Need circuit to produce 3 modes of audible confirmation

[Jason S]

I need to make a circuit that will produce either one, two or three short
consecutive beeps depending on the input signal it recieves.
For example:

Input 1 = One beep (could mean 'ON').
Input 2 = Two beeps (could mean 'OFF').
Input 3 = Three beeps (could mean 'ERROR')..

This is just an example. The 3rd one is not critical, but would be nice.

I want to be able to add it to practically any circuit which will provide
audible confirmation of the circuit's current operation. I was thinking of
using a PIC to do this, but is last resort (as I know absolutely nothing
about them).

Has anyone got any suggestions?

Thanks.

[petrus bitbyter]

"Jason S" <jst...@iprimus.com.au> schreef in bericht
news:43997...@news.iprimus.com.au...

A PIC, even the smallest one is the most simple solution... once you can
program it and program it. That's to say you need the skills and the
equipment. If you like to make more of this type of circuits the investment
wil sure pay off. Otherwise, a bunch of 555s may do the job as well.

- One astable (1) for the beep timing (let's say 0.4s on and 0.4s off once
enabled)
- One monostable (1). Triggered by 'ON', enabling the astable (1) for 0.6s.
- One monostable (2). Triggered by 'OFF', enabling the astable (1) for 1.4s.
- One monostable (3). Triggered by 'ERROR', enabling the astable (1) for
2.2s.
- One astable (2) optional to produce a 1kHz beebfrequency once enabled by
astable (1).

The most straightforward way to control astable(1) by the three monostables
is a 3-input OR-gate (or NAND).

petrus bitbyter

[petrus bitbyter]

"petrus bitbyter" <pieterkral...@enditookhccnet.nl> schreef in
bericht news:4399c38d$0$20780$e4fe...@dreader25.news.xs4all.nl...

Oops, forget the NAND as you have no inverted outputs.

petrus bitbyter

[Rich Grise]

Well, a PIC would certainly keep the chip count down! But, you have the
investment of maybe $100.00 for the development system, and the time it
takes to learn the programming language, which, if you're a programmer,
shouldn't be hard at all. And the act of selecting which PIC you want
to use is an exercise in itself. :-)

The only other way that springs to my mind is a pot-load of counters
and logic gates and stuff, which sounds like a dandy exercise for the
reader. ;-)

Good Luck!
Rich

[petrus bitbyter]

"Rich Grise" <rich...@example.net> schreef in bericht
news:pan.2005.12.09....@example.net...

Well, Rich,

I have to disagree on some points. As for a development system, you can find
all on the net. For free. A programmer (the equipment, not the man) can be
build for much less then $100. Free schematics and plans on the net. The
investment will be in time. Knowing nothing of PICS is not a real problem
but knowing nothing of micros and programming is a pretty high hurdle.

Another way I pointed out in a previous posting. If you use NE556 you can do
with two or three chips (and a pot of resistors and caps :)

petrus bitbyter

[Rich Grise]

On Fri, 09 Dec 2005 23:01:46 +0100, petrus bitbyter wrote:
> "Rich Grise" <rich...@example.net> schreef in bericht

>> On Fri, 09 Dec 2005 23:05:05 +1100, Jason S wrote:
>>
>>> I need to make a circuit that will produce either one, two or three
>>> short consecutive beeps depending on the input signal it recieves. For
>>> example:
>>>
>>> Input 1 = One beep (could mean 'ON'). Input 2 = Two beeps (could
>>> mean 'OFF'). Input 3 = Three beeps (could mean 'ERROR')..
>>>
>>> This is just an example. The 3rd one is not critical, but would be
>>> nice.
>>>
>>> I want to be able to add it to practically any circuit which will
>>> provide audible confirmation of the circuit's current operation. I
>>> was thinking of
>>> using a PIC to do this, but is last resort (as I know absolutely
>>> nothing about them).
>>>
>>> Has anyone got any suggestions?
>>
>> Well, a PIC would certainly keep the chip count down! But, you have the
>> investment of maybe $100.00 for the development system, and the time it
>> takes to learn the programming language, which, if you're a programmer,
>> shouldn't be hard at all. And the act of selecting which PIC you want
>> to use is an exercise in itself. :-)
>>
>> The only other way that springs to my mind is a pot-load of counters
>> and logic gates and stuff, which sounds like a dandy exercise for the
>> reader. ;-)
>

