Author Topic: Parallel Zener and TVS - Bad idea?  (Read 1840 times)

0 Members and 1 Guest are viewing this topic.

Offline laurengTopic starter

  • Newbie
  • Posts: 7
  • Country: au
Parallel Zener and TVS - Bad idea?
« on: October 29, 2021, 04:36:38 am »
Hi all,

I'm designing a circuit with an external input, and I'd like to protect against transient voltage (e.g. static, indirect lightning) and constant over-voltage (e.g. someone plugging too high a voltage into the input). I'm thinking of using a TVS and Zener in parallel, with the TVS handling surges and the zener handling DC over-voltage. E.g. if the line is nominally 12V but can tolerate up to 20V, use a zener with a 20V breakdown, and a TVS with a reverse standoff of 20V and a breakdown of ~22V. I get that a TVS is kind of like a zener anyway, but I'm assuming the zener will be better at dissipating power in a sustained over-voltage.

I get the feeling that it's not so straightforward using them both in parallel, though. Is there anything I'm missing, and what do I need to be careful of?

One alternative I've seen is using something like a polyfuse in front of the TVS to limit current through it in an over-voltage, but then your polyfuse is potentially seeing e.g. static transients, which might not be good for it.

Cheers!
 

Offline srb1954

  • Super Contributor
  • ***
  • Posts: 1111
  • Country: nz
  • Retired Electronics Design Engineer
Re: Parallel Zener and TVS - Bad idea?
« Reply #1 on: October 29, 2021, 06:30:12 am »
Hi all,

I'm designing a circuit with an external input, and I'd like to protect against transient voltage (e.g. static, indirect lightning) and constant over-voltage (e.g. someone plugging too high a voltage into the input). I'm thinking of using a TVS and Zener in parallel, with the TVS handling surges and the zener handling DC over-voltage. E.g. if the line is nominally 12V but can tolerate up to 20V, use a zener with a 20V breakdown, and a TVS with a reverse standoff of 20V and a breakdown of ~22V. I get that a TVS is kind of like a zener anyway, but I'm assuming the zener will be better at dissipating power in a sustained over-voltage.

I get the feeling that it's not so straightforward using them both in parallel, though. Is there anything I'm missing, and what do I need to be careful of?

One alternative I've seen is using something like a polyfuse in front of the TVS to limit current through it in an over-voltage, but then your polyfuse is potentially seeing e.g. static transients, which might not be good for it.

Cheers!
Simply having the TVS and Zener in parallel will not work as it is likely that the 20V Zener will be significantly damaged before the TVS starts clamping the voltage at >22V.

What you really need is a current limiting impedance between the TVS and the Zener to limit the peak Zener current to safe levels during a transient. The current limiting impedance could be an inductor for short duration transients or it would need top be a resistor for continuous overloads. The impedance/resistance needs to be calculated so that the peak current through the Zener, when the TVS is clamping to its maximum clamp voltage, is within the peak current/power rating of the Zener. Limiting the peak current through the Zener also ensures that the clamping voltage is better controlled providing better protection for downstream circuitry.

A polyfuse preceding the TVS is a good idea if you are expecting prolonged overvoltage conditions but the polyfuse will be too slow to act in the case of short transients and won't provide much additional protection for the TVS in that case.
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3591
  • Country: fr
Re: Parallel Zener and TVS - Bad idea?
« Reply #2 on: October 29, 2021, 10:57:07 am »
Please detail the application, power source and transients expected.

A 10 kW DC><DC will needs a differnet protection than a 1W.

A 12V lab PSU source will have a different requirement  than a 12V vehicle bus with ALT/GEN load dump transients.

A sketch or block of ALL system components would be helpful.

Jon

Jean-Paul  the Internet Dinosaur
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22432
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Parallel Zener and TVS - Bad idea?
« Reply #3 on: October 29, 2021, 01:36:45 pm »
The TVS is already a zener, so there's no need to use extra. :)

DC overvoltage depends on source impedance.  Some surges can be clamped, or even bypassed (ESD, EFT, even 8/20us induced surge to some extent), others are difficult or impossible to clamp (automotive load dump, crossed wiring overvoltage).  You may need to ride through or disconnect the overvoltage, instead.

Note that polyfuses take quite a long time to open, 100s ms -- in this time the TVS will have cooked off.  They used to make three-terminal paired parts, where the TVS heats the poly-goo making it open quickly and safely on overvoltage, but they're pretty much unavailable these days AFAIK.  Alas.  You may instead need an electric fuse or current limiter, or some other circuitry to handle it.  Or a good old fashioned crowbar, with a big enough SCR to handle clearing current of the fuse.

Also, note that polyfuses have maximum current and voltage ratings; current may be easily exceeded even well within voltage ratings.  Check the internal resistance, and add more if necessary.  Yes, this will reduce the efficiency of your system; such is the price for safe operation, and a hint that a more elaborate method (as above) may be desirable.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: thm_w

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3591
  • Country: fr
Re: Parallel Zener and TVS - Bad idea?
« Reply #4 on: October 29, 2021, 03:15:30 pm »
Bonjour, TAZ - Transient Absorb Zeners are very large die Zener with specs for max Vz at a specified transient eg 1 mS 1500 joules 12 V nom part clamps at 18V

Any series Z like a polyfuse will defeat the protection as the fuse Resistance will add substantial voltage to the protected bus.

Suggest to read tech apps notes on TAZ.

Kind Regards,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17167
  • Country: us
  • DavidH
Re: Parallel Zener and TVS - Bad idea?
« Reply #5 on: October 30, 2021, 01:26:39 am »
I'm thinking of using a TVS and Zener in parallel, with the TVS handling surges and the zener handling DC over-voltage.

A TVS is a type of zener diode.  (1) It is characterized differently, available in larger sizes, and processed to have a uniform junction preventing hot spots from forming like an avalanche rated rectifier.

Quote
One alternative I've seen is using something like a polyfuse in front of the TVS to limit current through it in an over-voltage, but then your polyfuse is potentially seeing e.g. static transients, which might not be good for it.

Some way is needed to limit the current or protecting against failure from applied over-voltage will just move the failure to a different spot.  I think a polyfuse is one of the better options in this case.

(1) Really avalanche diode for parts above about 7 volts.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf