Page 1 of 3

555 Timer triggered fropm a differentiator.

Posted: Tue Jun 24, 2008 8:06 pm
by frank_fg
:???:

Need your help, it been some time that I troubleshot a circuit.

Hi! I just bought a kit that has two 555 timers that run a servo.

IC2 is 50 Hz oscillator and IC3 is a one-shot.

IC2 pin 3 from the oscillator feeds C4 .001 ufd cap which is connected to
D3 1N914 diod and R4 10k resistor, which from a diferentiator circuit. that feed pin 2 of IC3 the one-shot.

The problem is the diferentiator circuit.

Pin 2 of IC3 is the trigger input which is the diferentiator circuit, is held high
all the time, remove R4 the circuit works find, put it back it don't.
I checked all the components all check out good.
Both IC's good. It look like the diod is not conducting, that is, the capacitor
charges to Vcc and the diod should conduct bring it to 0 volts then start
charging the cap again developing a voltage across R4.

So simple, I can not see it.

Thanks

Frank G.

Posted: Wed Jun 25, 2008 3:40 am
by MrAl
Hi Frank,

Just a guess, maybe the diode is in backwards?
Did you build the circuit yourself?

If you could draw the whole circuit that might help too.

Posted: Wed Jun 25, 2008 9:28 am
by frank_fg
Hi!

Sorry to say that the diode is not in backwards.

All the parts are in correctly, however there is that human factor.

what is throwing me off, is the fact that when I remove one end of the
resistor R4 I can see the capacitor and diode in action.

Tha capacitor is charge to 5v, then the diode is forward biased and start
to conduct, discharging the capacitor and so on. But putting the resistor
back it pull up the signal.

I have no way of drawing a schematic.

visualiz this:
output signal feeds a capacitor which is connected to the aniode side of a diode and a resistor, the other ends are connected to + 5 V.
the diferentiator signal feed the trigger of the one-shot, which is being held high.

It acts as if the resistor were shorted but its not.

I thought that maybe the one-shot trigger pin 2 was not sinking the current, so I replaced it, no change.

Thanks

frank G.

Posted: Wed Jun 25, 2008 10:52 am
by MrAl
Here is a differentiator circuit with resistor to plus 5 volts...

Code: Select all


 

                                 diode
 +5  o--------------------+-------|<|------+
                          |                |
                          |                |
                          |                |
                          R                |
                          |                |
                          |                |
Sig  o-----C--------------+----------------+-------o Sig Out
                          
                          
                          
GND  o---------------------------------------------o




It may however work better if the R is connected to ground (GND)
instead of +5 volts. It depends on the input signal.

Is this the circuit? Or, modify that drawing to be what you really
have there so we can take a look.


Alternate circuit...this might work better...

Code: Select all




 +5  o--------




Sig  o-----C--------------+----------------+-------o Sig Out
                          |                |
                          |                |
                          R      +---|>|---+
                          |      |  diode
                          |      |
GND  o--------------------+------+-----------------o






Posted: Wed Jun 25, 2008 4:04 pm
by frank_fg
Hi! Al

You got it, the only thing is the diode come first, but that works.

Well Al I have worked on some real neat circuits in the past, but that the problem in the past.

What I did was to make my own differenntiator network and substituted
it, and woulden you now it! it worked the same.

I'm at a lost in that the circuit looks good its what you might call textbook
circuit.

It a negative-going edge trigger feeding a comparator pin 2 of the 555.

I still say it something simple I can't see.

Thanks

Frank G.

Posted: Wed Jun 25, 2008 7:20 pm
by ringo47stars
The 555 and the ic connected to it have to get in sinc so you might try a different size diode or putting a capacitor across the positive and negative (large one) or both or positive and somewhere else to dampen the trigger from the ic.

Posted: Thu Jun 26, 2008 2:16 am
by MrAl
Hi again,

Frank, you havent yet drawn the circuit for us to see, so i can only
guess. If you draw the circuit we will solve this in 10 seconds. If
you dont draw the circuit it's going to take 10 posts instead.

Anyway, since you say that it is a negative going edge that triggers
the next stage then you do need the cap in series and the resistor
to pull up, but the diode can not be in series with the cap as you
seem to be suggesting it is now. Take out the diode and short it
(if it's in series with the cap) and everything works. A diode in this
application would be to protect the input of the next stage from
an over or undervoltage. In this case, when the previous stage goes
high the cap discharges completely (or nearly so), then when it goes
low the low appears at the input of the next stage for a short time
period depending on the size of the cap and the size of the resistor
and the input impedance of the next stage. All is well so far, but
now the cap is charged by the resistor in a way that makes the
right side positive (+5v approx), and now when the previous stage
goes back to positive the right side of the cap jumps up to +10v
(a voltage doubler circuit). This could blow out the next stage,
so a diode clamp is used to prevent the input of the next stage
from receiving +10v instead of the lower and safer +5v. To
get this protection the diode is connected with the anode to the
input of the next stage and the cathode to +5v.

Note that a diode in series with a cap does not work very well
anyway, because it depends on the leakage of the diode to operate,
if it operates at all.

The circuit i am recommending for your application would therefore
look like this:

Code: Select all


 

                                 diode
 +5  o--------------------+-------|<|------+
                          |                |
                          |                |
                          |                |
                          R                |
                          |                |
                          |                |
Sig  o-----C--------------+----------------+-------o Sig Out
                         
                         
                         
GND  o---------------------------------------------o





Note you may wish to include a small 100 ohm resistor in series
with the cap to limit the discharge peak current through the diode.
This current might not hurt the diode, but it may cause the +5v
line to rise for a short time period which could bother other parts
of the circuit. With the extra 100 ohm resistor this doesnt happen.


