Searching \ for '[EE:] Howto make a quality square wave clock circu' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/timers.htm?key=clock
Search entire site for: 'Howto make a quality square wave clock circu'.

Exact match. Not showing close matches.
PICList Thread
'[EE:] Howto make a quality square wave clock circu'
2004\06\17@154503 by tony

flavicon
face
Hi all,

I've been struggling with making a good quality sqare wave circuit for a
digital circuit I'm trying to build.

I am having to feed the clock circuit to ~15 IC's, included in that is a
CPU.  I am trying to get it to run at between 10 and 12 MHz, but it seems
that I run into a few problems I don't know how to solve....

I get a pretty good spike on the rise and fall of the clock above Vcc and
below GND, as I feed it to more of the IC's, for some reason, the Vil is
raising, so when it is low, instead of it sitting at about 0Vdc (or there
abouts), it is hovering at around 1Vdc, which is too close to comfort for
transition state for the IC's - especially if there is a bit of ringing
that can have a peak at above 1.5V causing a potential problem for the
CPU.

So, my questions to those that are much more capable engineers then I (I
am a technician more then an engineer and am doing this as hobby, not
work) - how can I clamp that Vil down to ground properly so its ringing
won't be able to go into thinking there has been a transition. And how can
I reduce some of the ringing?

I have been using TTL oscillator cans between 8 and 12 MHz feeding into a
74LS04 as well as trying a 7404 to buffer it, I have also tried building a
clock circuit utilizing an 8 MHz crystal and a 7404 along with some
resistors and caps (it has been more positive then other methods thus
far).  I have tried swapping out the 7404/LS for a 74AC04 which has made
the ringing worse (appears to be that it is too sensitive and is
amplifying my problems).

Where can I go to learn more about this and how to fix my issue?  Or does
anyone have any suggestions?

-Tony

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\06\17@155353 by rixy04

flavicon
face
TTL does not swing from Vcc to Gnd. You could get a little closer with a totem
pole made up of two transistors that will give you more drive and closer to
the rail swing. Are you sure the ringing is from the circuit and not the scope
probe? Have you calibrated the probe? Why do you need to swing to the rails
anyhow?
Rick

spam_OUTtonyTakeThisOuTspamELROYNET.COM wrote:

{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\06\17@162145 by tony

flavicon
face
Yes, I'm pretty sure the ringing is the circuit, I've checked the probes,
they are fully calibrated, and I have checked it on multiple scopes to
make sure I wasn't loosing my mind, the scope I was last using was a
200MHz scope.

I'm not looking for complete rail to rail swing, but I am looking for
closer to ground on low - the high is fine - where high is (at ~4V) is
high enough to run the rest of the circuits no problem, even with the
amount of ring that is present on the high (which is not nearly as much as
there is on the low), but the low, being where it is, I believe is causing
some of my problems.  If I could get the low swing down a little bit, and
a bit more stable, I'd be happy.  If it isn't a design problem from me, my
other thought is that there is enough resistance and crosstalk in the
wirewrap wires that it is causing me problems, but I am not sure; I am
having to run the clock wires a bit farther then I would like.

When I run a different speed (4MHz) where I don't seem to be seeing this
problem, there is no problem clocking through data, but at the higher
speed, data is either not clocked or clocked sporatically, which *appears*
to indicate a clock problem, like it isn't seeing that the clock is
staying low long enough and is hovering at too high a voltage point.

All of the devices are rated to operate normally (the other chips) at
14MHz or lower, I want 10MHz, but can survive at 8 MHz.

-Tony




{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\06\17@172140 by Tom

flavicon
face
At 03:21 PM 6/17/04 -0500, you wrote:
>Yes, I'm pretty sure the ringing is the circuit, I've checked the probes,
>they are fully calibrated, and I have checked it on multiple scopes to
>make sure I wasn't loosing my mind, the scope I was last using was a
>200MHz scope.
>

Just out of curiosity, how long is your scope probe ground lead?

At these freqs it needs to be less than 2 inches max (better shorter) or it
will show ringing that isn't really there.

As for fanout, you can always try a clock distribution system - 3 or 4
inverters, inputs from your clock source, outputs to different destinations.

Good luck!
Tom

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\06\17@175424 by llile

flavicon
face
Although you are using TTL, would CMOS parts be able to swing closer to
the rails?  I agree with the fanout problem - several parallel paths (all
with identical time delays) will solve this problem better than one buffer
would.

-- Lawrence Lile
Senior Project Engineer
Toastmaster, Inc.
Division of Salton, Inc.
573-446-5661 voice
573-446-5676 fax




Tom <.....kristKILLspamspam@spam@THEGRID.NET>
Sent by: pic microcontroller discussion list <PICLISTspamKILLspamMITVMA.MIT.EDU>
06/17/2004 04:25 PM
Please respond to pic microcontroller discussion list


       To:     .....PICLISTKILLspamspam.....MITVMA.MIT.EDU
       cc:
       Subject:        Re: [EE:] Howto make a quality square wave clock circuit??


At 03:21 PM 6/17/04 -0500, you wrote:
>Yes, I'm pretty sure the ringing is the circuit, I've checked the probes,
>they are fully calibrated, and I have checked it on multiple scopes to
>make sure I wasn't loosing my mind, the scope I was last using was a
>200MHz scope.
>

Just out of curiosity, how long is your scope probe ground lead?

At these freqs it needs to be less than 2 inches max (better shorter) or
it
will show ringing that isn't really there.

As for fanout, you can always try a clock distribution system - 3 or 4
inverters, inputs from your clock source, outputs to different
destinations.