> I have to disagree on some points. As for a development system, you can
> find all on the net. For free. A programmer (the equipment, not the man)
> can be build for much less then $100. Free schematics and plans on the
> net. The investment will be in time. Knowing nothing of PICS is not a
> real problem but knowing nothing of micros and programming is a pretty
> high hurdle.

OK, fair enough. I haven't personally attacked that hurdle, because I
haven't really had that much incentive to chase after it. :-)

> Another way I pointed out in a previous posting. If you use NE556 you
> can do with two or three chips (and a pot of resistors and caps :)

This is true - I made an SOS light once with about three chips - it was
very much like those amber flashers that you see on the "Under
Construction" signs, but this flashed "SOS" in Morse code

_ _ _ ___ ___ ___ _ _ _ _
___| |_| |_| |_| |_| |_| |_| |_| |_| |_____________| |_ ... etc.

I used "data selector logic" lifted almost directly from Don Lancaster's
TTL Cookbook.

The point of the project was, if you see flashing amber lights, on-off-on-
off, etc, then it means mechanical difficulties. If you see the SOS light,
it means a person in distress, so hurry up and call the ambulance. The
product never got off the ground, but it was fun to design!

And I got paid. :-)

Cheers!
Rich

[Ken Smith]

In article <43997...@news.iprimus.com.au>,

Jason S <jst...@iprimus.com.au> wrote:
>I need to make a circuit that will produce either one, two or three short
>consecutive beeps depending on the input signal it recieves.
>For example:
>
> Input 1 = One beep (could mean 'ON').
> Input 2 = Two beeps (could mean 'OFF').
> Input 3 = Three beeps (could mean 'ERROR')..
>
>This is just an example. The 3rd one is not critical, but would be nice.

How about a CMOS 22V10 and a couple of RCs ?

--
--
kens...@rahul.net forging knowledge

[Fred Bloggs]

I get two logic chips, a 555, three diodes, one transistor, two
resistors, the usual 2xR+C 555 astable timing, and a beeper:
View in a fixed-width font such as Courier.

.
. .------------------------------------------------------.
. | |
. | 3x 1N4148 |
. | --------- |
. IN 1----------------------+-|>|--+-[100K]-+----|RST OUT|----+
. | | | | | | |
. IN 2--------------------+-|-|>|--+ | | 555 | |
. | | | | | | | |
. IN 3--------------------|-|-|>|--+ | | ASTABLE | BEEPER
. | | | | | | | |
. .-|----------------|-|------+ | | ON TIME | |
. | | | | | | | =BEEP | |
. | | | | [100K] | | OFF TIME| ---
. | | 4017 | | | | | 1/2 BEEP| ///
. | | -------- | | | | ---------
. | | | Q0| | | --- \|
. | | |-- | | | /// 2N3904|---.
. | '---|CE | | | <| |
. | | | | | _ 4011 | |
. | '1'-|CLK Q1|--------| \ --- |
. | | | | | | o _ /// |
. | | | | '--|_/ '-| \ |
. | | | | _ | o--[100K]-+
. | | Q2|--------| \ .-|_/ |
. | | | | | o' |
. | | | '----|_/ |
. | | | |
. | _ | Q3|--------------------[100K]-'
. +-| \ | |
. | | o|RST . |
. +-|_/ | |
. | Q9|
. | |
. --------
.
.
.

Any one of the inputs goes high and the 4017 and 555 are taken out of
RESET. The end of each beeper pulse clocks the 4017, outputs of which
are AND'ed with appropriate input line as necessary. When required
number beeps has occurred, transistor RSTs the 555, freezing it in LOW
state, and circuit remains locked up until all inputs are
removed-allowing 4017 to RESET.

