My designs are fussy about noise and fussy about latchup, so there is almost always a series resistor or current limiter between the power source and the power pin on any device. This causes the electrical rules check to flag every such power pin. Is there a way to disable this? Putting a PWR_FLAG on every power pin is distracting clutter.
You could disable that particular check throughout the design. The other person I know on this forum who uses series decoupling will probably weigh in soon.
If you are putting latch up limiters in series with every power pin, just disable this ERC check. In the end ERC is only a crude circuit design check and has no effect on PCB manufacturability.
Thanks. Understanding that this is under the “schematic setup” icon was what I needed.
It’s not called series decoupling. just current limiting.
ERC is really there to stop you from forgetting to connect stuff. I never use it for anything else. Having it to stop you connecting outputs to outputs, power supplies together etc, that’s a much more basic error. If you do that accidently, you are highly likely to have errors the ERC will not catch, that’s why IMO it is a waste of time.
On the other hand- - It’s easy to forget to connect a pin, or net …
It’s series decoupling if the purpose is to attenuate crosstalk between chips that share power. If that’s the only purpose, the resistor may be too small to be a serious current limiter.
1 Like
Long ago, they were typically 10~100 Ohm and just limited current until a power supply crowbar triggered.
Military stuff
I think this words mean something else than Google translator tells me. It translates it to “I’ll probably weigh myself soon”.
I use 0603 1k ferrite bead on each IC power supply pin. There are 0603 1k signal ferrite beads (small nominal current like 200mA) and supply ferrite beads with lower Rdc and higher nominal current (like 800mA).
Using spectrum analyzer I measured how ferrite beads impedance in a function of frequency depends on DC current flowing through it (to isolate from DC supply I used the same ferrite bead so what I measured was two ferrite beads in paralel (for AC)). This supply ferrite beads not only have lower Rdc but retain their properties longer as the DC current flowing through them increases (for me at 100mA such 800mA FB is really better than such 200mA). So I use supply FB also at signal lines (one roll on the machine instead of two).
For noise it is much better to have 1k at 100MHz than resistor that typically can’t be so big at supply rail, I think.
In your case I would consider using FB and R in serie.
1 Like
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.