Breakdown and standoff voltage for TVS diodes in 3.3v system

Go To Last Post
10 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm running a atmega644 at 3.3v and have 2 io lines available to the end-user, usable as digital in or out or as analog input. I will need to protect these lines from ESD events.

I found a few threads discussing how to do this with this post being very informative.

https://www.avrfreaks.net/index.p...

So I set out to find an appropriate diode. The signals are not time critical on these lines.

My best guess for the values are:

The Reverse Standoff Voltage is 3.3v
Voltage - Breakdown: 3.5 or 3.6

Problem is that there are only a few parts, can I go to breakdown of 5v?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That is a real problem because there are few transient suppressors under 5V. Even USB has to be rated over 5V.

I think I would be tempted to use Schottky diodes from the signal line to Vcc and Ground. With a pair, neither will ever see more than Vcc+0,3V (plus higher current forward drop) as a reverse voltage.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I was afraid of that. Would things be easier if I only use the pins as inputs?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No, actually the opposite. Somewhat. When a pin is an output, there is already a (pretty) low resistance to either Vcc or Ground. It will be limited by the max current carrying capacity of the internal traces within the chip. So. for that reason, alone, it may be desirable to add in external diodes. But, having a pin as an output should make it easier.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I can't see why you could not use opto-couplers for either input or output coupling to a 3.3V system. This immediately provides good isolation & protection from ESD and additional clamping & limiting can be done on the galvanic decoupled side.
In the case of lightning, the opto's might be taken out, but the micro saved. Of course a really good direct strike might cause a few other issues! :)

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for those ideas.

Taking a step back and reading that thread again. https://www.avrfreaks.net/index.php?name=PNphpBB2&file=viewtopic&t=89249&postdays=0&postorder=asc&start=20.

JayJay proposes this schematic:

I realized I missed something crucial in that schematic, the capacitor, and the fact that this ESD is by nature very short lived. Thus even though the clamping voltage of TVS diodes never stays within the atmega allowed range, the capacitor will suppress/filter out the voltage spike.

So I think I'll go with this diode:
http://www.st.com/stonline/books...
a 220 ohm (or thereabouts) resistor.
and a 100nF cap.
r3 removed.

Last Edited: Mon. Jun 21, 2010 - 10:41 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
This immediately provides good isolation & protection from ESD and additional clamping & limiting can be done on the galvanic decoupled side.

Lee, you're forgetting that optos have a led and capacitance! The led needs some protection and for fast transients, the capacitance from in to out can couple the nasties. So, optos aren't quite the magic bullet one would hope for.

I was reminded of this when I was designing a motor controller. The AVR based board I was designing didn't mind the transients, but the opto coupled usb->485 adapter did. The usb chip would lock up randomly.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Can't you just have the TVS diode then a series resistor to the mcu (perhaps a capacitor also for some extra filtering)?

Use a 3.3V TVS then a 10k series resistor.
This way the TVS will clamp to around 5V, then the series resistor along with the AVR internal diode can deal with the rest.

Perhaps look at something like a CESD3V3D5 for the TVS.

You want the breakdown outside your operating area a little otherwise the diode will leak current.

oddbudman

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

the resistor and capacitor are the first line of defence by slowing the rise time of the discharge. The TVS diode just cleans up the rest.

The TVS diode is also good for longer term overloads but make sure the R can dissipate enough heat.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Running a few simulations it appears that 270 ohm resistor and 100nF cap with a 3.3/5 tvs will keep things within a safe range and still allow signals as fast as 1kHz to pass without too much distortion.

Thanks everyone for the answers. I'm probably ordering the first few 4 layer pcbs this week! Then it's on to the FCC testing!