[IC] Electrostatic charge in switched-capacitor circuit

Dear good guys,

We have a question in switched-capacitor circuit that is

when we design switched-capacitor circuits, do we have to consider the
problem of "electrostatic charge" ?

does it is possible the capacitor was "full of electrostatic charge" before
circuit connected to power supply?

does it is possible the full of electrostatic charge capacitor causes IC
borken?


Thank you very much!

Best regards,

Boki.

 

Jamie, thank you very much, but I am a little confusing about it.

You mean there is a switch/delay circuit that perform discharge task first,
and then while switched-capacitor circuit start work?



Thank you very much!

Best regards,

Boki.

 

yes, it can cause problems like current surges.
that is one reason you many use a timer to delay
on a relay mean while a resistor is used to get the
charge in the cap going so that when the relay
does switch in it will not surge it.

 

actually the other way around.
you use a resistor to slowly charge the cap to get it
out of the region of where the inrush surge would be
tomuch for the ic. i am not sure exactly what your
problem is ? maybe you have highvoltage building on the
cap while it is left unloaded ? if that is the case you could
put a drain resistor on the cap (always).

 

Subject: [IC] Electrostatic charge in switched-capacitor circuit

I'm assuming you're designing a switched capacitor filter or something of that
sort, and that you're using a switched-capacitor filter IC to do the job. Any
additional information you might provide would be helpful.

Any residual charge on the cap when the power supply is turned off will bleed
down through the substrate. A couple of points:

1) Of course, you should always be sure the cap is discharged before inserting
it in the circuit during assembly.

2) Especially if you have a switching power supply, look carefully at how fast
the supply creeps down at turn-off. A lot of inputs won't have much to limit
input current in the even the cap has a higher voltage than the + rail or lower
than the - rail. Arrange so the power supplies drop off slowly, and evenly if
possible to keep that cap bleed-off slow.

3) If you've got dual supplies, this can be much more of a problem. In dual
supplies, if the two turn off unequally, sneak paths through the power supply
load can cause one of the power supply rails to be reversed in its relation to
GND. This is bad news. If this might be a problem, try two back-biased diodes
protecting each power supply rail between the power supply output and
GND/common. This also protects things at turn-on, for the same reasons.

Power Supply Diode Protection
____
V+ | | +5V
o-------o------|7805|-----o-------o-----o
| |____| | |
| | | |
--- | --- -
--- | --- ^
| | | |
o-------o--------o--------o-------o-----o
| | | | GND
--- | --- |
--- | --- -
| _|__ | ^
| | | | |
o-------o------|7905|-----o-------o-----o
V- |____| -5V

This is general advice which applies to all ICs that have inputs that can be
connected to a charged cap at turn-off. If power supply turn-on and turn-off
aren't the problem (that's the biggie), I've found most reliability problems
with switched capacitor circuits to be with ESD.

Sorry -- wish I could be more helpful, but you haven't given us a lot to run
with.

Good luck
Chris

 

Wow, thank you very much for such clear description.
It will be very useful.

When I measured the sc chip, I will report to every body ~ thank you very much~

Best regards,
Boki.

 

Is this right? : )
C
vin -------/ ----||---- \------->OP input
| | |
| | |
/ | \
| | |
| | |
GND | GND
|
R
|
\ Switch_R
|
GND


Regards,
Boki