Here are the waveforms and the circuits for negative edge triggering:


Image


From these you can see that this should work very nicely.
If the input of the next stage has significant impedance, you may have
to increase the value of the capacitor.

Posted: Thu Jun 26, 2008 11:08 am
by frank_fg
Hi! Al

How did you get the photos on the forum?

Do you get the magazine Servo?
If you do, it is on May issue on page 34, called Servo buddy.

I'm sorry I did not say it before, but I did not want to give it away.

But I see, that I'v taken so much of your time, and I do appriciate.

IC2 555 timer is the osillator which enables you to 50 Hz or 250 Hz.

Well I have been working on 50 Hz side, in that the servo would only
run CW, and R6 pot had no control at all.

Well I did something different, I jumpered R3a and R3b to gat 250 Hz,
signal and there it was, the trigger signal, but when I removed it, it went
back to 5 v. and stayed there. It looks like around 20 usec negative pulse
What is that telling me?

On 250 Hz it works and on 50 Hz it doesn't.

Now to find out why R6 has no control.

A lot of workm just to get my servo running.

Thanks
Al

Frank G.

Posted: Thu Jun 26, 2008 11:28 am
by MrAl
Hi Frank,

To get an image in the post you click "Add image to post" which is
written right under the window you type your message inside of.
Look immediately under the window and you'll see it.
When you click that, you get a box that allows you to upload a picture
and then that gets shown in your post later.

Sorry i dont yet get the magazine, but if you draw the schematic
i will be able to help you right away. You can also send a scan in
my PM box and i'll take a look if you prefer to do that.
Dont give up yet you're almost done.

Posted: Fri Jun 27, 2008 12:15 pm
by frank_fg
Hi! Al

Good news!

Fixed my problems a and b, the a problem was me, the brain was not engaged, as I was drinking my coffee this morning it accurred to me
that the servo was turning counter clockwise only, which tells me that the one-shot had to been working, and the negative triggr was so small and possibly the brightness of my scope was not high enough that I could'n see it.

b) was the 100 k pot, that was really 100 Ohms, that would explain
why I got no response from it, replaced it with one of my own.

Thank you so much for your help and you were a lot of help, in the old
days just talking to someone was a big help.

Don't go away yet maybe you could help me with the next one.

As I said the servo circuit is working now, 'But' the control pot is ultra
sensitive, the second I turn it, the servo goes 360 degrees more or less.

In that I am using it as a steering system, I would like a smoother control.
The 100 k pot is already paralleled with 220K ohms, I was thinkingin of trying diferent resistances to narrow the range.

What do you think?


Frank G.

Posted: Fri Jun 27, 2008 1:57 pm
by MrAl
Hi Frank,

Sounds good now. I am happy you made some progress so now
you are probably much happier about your circuit.
I also have found that sometimes just talking about the circuit
helps to bring things out into the open that you didnt think about
before talking about it. Thankfully we can do that here pretty
well, and if you learn how to draw and post your circuits you'll
be that much farther too, and others can chime in with their ideas
too.

Since your pot isnt controlling your servo well enough and you think
you need a lower value, you could try the parallel resistor but
unfortunately that doesnt always work for every circuit. It depends on
the load and the pot value. You could try it, and if it doesnt work
you'll have to try different value pots. Unfortunately i still cant see
your circuit so it's hard to give any fast fixes...if you post a circuit
it will speed things up greatly i can tell you that much.
Sometimes you can increase the resolution of the adjustment at
the expense of the range, but again i'd have to see the circuit first
to know which technique would work correctly.

Posted: Fri Jun 27, 2008 4:24 pm
by frank_fg
Hi! Al

Hope you thi9s works.



Image

Image

Posted: Fri Jun 27, 2008 4:29 pm
by frank_fg
Hi!


Looks like it worked.

The valuase are:

R5 = 33k
R6 = 100k
R7220k

C6 = 0.022 ufd

It seems more stable at 250 Hz,
Jitters at 50HZ

thanks

Frank G.
\

Posted: Sat Jun 28, 2008 1:36 am
by MrAl
Hi Frank,

Oh good, that picture seems to be clear enough too.

Now just one more thing...

Do you know what the required pulse widths are out of IC3 pin 3
are supposed to be?
That is, at one extreme of the pot and the other extreme (min and
max pulse width), what the pulse widths should be.

While we are at it, what are the other values of the parts associated
with both IC2 and IC3? (resistor values, cap values).

Posted: Sat Jun 28, 2008 7:23 am
by frank_fg
Hi! Al

On another project working with a servo, I turn the pot until it was were
I wanted it, measured the resistance, and did the same thing on the other side. It worked, but it got there fast. I don't want this to happen her.

Spec's are:

Oscilliation frequency 50 Hz and 250 Hz, At 0.8 ms to 2.5 ms pulse width

Parts:

R2 = 22.0k 1%, R3a = 20.0k 5%, R3b = 133k 1%, R4 = 10k
R5 = 33k, R6 = 100k pot, R7 = 220k,R8 = 100, all are 5%

C3 = 0.1 ufd 50v, C4 = .001 ufd, C5 = 0.1 ufd, C6 = 0.022 ufd,
C7 = 0.1 ufd, C8 = 0.1 ufd, all monolthic

D3 = 1N914, IC's 555 CMOS

Thanks

Frank G.