Good luck!
Tom

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.



--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\06\17@230034 by Richard Graziano

picon face
Ringing occurs when the source impedance and the load impedance are
different. Sometimes, the ringing is in the scope measurement because the
signal being measure is not properly terminated, i.e., 50 ohms, depending on
the measurement design.

Good square wave clocks are generally developed by dividing a frequency with
a counter or flip flop.  For example, if you need a stable 20 MHz clock,
start with a 40 MHz oscillator and use a D-Type flip flop to divide by 2.
The more you divide down, the better the stability with respect to jitter.
However, ringing can still be a problem and careful attention to impedance
is needed.




{Original Message removed}

2004\06\19@060830 by tony

flavicon
face
> Ringing occurs when the source impedance and the load impedance are
> different. Sometimes, the ringing is in the scope measurement because the
> signal being measure is not properly terminated, i.e., 50 ohms, depending
> on
> the measurement design.
>
> Good square wave clocks are generally developed by dividing a frequency
> with
> a counter or flip flop.  For example, if you need a stable 20 MHz clock,
> start with a 40 MHz oscillator and use a D-Type flip flop to divide by 2.
> The more you divide down, the better the stability with respect to jitter.
> However, ringing can still be a problem and careful attention to impedance
> is needed.
>

This is the way I'm going at this point, I've taken one of my 40MHz cans
and thrown it thru a couple of flip-flops getting it down to 10 MHz and it
looks fairly clean, much cleaner then what the 10MHz can was putting out -
thank you for the information.

Now I will probably feed it thru a set of inverters to help to increase
fanout hopefully without loading down the clock itself (which is what Tom
had suggested, and was already the direction I was heading before, I
thought it was a good idea then, and when he had mentioned it I felt I was
on track with my way of spreading the signal to the different parts of the
circuit.

-Tony

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu

2004\06\19@081514 by Peter van Hoof
picon face
I like to use the 4049 or the 4050 for the buffer , has a nice swing and
high current drive, they can be had from digikey for a couple dimes.

Peter

{Original Message removed}

2004\06\19@084508 by Richard Graziano

picon face
Remember that some inverters are not drivers.  You can also use a
noninverting driver. The more stages you feed through the more signal delay
you generate.  Parallel inverter drivers or noninverting drivers, whichever
is appropriate, will increase the sourcing capability and buffer the clock
while not increasing the signal delay time as much.
{Original Message removed}

2004\06\21@040229 by Vincent Vega

picon face
Tony, most of your problems are likely to be layout
and loading related. Without seeing your circuit is
hard to help you out, because it can be whatever.
Check this book in your local library:
High Speed Digital Design: A handbook of black magic.
By Howard W. Johnson ISBN  0-13-395724-1


Don't be scared by the *black magic* stuff, there's nothing
magic and certainly not black about engineering, just a
little bit of thinking. That's all you need.
Best regards
VV

tonyspamspam_OUTELROYNET.COM wrote:

>Hi all,

>I've been struggling with making a good quality sqare wave circuit for a
>digital circuit I'm trying to build.

> am having to feed the clock circuit to ~15 IC's, included in that is a
>CPU. I am trying to get it to run at between 10 and 12 MHz, but it seems
>that I run into a few problems I don't know how to solve....

[snip]

>Where can I go to learn more about this and how to fix my issue? Or does
>anyone have any suggestions?

>-Tony



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\06\21@040439 by Vincent Vega

picon face
What about the line in between? ;-)
Best regards
VV

Richard Graziano <@spam@rgrazia1KILLspamspamROCHESTER.RR.COM> wrote:

>Ringing occurs when the source impedance and the load impedance are
>different.



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\06\21@114647 by tony

flavicon
face
> Tony, most of your problems are likely to be layout
> and loading related. Without seeing your circuit is
> hard to help you out, because it can be whatever.
> Check this book in your local library:
> High Speed Digital Design: A handbook of black magic.
> By Howard W. Johnson ISBN  0-13-395724-1
>
>
> Don't be scared by the *black magic* stuff, there's nothing
> magic and certainly not black about engineering, just a
> little bit of thinking. That's all you need.
> Best regards
> VV

LoL, I had seen that book title once before and thought about it, but
didn't have the cash at the time to buy it, then managed to forget what it
was called - thanks for sending it, now I can find it again :)

I've had more luck now with using flip flops and buffers, but I still will
probably get the book as I would like to learn more about creating solid
high speed circuits.

-Tony

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\06\21@121012 by David VanHorn

flavicon
face
At 10:46 AM 6/21/2004 -0500, KILLspamtonyKILLspamspamELROYNET.COM wrote:

>> Tony, most of your problems are likely to be layout
>> and loading related. Without seeing your circuit is
>> hard to help you out, because it can be whatever.
>> Check this book in your local library:
>> High Speed Digital Design: A handbook of black magic.
>> By Howard W. Johnson ISBN  0-13-395724-1

It's a good reference, but it may be hard to find in your library.
There's always Amazon.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\06\21@222005 by Russell McMahon

face
flavicon
face
> >> Tony, most of your problems are likely to be layout
> >> and loading related. Without seeing your circuit is
> >> hard to help you out, because it can be whatever.
> >> Check this book in your local library:
> >> High Speed Digital Design: A handbook of black magic.
> >> By Howard W. Johnson ISBN  0-13-395724-1

Or see, as per my post yesterday

> A very good resource on the general topic of digital signal integrity.
>
> http://www.signalintegrity.com/pubsChron.htm


       RM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

More... (looser matching)
- Last day of these posts
- In 2004 , 2005 only
- Today
- New search...