AVR-Chat

usually, bypass capacitors are placed at all vcc pins of ic's and are 
layouted as near as posible to filter out voltage ripples...i tried to 
measure the ripples using digital scope and it was around 300mV...but 
i think my scope is not accurate since i get a 300mV ripples even when 
my probe is not connected or connected to ground...my question is, 
What is the tolerable voltage ripples? there is this called embeded 
capacitance where in the VCC plane has embeded capacitance.By using 
this, the .1 uF bypass capacitors are elliminated...Any idea on how to
compare its performance with the one with .1 uF bypass capacitors?
regards

mago

On 12/13/06, magzky02 <magzky02@yahoo.com> wrote:
>
> usually, bypass capacitors are placed at all vcc pins of ic's and are
> layouted as near as posible to filter out voltage ripples...i tried to
> measure the ripples using digital scope and it was around 300mV...but
> i think my scope is not accurate since i get a 300mV ripples even when
> my probe is not connected or connected to ground...


Definitely a problem, you may have large circulating currents.
Do you get this if you connect the probes only to the scope's ground?

my question is,
> What is the tolerable voltage ripples?


Yes.
Only you can say what's tolerable really.. The chip can take a fair bit, but
the question also involves your external hardware.  In one project I worked
on, I had a 24 watt chopped stepper driver less than an inch from an
amplifier that had to pick up microvolt sized signals from a magnetic read
head.  This also had a small flyback switcher, and 12 mhz micro.  With
careful layout and bypassing, on a two layer board, my noise on the power
rails was around 50mV. But everything matters, even how you connect two
bypass caps sitting next to each other, in parallel.


 there is this called embeded
> capacitance where in the VCC plane has embeded capacitance.By using
> this, the .1 uF bypass capacitors are elliminated...Any idea on how to
> compare its performance with the one with .1 uF bypass capacitors?
> regards


Unless your board is HUGE, just forget it. There isn't enough capacitance to
matter.

>




-- 
Feel the power of the dark side!  Atmel AVR


[Non-text portions of this message have been removed]

On Wed, 13 Dec 2006 08:28:29 -0000
 "magzky02" <magzky02@yahoo.com> wrote:
> usually, bypass capacitors are placed at all vcc pins of
> ic's and are 
> layouted as near as posible to filter out voltage
> ripples...i tried to 
> measure the ripples using digital scope and it was around
> 300mV...but 
> i think my scope is not accurate since i get a 300mV
> ripples even when 
> my probe is not connected or connected to ground...my
> question is, 
> What is the tolerable voltage ripples? there is this
> called embeded 
> capacitance where in the VCC plane has embeded
> capacitance.By using 
> this, the .1 uF bypass capacitors are elliminated...Any
> idea on how to
> compare its performance with the one with .1 uF bypass
> capacitors?
> regards
> 
> mago
> 
> 

There was a very similar question to this on AVRFreaks very
recently.

As long as the ripple voltage stays within the specified
voltage limits of the processor, IT is OK.

If you are doing a PWM output, the ripple will be
superimposed on the high logic output of the PWM signal.
Hopefully, your PWM averaging filter will also remove the
ripple since it should be even higher in frequency than the
PWM. 

I am willing to bet that the "ripple" you see on the
oscilloscope is AC line frequency, not processor clock.

Jim
---------------------------------------------------------------
The Think Different Store
http://www.thinkdifferentstore.com/
For All Your Mac Gear
---------------------------------------------------------------

But what about spurious spikes and surges caused by noise from the mains getting through?

I remember an old home computer I had that would crash every time  the fridge switched on. Clearly poor  noise filtering.

http://www.ckp-railways.talktalk.net/pcbcad21.htm
 

Jim Wagner <jim_d_wagner@applelinks.net> wrote:                                  On Wed, 13 Dec 2006 08:28:29 -0000
  "magzky02" <magzky02@yahoo.com> wrote:
 > usually, bypass capacitors are placed at all vcc pins of
 > ic's and are 
 > layouted as near as posible to filter out voltage
 > ripples...i tried to 
 > measure the ripples using digital scope and it was around
 > 300mV...but 
 > i think my scope is not accurate since i get a 300mV
 > ripples even when 
 > my probe is not connected or connected to ground...my
 > question is, 
 > What is the tolerable voltage ripples? there is this
 > called embeded 
 > capacitance where in the VCC plane has embeded
 > capacitance.By using 
 > this, the .1 uF bypass capacitors are elliminated...Any
 > idea on how to
 > compare its performance with the one with .1 uF bypass
 > capacitors?
 > regards
 > 
 > mago
 > 
 > 
 
 There was a very similar question to this on AVRFreaks very
 recently.
 
 As long as the ripple voltage stays within the specified
 voltage limits of the processor, IT is OK.
 
 If you are doing a PWM output, the ripple will be
 superimposed on the high logic output of the PWM signal.
 Hopefully, your PWM averaging filter will also remove the
 ripple since it should be even higher in frequency than the
 PWM. 
 
 I am willing to bet that the "ripple" you see on the
 oscilloscope is AC line frequency, not processor clock.
 
 Jim
 ----------------------------------------------------------
 The Think Different Store
 http://www.thinkdifferentstore.com/
 For All Your Mac Gear
 ----------------------------------------------------------
 
     
                       

 		
---------------------------------
 All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine

[Non-text portions of this message have been removed]

On 12/13/06, np np <harrabylad@yahoo.co.uk> wrote:
>
> But what about spurious spikes and surges caused by noise from the mains
> getting through?
>
> I remember an old home computer I had that would crash every time  the
> fridge switched on. Clearly poor  noise filtering.


That's a whole different issue.


[Non-text portions of this message have been removed]

See that wire on your scope probe?  That is an antenna, get rid of it.
 Take of the probe clip (witches hat) and wrap a bare wire around the
metal ground sleeve.  Probe again using this wire to ground, it should
be as short as possible, less than an inch.  I bet things look different.

Rich
--- In AVR-Chat@yahoogroups.com, "magzky02" <magzky02@...> wrote:

>
> usually, bypass capacitors are placed at all vcc pins of ic's and are 
> layouted as near as posible to filter out voltage ripples...i tried to 
> measure the ripples using digital scope and it was around 300mV...but 
> i think my scope is not accurate since i get a 300mV ripples even when 
> my probe is not connected or connected to ground...my question is, 
> What is the tolerable voltage ripples? there is this called embeded 
> capacitance where in the VCC plane has embeded capacitance.By using 
> this, the .1 uF bypass capacitors are elliminated...Any idea on how to
> compare its performance with the one with .1 uF bypass capacitors?
> regards
> 
> mago
>