8

For 220 V AC mains voltage, is it good practice to replace a 300 V MOV with a 300 V bidirectional TVS diode (like in here http://www.littelfuse.com/products/tvs-diodes.aspx), or connect both in parallel? Are there any points that need to be considered?

enter image description here

downthewire
  • 127
  • 2
  • 3
  • 8
  • 1
    A MOV is a TVS. If you're talking about putting a Zener type TVS in parallel with a MOV, there isn't much to be gained there. – Matt Young Feb 01 '15 at 17:23
  • 3
    Can you explain why? Can't I benefit from the faster response time of the TVS diode? Keep in mind I'm asking both about replacement and connection in parallel. – downthewire Feb 01 '15 at 18:07

2 Answers2

24

It comes downto what are you trying to protect against

There are four main types of transient suppression devices

Gas Tube

  • Protection time: > 1us
  • Protection Voltage: 60 - 100V
  • PowerDissipation: Nil
  • Reliable Performance: No
  • Expected Life: Limited
  • Other: Only 50-2500 surges, can short powerlines

MOV

  • Protection time: 10 - 20ns
  • Protection Voltage: > 300V
  • PowerDissipation: Nil
  • Reliable Performance: No
  • Expected Life: Degrades
  • Other: Fusing required. Degrades

Avalanche TVS

  • Protection time: 50ps
  • Protection Voltage: 3-400V
  • PowerDissipation: low
  • Reliable Performance: yes
  • Expected Life: long
  • Other: Low power dissipation. Bidirectional requires dual

Thyristor TVS

  • Protection time: <3ns
  • Protection Voltage: 30-400V
  • PowerDissipation: Nil
  • Reliable Performance: yes
  • Expected Life: long
  • Other: High Capacitance

http://www.onsemi.com/pub_link/Collateral/HBD854-D.PDF

( http://web.archive.org/web/20051001082352/http://www.onsemi.com/pub/Collateral/HBD854-D.PDF )

http://www.vishay.com/docs/88440/failurem.pdf

  • Sorry, I should have been more clear, I'm talking about TVS diodes like in here : http://www.littelfuse.com/products/tvs-diodes.aspx. – downthewire Feb 01 '15 at 18:05
  • I figured that. My point still stands, what you pick is dependent on what you are trying to protect against. The only real reason to supplement a MOV with a TVS is if response time is required and then you would have the TVS voltage slightly higher, but then having the MOV there then becomes questionable –  Feb 01 '15 at 18:11
  • Gas tubes will need at least 10us to completely fire. The thing with transient voltage protector is they differ in terms of speed, the amount of current they can swallow and how long a pulse they can handle. Adittionally some of them like gas and MOVs are guaranteed to fail in "short circuit". A TVS diode may or may not fail in that way. – iggy Feb 01 '15 at 18:12
  • 1
    IEC 60950-1 explicitly proscribes MOVs instead of TVS Diodes, when used as surge suppressors. – avl_sweden Jul 07 '15 at 15:51
  • 9
    "Proscribe" means "Forbid". The appropriate word here is "prescribe". Exactly opposite meaning. – user1713481 Sep 06 '15 at 18:57
  • @avl_sweden: I can't find any reference to that. Can you point me to anything? – Rev Mar 17 '17 at 12:30
  • 1
    "1.5.9.1 General It is permitted to use any type of surge suppressor, including a voltage dependent resistor in a SECONDARY CIRCUIT. If a surge suppressor is used in a PRIMARY CIRCUIT , it shall be a (VDR), and it shall comply with Annex Q." – avl_sweden Apr 03 '17 at 07:54
  • And yes, I meant prescribe, not proscribe. I'm not a native english speaker :-) – avl_sweden Apr 03 '17 at 07:56
  • 1
    The standard then goes on to state: "NOTE 1 A VDR is sometimes referred to as a varistor or a metal oxide varistor (MOV). Devices such as gas discharge tubes, carbon blocks and semiconductor devices with non-linear voltage/current characteristics are not considered as VDRs in this standard." – avl_sweden Apr 03 '17 at 07:57
  • All this is from IEC 60950-1 Second edition 2005-12.

    I don't know if there is a later version.

     – avl_sweden Apr 03 '17 at 07:57
  • From my reading, saying MOV's have "nil" power dissipation is wrong; actually they do more than shunt current, they have a significant resistance while conducting (generally more/much more than TVS diodes), and can get very hot. You can also get them in huge sizes to absorb tens of thousands of Joules. – Ralph Versteegen Dec 01 '18 at 13:40
3

The overvoltage protection circuit subject is complex and there simply isn't a one size fits all remedy. Things to consider are: where will the circuit be located in relation to the service entrance, are you attempting to protect for short and/or long duration surges, and what type of surge are you planning on protecting against? From my research, the best attempt at a one size fits all overvoltage protection scheme would include line conditioning filtering (inductors/capacitors), (MOV) Metal Oxide Varistors with integral thermal disconnect fuses and Silicon Avalanche Suppressor Diodes. I would talk to the engineers at Littlefuse for advice. They have a large library of downloadable PDF's pertaining to Overvoltage Suppression.

Neil_UK
  • 166,079
  • 3
  • 185
  • 408
David
  • 31
  • 1