[Fred Bloggs]

That's nonsense!- why do that when you can use the serial port on your PC.

[Don Foreman]

On Sat, 10 Dec 2005 01:26:08 GMT, Rich Grise <rich...@example.net>
wrote:

>
>This is true - I made an SOS light once with about three chips - it was
>very much like those amber flashers that you see on the "Under
>Construction" signs, but this flashed "SOS" in Morse code
>
> _ _ _ ___ ___ ___ _ _ _ _
>___| |_| |_| |_| |_| |_| |_| |_| |_| |_____________| |_ ... etc.
>
>I used "data selector logic" lifted almost directly from Don Lancaster's
>TTL Cookbook.
>
>The point of the project was, if you see flashing amber lights, on-off-on-
>off, etc, then it means mechanical difficulties. If you see the SOS light,
>it means a person in distress, so hurry up and call the ambulance. The
>product never got off the ground, but it was fun to design!
>
>And I got paid. :-)
>

I made a flashing bike light for my daughter in NYC using a red 1-watt
Luxeon. She said she wanted bright. This sucker is bright!

I thought about having it flash: di di di di dah ..... dah dah di dah
(4Q) to show a little New Yawk attitude -- but I didn't have the
microcontroller ability to do that. Still don't but hope to soon.
Heck, a SOT23-6 10Fxxx PIC would have taken less space than the 555
I used.

[Jason S]

"Fred Bloggs" <nos...@nospam.com> wrote in message
news:439A4968...@nospam.com...

Hi Fred, thanks for the schematic - very helpful. Did you come up with that
yourself? Just out of curiosity, have you tested the circuit? I haven't
had a chance to build it yet, but can't wait to see how it works - and
sounds like it will do the job. Most others have mentioned a PIC but I
haven't got the time for that. I will let you know how I went with your
schematic. Thanks again.

Jase.

[Jason S]

Thanks to everyone who has made suggestions.

Jase.

[Fred Bloggs]

No need to breadboard, these parts are too familiar. I did a
confirmation check in mixed-mode SPICE and it works as expected. What
kind of beeper are you working with?- there are several types- some are
self-oscillating so that all the 555 has to do is apply voltage to it,
others are just transducers and you have to supply the oscillatory
input. In the second case you will have to make a 555 oscillator at
2-3KHz to drive it, and use the 555 on the schematic to drive its RST
input- gating it on and off. I suppose a beep duration of 500ms ON and
250ms OFF should be about right- and that would be something like this:

View in a fixed-width font such as Courier.

.
.
. TO <---------------------.
. 4017 |
. /CE |
. 5V |
. | |
. .-----+--[270K]---+ |
. | | | |
. | [270K] +------------+ |
. | | | VCC | |
. | | | | |
. FROM >----|-----|----|RST OUT|-+-> TO BEEPER
. LOGIC | | | | .
. | +----|THRESH | |
. | | | | |
. | +----|TRIG DIS|-. |
. | + | | | | |
. | === | GND | | |
. | 1U| +------------+ | |
. | | | | |
. | '-----------+ | |
. | | | |
. | GND | |
. | | |
. '-------------------------' |
. |
. |
. |
. .---------------------------------'
. |
. |
. |
. | 5V
. | |
. .----|---+--[27K]----+
. | | | |
. | | [270K] +------------+
. | | | | VCC |
. | | | | |
. | '---|----|RST OUT|--------.
. | | | | |
. | +----|THRESH | |
. | | | | |
. | +----|TRIG DIS|-. PIEZO
. | | | | | |
. | === | GND | | |
. | 1n| +------------+ | |
. | | | | |
. | '-----------+-------|------'
. | | |
. | GND |
. | |
. '----------------------------
.
.

[Ken Smith]

In article <439AEEB8...@nospam.com>,
Fred Bloggs <nos...@nospam.com> wrote:

In my never ending battle to save parts I suggest:

Modified version:
>
>.
>.
>. TO <-+-------------------.

