[SI-LIST] Re: decoupling capacitors

Juan,
Sometimes the datasheet will tell you the transient-current requirements
for a part.  For example, a microprocessor that has a "sleep mode" will
show a:
*  Sudden increase in its core-current draw going from sleep mode to 
   operating mode.  (You usually have no way to control this, except
   not using sleep mode.)
*  Sudden decrease in its core-current draw going from operating mode
   to sleep mode.  (If the part has multiple power-down modes, you may
   be able to go through these in sequence to reduce this transient.)

For most parts, the transient current draw will be the sum of:
1.  Current to drive outputs from low-to-high.
2.  Internal "cross-over" currents.

You can estimate the first of these from the sum of the maximum output
short-circuit currents (Ios) for all the outputs that *could* switch
low-to-high at the same time.  

Cross-over current is harder to get a handle on.  Part of it is due to
the internal design of a chip, and a so-called "low emissions" ASIC
family may actually be worse from a Radiated Emissions standpoint than
a plain ASIC family.  (We had this happen at a previous employer.  It
was most irksome.)  

Especially with CMOS parts, the cross-over current may be very sensitive
to the edge rate (dV/dt) on the inputs.  Back in 1972, when the CD4050's
cost $3 each in quantity, I blew up two tubes of the parts experimenting
to determine why they were going "poof" on a controller board we were
developing.  I discovered that that they could momentarily  draw over
0.3A from a +15V supply:
*  With one out of six inputs slowly transitioning from high-to-low or
   vice versa.
*  With one out of six outputs loaded by more than 100-200pF.

With the package only rated for about 0.3W, many CD4050's wouldn't
survive even one test cycle.  Instead there would be a nice crater on
top, over what used to be the chip...

Some datasheets will specify that inputs must transition faster than x
V/ns, to prevent this problem.

I hope this explanation helps.
                                                John Barnes
                                                dBi Corporation
 
Juan Manuel wrote:
> 
> Hi, everybody.
> I'm looking for a method for calculating the appropiated capacitance of 
> bypassing capacitors. The most common method is this :
> c=I*N*dt/DV
> The problem that I find in this method is: What is the typical value ofthe I 
> ( current needed to switch a output), I know that this current is function of 
> the logic families.
> someone knows how obtain these instantanoeus current?
> Thank beforehand.
> juanma
> 
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
> 
> or to administer your membership from a web page, go to:
> http://www.freelists.org/webpage/si-list
> 
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
> 
> List archives are viewable at:
>                 http://www.freelists.org/archives/si-list
> or at our remote archives:
>                 http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
>                 http://www.qsl.net/wb6tpu
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: