0
\$\begingroup\$

I have a PCB with 3.3V data lines and power supply lines. Both require ESD TVS diodes. In the past I have used the same PESD0402-140 datasheet diode for both, but I feel like this is likely not a good idea. The power supply can have up to 2A. Because this diode is low capacitance and the lines are at the same voltage, is there a problem with doing this?

\$\endgroup\$
4
  • \$\begingroup\$ How do you intend to use the same diode for both? Wouldn't that require tying the lines together, shorting the data line to the power supply? \$\endgroup\$
    – Hearth
    Oct 17 at 0:18
  • 3
    \$\begingroup\$ @Hearth Presumably the same part number and not the same individual diode. \$\endgroup\$
    – DKNguyen
    Oct 17 at 2:10
  • \$\begingroup\$ @DKNguyen right. I could have been more clear on this but I meant using multiple of the same diode on the pcb. \$\endgroup\$
    – Feynman137
    Oct 17 at 2:35
  • \$\begingroup\$ Close/open to clear close votes after editing. \$\endgroup\$
    – Russell McMahon
    Oct 30 at 10:26
4
\$\begingroup\$

The TVS won’t fire unless its threshold is reached.

The question becomes, what kind of abnormal voltage do you expect on the 3.3V line? If it’s ESD, a TVS applied to it will be dissipating the same energy as a hit on the data lines.

If there’s a possibly of overvoltage from the power source, that needs a different type of protection than a TVS (or in addition to the TVS.)

Assuming overvoltage isn’t a problem, then using the same TVS for data and power is fine.

\$\endgroup\$
1
\$\begingroup\$

In line of principle the component would be the same. However there are performance limits and parasitics to keep in mind.

Suppressors for data lines usually have less capacitive influence (often on the pF order) to avoid distortion in the signal path: an 'idle' TVS is substantially a junction capacitance plus some leakage current; all of this in the datasheet.

On supply lines you don't have this issue (often capacitance in power lines is beneficial, actually). So suppressor designed for power lines are optimized for dissipation and low leakage. The idea is that arguably on power lines transients are stronger.

You need consider the kind, shape and amplitude of the expected transient (the standard one is the 8/20µs if to want to look at how is made). Your 2A supply are not even entering your TVS… it's in shunt position, not in series like a choke. However look out for the minimum breakdown voltage since it could start conduction even at nominal voltages (look at the standoff voltage, the minimum voltage required for trigger it)

In practice since the clamp power is limited and the TVS is everything but a precision component (varistors are even worse) you'll need to derate. Your part is actually an ESD suppressor (different wave shape) but the idea is the same:

  • You can put on it 14V without any issue (standoff)
  • It triggers more or less at 250V
  • When it trigger you'll still have 40V to handle somehow

The 'handle' part for ESD on signal lines is usually 'let the clamp diodes eat it'. Since clamp diodes are toward the supply theorically you still would have 40V on the supply line. Luckily you should have lots of bypass capacitors and these usually suck up on line transients.

So especially for ESD unless you have special endurance requirement a data line TVS is usually applicable to a power line. For other kind of transients (like EFT, switching noise and worse of all the automotive load dump) there are tighter components available,

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.