how is a TVS different? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

how is a TVS different?

1 2 next

Move Thread

LAN_403

IamSmooth

Sat Sept 19 2009, 04:58PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

what is the difference in using a TVS vs a regular fast acting diode to suppress voltage spikes? For example, I used on on a small audio transformer that I was pulsing with a FET. I had used it to clamp the kickback spikes. When does one really want to use a transient voltage suppressor instead of a plain diode?

Finn Hammer

Sat Sept 19 2009, 05:26PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

A TVS shares properties of the Zener diode: It doesnÂ´t start to conduct before a certain voltage has been reached. Therefore it is able to clamp spikes appearing above a DC bias.
An ordinary diode would start conducting already ta 0.3-0.6 volts.

Hope this helps,

Cheers, Finn Hammer

Proud Mary

Sat Sept 19 2009, 05:33PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Hei Smoothie,

TVS diodes are essentially transient over-voltage protection devices, which clamp very sharply when a certain voltage is exceeded. For example, a 5V3 TVS diode between the positive and negative rails of a 5V supply would clamp very sharply (avalanche breakdown occurs in a matter of picoseconds) if the voltage should reach 5V3 due to surge or spike. Ideally, before the clamping point the TVS diode will look like an open circuit, but when Vclamp has been reached, it looks like a closed circuit across the supply.

If your PSU was born under a wandrin' star, a nice meaty TVS diode will simply blow the fuses for you if the upward voltage drift should reach Vclamp long before you smell the smoke.

An interesting feature of the TVS diode is that it is available in both unidirectional and bidirectional forms -the bidirectional type has a lot in common with the old dual common kathode zener - or individual zeners back to back in a clipper circuit.

They're cheap, and available in voltages from about 3V to 400V or so, and unlike MOVs they don't get all worn out and sorry for themselves after repeated breakdowns. They'd save a lot of costly MOSFETs on this forum if folk placed them at centre stage in the battle against the evil inductive kick-back spike.

IamSmooth

Sat Sept 19 2009, 06:47PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

I understand that a diode will start conducting near 0.3v. What I would like to know is why would one choose a TVS for this circuit, rather than use a fast-acting diode. It seems to me that either one would protect the mosfet from kick-back.

I used for this circuit a unidirectional 62v TVS. The transformer had a small kick-back which caused a larger voltage on the primary allowing me to get an even higher voltage on the secondary for charging a capacitor. The maximum drain-source voltage

is 500v. Does this 500v maximum allow for a negative drain voltage?

Sulaiman

Sat Sept 19 2009, 07:42PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

If a fast diode such as UF4007 is put in series with the TVS
less power will be lost due to charging and discharging the TVS each cycle.
The voltage rating of the TVS should be just below maximum Vds of the fet,

e.g. For an IRFP460 that has Vds max = 500V
you may think a (for example) 1.5KE440 would be good, rated for 440 V clamping.
that's for 1mA, if the peak primary current is for example 3.5 A
the TVS would clamp to 603 V (datasheet)
a 1.5KE350 with 4A would clamp at 482 Volts, a better choice.

You want the highest flyback voltage possible to allow the highest secondary voltage.

If you just had a simple diode instead of the TVS it would do no good at all,
mosfets have a diode from source to drain as part of internal structure
so the diode is redundant,
Sometimes a diode across the winding is used to KILL flyback spikes
for a flyback inverter that would be useless, it's main use is things like relay coils etc.

Proud Mary

Sat Sept 19 2009, 11:22PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Whoops, Smoothie, I don't think anyone has actually said that TVS is short for Transient Voltage Suppressor,which makes it easier to remember.

...

Sun Sept 20 2009, 04:42AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

It is my understanding that for all intensive purposes a zener diode is the same thing as a tvs. If one wanted to get picky, the TVS relies on 'avalanche breakdown' on the semiconductor, where the zener diode relies on 'zener breakdown' which are subtly different mechanisms to cause a diode to break down--the only practical difference between them being that the temperature coefficient of zener breakdown is positive (higher temperature -> higher breakdown voltage), where the temp coefficient for the avalanche mechanism is negative (higher temperature -> lower breakdown voltage).

As to the comparison between a fast diode and a tvs; the TVS will let the primary 'ring' up to a high(ish) voltage before it begins to clamp like the fast diode would. The way to pick which one to use is simple--if you are expecting your transformer to act as a classic transformer (ie, Vout = Vin * turns ratio) then use a fast diode, if you are expecting it to act as a flyback (ie, Vout = way bigger than Vin) then use the TVS.

Dr. Dark Current

Sun Sept 20 2009, 08:07AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

The MOSFET protect itself from overvoltage destruction by mechanism called avalanche breakdown. It can usually clamp the same current as it can switch, limited only by junction temperature.

Mates

Sun Sept 20 2009, 08:29PM

Registered Member #1025 Joined: Sun Sept 23 2007, 07:53PM
Location: Czech Rep.
Posts: 566

Dr. Kilovolt wrote ...

The MOSFET protect itself from overvoltage destruction by mechanism called avalanche breakdown. It can usually clamp the same current as it can switch, limited only by junction temperature.

I disagree Jan, most of todayâ€™s FET and IGBT transistors are indeed avalanche rated, but the energy they can handle in avalanche mode is much smaller than the switching energies in case the avalanche switch is repetitive. I have a huge collection of HV spikes destroyed FETs which all could have been saved in case I would use TVS. Avalanche mode is definitely unwanted situation for every transistor and using TVS is something I would always recommend even if the used transistors are avalanche rated.

Dr. Dark Current

Sun Sept 20 2009, 09:59PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Mates wrote ...

Well... I have seen your line-powered single switch drivers... I guess then, they simply overheated by the huge amount of avalanche...
If you look at a few MOSFET datasheets, you'll usually see something along the lines that the maximum avalanche current is the same as maximum on-state current, and that the avalanche dissipation is limited only by maximum junction temperature allowed.

On the other hand, IGBTs are usually NOT avalanche rated.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob