-------
From: John Fields <>
Newsgroups: alt.binaries.schematics.electronic
Subject: Need to mod door bell
Date: Mon, 05 Jan 2004 09:25:08 -0600

with attachment DINGDONG.PDF

That showed a neat circuit, but I assumed John had accidentally
ommitted the usual diode. Like mine, his circuits had previously
included it. Hence my follow-up post.
=========================
Minor point: diode across R2 to limit +ve excursions?

--
Terry Pinnell
=========================
Eh?

The Trigger Input is to the base of a lateral PNP and can go above
rail without any harm whatsoever.

....Jim Thompson
=========================
On both standard and CMOS 555 types? In which case I'll cease using
one on edge-triggered 555 monos. Most published circuits I've seen
appear to do so. Typical text like this one: "A small signal diode is
connected across the differentiator to ensure the 555 input signal
never exceeds 9V."

Terry Pinnell
=========================
The CMOS version specifies Vpin(max) as Vs+0.3V, so a silicon diode
isn't going to do squat.

I'd add a series R to the pin to limit current to, say, 1mA, using the
internal ESD diode as the actual catcher.

....Jim Thompson
=========================
Maxim's 7555 data sheet states that it's the CMOS parasitic SCR going
into latchup which limits the pin's positive excursion to 0.3V above
the
positive supply. If that's the case, then I don't see how putting 1mA
through the ESD diode is going to keep the pin below that 0.3V any
more
than an external silicon junction diode would.

As an alternative I'd try this:

V+ V+ V+
| |K |
[R] [CR] [R] +---
| | | |__
| <--O--[C]--+--[<CR]--+-----O|TR
| |
O +---
| 7555
|
GND

--
John Fields
=========================
You're mis-reading the data sheet... they want you to avoid +0.3V
*from a low impedance source* which could cause latch-up.

If you'll look at some logic devices they usually specify a max
current *before* latch-up can occur.

But your method definitely avoids any forward current.

....Jim Thompson
=========================
How about a Ge diode?

Terry Pinnell
=========================
(Thread then wandered OT to brief discussion about sourcing Ge diodes.
As far as I know, my question - which on reflection I'd extend to
Schottky types too - was never explicitly answered.)

10. ### Tim DicusGuest

So what Jim is saying is that I can connect the trigger input pin 2 to my automobile ignition coil (about 400v flyback last time I
measured) and it will do ok?

It is possible that the junction will take a small amount of reverse voltage without harm, but since nobody seems to know where that
limit is, I will use the protection diode. Even though I know it is almost impossible to do, I try to design my circuits to be
"bulletproof".

I have been at this for a while, and I have yet to see a semiconductor junction that does not have limits in both directions.

But there is one thing I know for certain:
I know nothing for certain!

Tim

11. ### Fred BloggsGuest

That is true, but that PNP buffers the input of a differential pair, the
other input of which is connected to 1/3 Vcc. The IC is specified to
Vcc,max=18V making 1/3Vcc=6V. If you have no problem with running TRIG
to Vcc, then this puts 18-2xVeb-6~10.7V reverse voltage across the
series BE junctions on the TRIG side. Allowing for a shorted CONT input
would make this 16.7V of input reverse bias. The transistors probably
take 30V, so unless there is something else going on, the clamp diode is
a waste of space.

12. ### John WoodgateGuest

I read in sci.electronics.design that Tim Dicus <>
What is on the schematic is very often far less than what is actually in
the device. This N-type base and its connecting track must be isolated
from the substrate, I suppose. What isolates them?

13. ### Fred BloggsGuest

Connecting track? You must mean metalization- and this runs on top of a
silicon dioxide layer or similar insulator....

14. ### Tim DicusGuest

Hi John,

Good to have your input. I am aware that most of these are "functional diagrams" rather than true schematics.

But in that light, I would hope that if there were overvoltage protection, that it would be refered to as a "function" by the
manufacturer.

"...isolated from the substrate, I suppose"? Is there a point you are trying to make? Please enlighten me! I want to insure my
designs are reliable without wasting money on parts I don't need.

Thanks,

Tim

15. ### John WoodgateGuest

I read in sci.electronics.design that Tim Dicus <>
Well, in some cases parts are isolated by what is normally a reverse-
biased diode, and that is exactly what you want, isn't it?

16. ### Tim DicusGuest

Hi Fred,

Your argument is very logical. However, I would logically deduce from the Vcc max rating that the transistors are rated at 18 volts,
rather than "probably 30".

I agree the trigger input will take more than Vcc. I know that from experience. But how much more (and for how long) is the question
I have. From experience, I know it will take at least .75 volt above Vcc, at least intermittently.

A 2N3906 shows a base-emitter junction voltage limit at -6 volts (PNP), but shows no "base-emitter reverse voltage" limit. Can one
assume it has no reverse voltage limit? Or would one assume the limit would be +6 volts? Or +40 volts?See the following datsheet:
http://www.semiconductors.philips.com/acrobat/datasheets/2N3906_3.pdf

Once again, my concern here is I want my designs to be reliable without wasting money on parts I don't need.

Thanks,

Tim

17. ### Tim DicusGuest

Yes, sir! That is EXACTLY what I need! If you would be so kind as to point me in the direction of where I can be assured that the
LM555 trigger input pin 2 has that feature.

Thanks,

Tim

18. ### Fred BloggsGuest

The National single supply comparators and opamps using the same process
are spec'ed to 30V and that is where I pulled that number from. For some
reason they give no spec in their Abs Max rating chart for the 555- so
you're right in saying it can't be assumed.

19. ### Tim DicusGuest

Thanks, Fred.

I have seen two other datasheets on the 555, and none have given the trigger input limit in their ratings.

I find that a little odd. Seems to me that is one of the most frequently used externally-driven inputs on that IC.

Tim

20. ### John WoodgateGuest

I read in sci.electronics.design that Tim Dicus <>
I don't know that it does, I just suggest that you MIGHT be lucky in
that a parasitic device that isn't on the functional schematic might do
exactly what you want.

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Continue to site