>. 4017 | |
>. /CE | |
>. | 5V |
>. | | |
>. | | |

>. | | |
>. [270K] +------------+ |
>. | | VCC | |
>. | | | |

>. FROM >----------|----|RST OUT|-+-> TO BEEPER

>. LOGIC | | |
>. +----|THRESH |
>. | | |

>. +----|TRIG DIS|--------

>. + | | | |
>. === | GND | |
>. 1U| +------------+ |
>. | | |
>. '-----------+ |
>. | |
>. GND |
>. |
>. |

>. |
>. |
>. |
>. .---------------------------------'
>. |

>. | --------------------------

>. | | |
>. | | 5V |
>. | | | |
>. | | | |

>. | | | |
>. | [270K] +------------+ |
>. | | | VCC | |
>. | | | | |

>. '---+ |RST OUT|--------+

>. | | | |
>. +----|THRESH | |
>. | | | |

> . +----|TRIG DIS|- PIEZO

> . | | | |
> . === | GND | |
> . 1n| +------------+ |
> . | | |

> . '-----------+--------------'

> . |
> . GND
> .
> .

> .
>
BTW: I'd use a ceramic for the 1uF. There is a lot less risk of putting
it in backwards that way.

[Fred Bloggs]

And here I opened your post on the outside chance you would tell me
something new. I don't really care for the 50-50 beep alert as it lacks
sufficient enthusiasm in my estimation, so I will splurge and use the
additional resistor thank you very much. But here's something novel- a
diode used to practically eliminate the extra long first pulse effect.
SPICE says it works, and I don't see why the real world should disagree
as long as the output is effectively unloaded. The price you pay is loss
of the ratiometric switching thresholds but that is not required in the
majority of applications. You can do whatever you want with the piezo
astable- it's neither here nor there.

View in a fixed-width font such as Courier.

.
.
. TO <-----------------------------.

. 4017 |
. /CE |
. 5V |
. | |

. .---[330K]--+ |
. | | |
. +-----------|---------. |
. | | | |
. [120K] +------------+ | |
. | | VCC | | |
. | | | | |
. FROM >----------|RST OUT|---------+
. LOGIC | | | | |
. CKT +----|THRESH | | |
. | | | | |
. +----|TRIG DIS|---' |
. | | | |
. | | | 1N4148 |
. | | CONT|--|>|----+
. | | | |

. + | | | |
. === | GND | |
. 1U| +------------+ |
. | | |
. '-----------+ |
. | |
. GND |

. \|/
.
. to RST of
. piezo astable
.
.

[Ken Smith]

In article <439B9908...@nospam.com>,
Fred Bloggs <nos...@nospam.com> wrote:

This will work too:

>
>.
>.
>. TO <-----------------------------.
>. 4017 |
>. /CE |
>. 5V |
>. | |
>. .---[330K]--+ |
>. | | |
>. +-----------|---------. |
>. | | | |
>. [120K] +------------+ | |
>. | | VCC | | |
>. | | | | |

>. FROM >+---------|RST OUT|---------+
>. LOGIC | | | | | |
>. CKT --- +----|THRESH | | |
>. --- | | | | |
>. 0.5 | +----|TRIG DIS|---' |
>. ----+ | | |
>. | | | |
>. | | CONT| |

>. | | | |
>. + | | | |
>. === | GND | |
>. 1U| +------------+ |
>. | | |
>. '-----------+ |
>. | |
>. GND |
>. \|/
>.
>. to RST of
>. piezo astable

When RST goes high, the 0.5u puts the trigger at Vcc/3.

BTW: You can make pulsing beep with just one 555. Unfortunately, there
is no easy way to do the CD4017 running from it. It looks like this:

100K 10K 100K 10K 1u
Vcc -/\/\/-+-/\/\-+-/\/\-+---+--/\/\--!!--+---- Output
! ! ! ! !
! 10u --- 0.1--- ! !
! --- --- ! !
! ! ! +- !
! GND GND ! ! !
! ! ! !
--------------------------------------
! D T T Q !
! i r h !
! s g r !

[Jason S]

Hi Fred,

To answer your question, I will be using a self-driven piezo buzzer which
DOES NOT oscillate (I thought that was what I was supposed to use?), so the
555 will apply power to it when a beep is required (via the usual
resistor/transistor/diode arrangement).

I did a breadboard test of your original schematic this morning, and it
seems to be operating PARTIALLY o.k - Not sure if I did anything wrong -
there's wires everywhere! arrghhh!. There's bit of a problem that I have
encountered....

Problem # 1 :
The 555's outpin pin is high if there are no inputs connected (standby) -
isn't it supposed to be low? As soon as I connect an output, it works o.k
but as soon as the input is disconnected (when the 4017 resets), the 555's
output pin goes high again and stays high (causing the buzzer to sound
continuously!). Not sure what to do here Fred~! I used a logic probe and
the NAND gate that connects to the RST pin of the 4017 intermittently gives
half high half low, and found that the circuit then doesn't 'reset'. Funny
how it's only sometimes.

Problem # 2:
Also if there is already an input connected BEFORE I turn the power on to
the circuit, buzzer will beep 7, 8 or 9 times before it stops!
i.e. Input 1 = 7 beeps. Input 2 = 8 beeps. Input 3 = 9 beeps.
However, the 4017 seems to be going through only 1 cycle (which is good),
because according to the logic probe, each of the outputs go 'high' only
once. Weird.

Let me confirm the following:
1) All gates used are of NAND type (4011). Note that your text mentioned
'AND'd'!).
2) The 4017's CLK/Clock pin is connected to the positive rail at all times.
???
3) Output from the 555 is connected to the CE/Clock Enable pin of the 4017.
4) The inputs - finding that circuit doesn't work properly if any unused
input is 'floating' (i.e. not connected to ground). Are the diodes in the
correct spot?
5) What's the deal of not having a square wave for the 555 output? Does it
matter? My output seems pretty square.

Thanks again,

Jason.

"Fred Bloggs" <nos...@nospam.com> wrote in message

news:439AEEB8...@nospam.com...

[Fred Bloggs]

Jason S wrote:
> Hi Fred,
>
> To answer your question, I will be using a self-driven piezo buzzer
> which DOES NOT oscillate (I thought that was what I was supposed to
> use?), so the 555 will apply power to it when a beep is required (via
> the usual resistor/transistor/diode arrangement).
>
> I did a breadboard test of your original schematic this morning, and
> it seems to be operating PARTIALLY o.k - Not sure if I did anything
> wrong - there's wires everywhere! arrghhh!. There's bit of a problem
> that I have encountered....
>
> Problem # 1 : The 555's outpin pin is high if there are no inputs
> connected (standby) - isn't it supposed to be low? As soon as I
> connect an output, it works o.k but as soon as the input is
> disconnected (when the 4017 resets), the 555's output pin goes high
> again and stays high (causing the buzzer to sound continuously!).
> Not sure what to do here Fred~! I used a logic probe and the NAND
> gate that connects to the RST pin of the 4017 intermittently gives
> half high half low, and found that the circuit then doesn't 'reset'.
> Funny how it's only sometimes.

You are using the bipolar 555 and the typical RST pin current is listed
as 0.1mA, for which the 100K resistors are much too high. If you want to
keep the signal input impedance of the circuit high then you will have
to use the circuit attached below or go to a CMOS 555.

>
> Problem # 2: Also if there is already an input connected BEFORE I
> turn the power on to the circuit, buzzer will beep 7, 8 or 9 times
> before it stops! i.e. Input 1 = 7 beeps. Input 2 = 8 beeps. Input 3
> = 9 beeps. However, the 4017 seems to be going through only 1 cycle
> (which is good), because according to the logic probe, each of the
> outputs go 'high' only once. Weird.

If live inputs while the circuit is unpowered are a real possibility
then you should put a 10K in series with the input of the NAND driving
the 4017 RST input, and you can add a C+R from VCC to one input of that
NAND for a power on RST if you don't like the beeps.

>
> Let me confirm the following: 1) All gates used are of NAND type
> (4011). Note that your text mentioned 'AND'd'!).

Yep-

> 2) The 4017's CLK/Clock pin is connected to the positive rail at all
> times. ???

Yep-

> 3) Output from the 555 is connected to the CE/Clock Enable pin of
> the 4017.

Yep-

> 4) The inputs - finding that circuit doesn't work properly if any
> unused input is 'floating' (i.e. not connected to ground). Are the
> diodes in the correct spot?

Either go with a CMOS 555 or use circuit below.

> 5) What's the deal of not having a square wave for the 555 output?
> Does it matter? My output seems pretty square.

It's only personal preference- if you're satisfied with the sound then
go with it.

As I said- either go with CMOS 555 and original circuit, or keep your
bipolar 555 and mod the circuit like shown below. You add a pull-up
resistor from 555 RST to Vcc, the NAND driver for the 4017 RST also
drives the transistor base OR circuit through a 10K and a 10K replaces
the 100K in the other drivers of the base circuit. A power on RC has
been added to RST the 4017 at turn-on. The operation is basically the
same except that now when an IN goes high, it indirectly causes 555 RST
high through the NAND+ transistor combination, instead of directly. The
555 is now RST after required number of beeps OR when the inputs are
removed.

View in a fixed-width font such as Courier.

.
. 5V
. 1U |
. GND--||--+----[100K]-------+
. | |
. .------------------------------------------------------.
. | | _ | |
. | 3x 1N4148 '---| \ [10K] |
. | | o-. | --------- |
. IN 1------[10K]--+-|>|--+---|_/ | +----|RST OUT|----+
. | | | | | | | |
. IN 2------[10K]+-|-|>|--+ | | | 555 | |
. | | | | | | | | |
. IN 3-----------|-|-|>|--+ | | | ASTABLE | BEEPER
. | | | | | | | | |
. | | | [100K] | | | ON TIME | |
. | | | | | | | =BEEP | |
. | | | GND | | | OFF TIME| GND
. | | | | | | 1/2 BEEP|
. | | | | | ---------
. | | | | |
. | | | | |
. | | '--------. | |
. | '--------. | | |
. | | | | |
. .---------------------------+-[10K]--------.
. | | | | | |
. | | --4017-- | | | |
. | | | Q0| | | \| |
. | | |-- | | | 2N3904|---+

. | '---|CE | | | <| |
. | | | | | _ 4011 | |

. | '1'-|CLK Q1|--------| \ GND |
. | | | | | | o _ |
. | | | | '--|_/ '-| \ |
. | | | | _ | o--[10K]--+

. | | Q2|--------| \ .-|_/ |
. | | | | | o' |
. | | | '----|_/ |
. | | | |

. | | Q3|--------------------[10K]--'
. | | |
. '-----|RST . |
. | |

[Fred Bloggs]

That would be 10K in series with each IN line:

View in a fixed-width font such as Courier.

.
. 5V
. 1U |
. GND--||--+----[100K]-------+
. | |
. .------------------------------------------------------.
. | | _ | |
. | 3x 1N4148 '---| \ [10K] |
. | | o-. | --------- |

. IN 1-----[10K]---+-|>|--+---|_/ | +----|RST OUT|----+
. | | | | | | | |
. IN 2-----[10K]-+-|-|>|--+ | | | 555 | |
. | | | | | | | | |
. IN 3-----[10K]-|-|-|>|--+ | | | ASTABLE | BEEPER

[Fred Bloggs]

Jason S wrote:
> 4) The inputs - finding that circuit doesn't work properly if any unused
> input is 'floating' (i.e. not connected to ground). Are the diodes in the
> correct spot?

You can't let the inputs 1 & 2 "float" since these connect to the logic
NANDs on the output of counter, the logic requires well-defined voltage
levels , either H or L, voltages in between cause them to act up. You
have to put pulldowns to GND there- say 1M ohms- put 1M's right at the
input to the circuit on all IN's for consistency. Where are your inputs
coming from anyway? The diodes are a logic OR and they are in the
correct spot.

[Ken Moffett]

Rich Grise <rich...@example.net> wrote in
news:pan.2005.12.09....@example.net:

Wrong...wrong...wrong! Google "PicAxe". Free PBasic programming software,
program loads from a serial port directly (through two resistors) to the
chip, and 8 pin chips are cheap (even with shipping from GB). They were
intended for teaching uP's in elementary and high schools, but are great
for real world projects. I have several controlling science displays. These
are "PIC's" with an onboard PBasic interpreters. The 8-pin, 4MHz
PicAxe08/08M will easily do what you described and more.

[Jason S]

yup, got it =)

Jason.

"Fred Bloggs" <nos...@nospam.com> wrote in message

news:439C4D16...@nospam.com...

[Jason S]

"Fred Bloggs" <nos...@nospam.com> wrote in message

news:439C605E...@nospam.com...

Ok, yeah I thought the case. I'll ammend my printed copy.
The inputs may or may not be coming from other logic gates, so it's best I
connect the 1M drop-downs I think.

Jason.

[Jason S]

Hi Fred,

Thanks so much for the assistance with this.

Your modification below actually worked a lot better than the original (I
decided to go with the modified schematic because I didn't have a CMOS 555
handy). It's working more normally now =) - I am wrapped!
But there's just some very minor issues now that I hope you could sort out
for me...

1. The first beep appears to be longer than the second and third (I know
you mentioned something earlier about a diode, but I didn't quite
understand). I take it that's because the 555 has just come out of a reset?
So as a result, if I connect, say, the 3rd INPUT for three beeps, I hear:
beeeep! beep! beep!

2. With the 10K resistors connected in series and also the 1M drop-downs
for each input like you said, once I disconnect an input (after beeps have
finished), I would hear a very brief 'Blip!' from the buzzer. I actually
thought maybe the 1uF cap had something to do with it, but still does it
even if I remove it.

3. Oh, at power on with an input already connected, it is still beeping 6-9
times. > Not a big issue however. Mainly concerned with the above two.

Kind regards,

Jason.

"Fred Bloggs" <nos...@nospam.com> wrote in message

news:439C4BEC...@nospam.com...

[Jason S]

"Jason S" <jst...@iprimus.com.au> wrote in message
news:439d6...@news.iprimus.com.au...

>
> 3. Oh, at power on with an input already connected, it is still beeping
> 6-9 times. > Not a big issue however. Mainly concerned with the above
> two.
>

Actually forget number 3 (above). It's working fine at startup now.

[Fred Bloggs]

Jason S wrote:
> Hi Fred,
>

> Thanks so much for the assistance with this.
>
> Your modification below actually worked a lot better than the original (I
> decided to go with the modified schematic because I didn't have a CMOS 555
> handy). It's working more normally now =) - I am wrapped!
> But there's just some very minor issues now that I hope you could sort out
> for me...
>
> 1. The first beep appears to be longer than the second and third (I know
> you mentioned something earlier about a diode, but I didn't quite
> understand). I take it that's because the 555 has just come out of a reset?
> So as a result, if I connect, say, the 3rd INPUT for three beeps, I hear:
> beeeep! beep! beep!

Right- the 555 pin 5 ,CONT, is the first node of an internal R-R-R
voltage divider to GND which is the threshold for the THRESH switch
point, and 1/2 the CONT voltage is the threshold for the TRIG switch
point. By connecting a small diode from CONT to OUT with anode at CONT,
when OUT is high, the diode is back biased and does not affect the trip
point for THRESH, but when OUT is low, which means the timing capacitor
is being discharged downwards towards the TRIG threshold, the diode
pulls the internal CONT node down to 0.7V making the new TRIG threshold
~0.35V. In your application, when you remove RST, the first beep lasts
for the time it takes the timing capacitor to charge from 0 to THRESH
threshold of 2 x Vcc/3, then goes off for the time it takes to discharge
to 0.35V, then back on for the time it takes to re-charge from 0.35V to
THRESH threshold, then off for time to discharge back 0.35V, etc...All
the ON times are about the same for practical purposes this way. Without
the diode the second and third on times would be of duration the time it
takes to charge from 1/3 Vcc to 2/3 Vcc- and they sound noticeably
shorter than the first beep.

>
> 2. With the 10K resistors connected in series and also the 1M drop-downs
> for each input like you said, once I disconnect an input (after beeps have
> finished), I would hear a very brief 'Blip!' from the buzzer. I actually
> thought maybe the 1uF cap had something to do with it, but still does it
> even if I remove it.

When a last input is removed, the input NAND gate RSTs the 4017 and
drives the base of the 2N3904 all at the same time. The RST of the 4017
removes its base drive from the transistor tending to turn it off while
the NAND is applying its drive turning it ON. The NAND base drive part
should be solidly on several hundred nanoseconds before the 4017 outputs
begin to remove their base drive so that I am at a loss to explain why
you are hearing a blip. If you are manually removing a clip lead or
using a switch, which has bounce at turn-off, then an intermittent
disconnect-connect-disconnect etc jittering of the input contact,
however brief, may be causing the blip you are hearing. You can
eliminate the effect, for reasonable jitter of input connection times,
by adding a filter capacitor from collector to base of the 2N3904. This
will make the transistor have a fast on- slow-off characteristic and
help prevent the blip. A schematic with all the changes:

View in a fixed-width font such as Courier.

.
. 5V
. 1U |

. GND----||--+----[100K]----+---------.
. | | |
. [10K] 10K] |
. | | |
. .-----------------------------------------------------.

. | | _ | | |
. | 3x 1N4148 '---| \ | | |

. | | o-. | --------- |
.IN1----+----[10K]---+-|>|--+---|_/ | +----|RST OUT|--+---+
. | | | | | | | | | |
.IN2--+------[10K]-+-|-|>|--+ | | | 555 | | |
. | | | | | | | | | OSC | | |
.IN3+--------[10K]-|-|-|>|--+ | | | | | BEEPER
. | | | | | | | | | | | | |
. 1M | | | | | [100K] | | | | | |
. | | | | | | | | | .-|CONT | | |
. |1M | | | | GND | | | | | | GND
. | | | | | | | | | | | |
. | |1M | | | | | | --------- |
. | | | | | | | | | | |
. | | | | | | | | | gnd |
. '-+-' | | '--------. | | | |
. | | '--------. | | | '-----|>|------'
. GND | | | | |
. .---------------------------' | +
. | | | | +----||--.
. | | --4017-- | | | 1U |
. | | | Q0| | | \| |
. | | |-- | | | 2N3904 |------+

. | '---|CE | | | <| |
. | | | | | _ 4011 | |
. | '1'-|CLK Q1|--------| \ GND |
. | | | | | | o _ |
. | | | | '--|_/ '-| \ |
. | | | | _ | o--[10K]--+
. | | Q2|--------| \ .-|_/ |
. | | | | | o' |
. | | | '----|_/ |
. | | | |

. | | Q3|--------------------[10K]--+
. | | | |
. +-----|RST . | |
. | | | |
. | | Q9| |
. | | | .-----------[10K]--'
. | -------- |
. | |
. '-----------------------'
.
.
.

[Jason S]

Fred,

It works! Hooray! Thank you so so much. You have been great.

The diode worked very well to fix the longer beep problem -thank you-, but
as with the cap to fix the 'blips' - it failed miserably... actually made it
worse. So I left it as it was, but connected a transistor to each input and
works a lot better now.

Appreciate your help.

Regards,

Jason.

"Fred Bloggs" <nos...@nospam.com> wrote in message

news:439D86BB...@nospam.com...