dimensioning of varistor for surge protection?

** Groper Alert !

** Why ?

An overvoltage spike on the AC supply merely causes a triac to trigger on for the remainder of the half cycle.

If that is gonna cause a disaster - you need to redesign the whole set up.

** What are you protecting against ?

Like home security - how much you need depends how badly folk are trying to break in.

...... Phil

Yes, if the voltage is not noo high. But a surge will nevertheless damage the device. We have several shorted triacs that has been investigated (by X-ray). Still I need to protect it! Someone suggested a varistor combined with a gas discharge tube (in series or in parallell?). But how to dimension this protection. Another component I have found is the Sidactor. Does anyone has any experience with that device?

** Over snipping, non quoting Groper Alert !

** How asinine.

Once triggered, the voltage across a triac drops like a ton of bricks.

** But you have no idea what sort - voltage or current.

** ROTFLMAO - what bollocks.

** From what ?

Why do YOU refuse to respond to this crucial question.

Makes you look a LIAR and a FOOL.

** WHATTTTTTTTTTT !!!!!

No I know you ARE a donkey brain..

Heeee haaaaww, heeeee haaaawww ......

....... Phil

I dont want any more stupid answers from you, Phil Allison. If you don't have any answer don't answer! Aren't there any serious persons reading this group.

Phil Allison skrev:

Well..I'll spare some of my "hack" hobby knowledge..it's not much help.. Some of which may not be applicable..

1) Crack open similar equipment and look at the surge protection electronics...Maybe a computer power supply but thats 120VAC.. 2) I've seen some extreme surge electronics that use just about every device MOV, TVS, gap, inductor, fuse, crowbar, varistor, resistor,transformer,zener.... My guess was somebody also didn't know how much protection and so overkill was done. 3) Sometimes my designs have a space constraint..For example..If I have a cubic inch of nothing...I'll fill it with overkill surge suppression.

D from BC

We don't have enough information. If a 800 volt triac is blowing out, and the protection varistor is blowing out, you have some really serious transients there.

I would put a line transient monitor box on the line and capture a few transients first. You can't protect against things you don't know the size and shape of.

You're correct that you can't protect against a direct lightning strike.

Varistors have a weak voltage cutting. Maybe it is possible that in your very low ohmic source circuit the actual voltage on the triac goes much higher than 800 volts and may kill the triac on the first strike or maybe due stress in a later time...

If the varistors exploded then this means simply that the pulse had much more energy than 84J ! Was it correctly voltage rated? For 230 volts you need at least an 400 volts type. If it is too low-rated it will be pre-heated and that will reduce the energy absorbing rating dramatically.

Varistors can absorb *much* energy. I think they are the most capable in this value. A spark gap, TVS, thyristor will reflect the energy back to the source! There must be a current-limiting circuit !

Circuits around the varistor can be mechanically protected by sleeving the varistor.

Maybe a simple non-saturating air-coil in series to your circuit will enough current-limit for the varistor and the varistor can still act as an voltage protector for the triac.

As suggested: Try to record the pulse events to see what is going on and analyze the possible sources.

regards - Henry

--

schrieb im Newsbeitrag news: snipped-for-privacy@o58g2000hsb.googlegroups.com...

snipped-for-privacy@swipnet.se ha scritto:

I think You must identify the source of surge (hypothetical) (the varistor is slow...)

If You don't identify this, You have a lot of possibility....

So, read this interesting application

and this...

Emanuele

--
*** YOUR ELECTRONICS OPEN SOURCE ***

http://dev.emcelettronica.com;---------------------------------------------------------

Take a look at a TrippLite Isobar and decide how much of that you can afford or have room for.

a sidactor is basically a triac with no gate. since they're built to be triggered by over voltage that may be in your favor.

Bye. Jasen

It's important to know the characteristic of the surge you are addressing. The energy level, impedance and duration will determine what is needed for minimal protection.

A combination of limiters would be required for general protection, in association with inserted series limiting impedances. These would likely be inductors capable of operating for the required interval in series with the line and resistors in series with shunt limiting elements to reduce peak energy absorption.

Within the surge, individual elements will be acting at different times, with the aim of reducing peak voltage or dv/dt from occuring across the victim.

Much literature out there. The GE-Harris-Littelfuse collection has the deepest selection of devices and application history to draw from.

RL

Of course you CAN protect from a direct lightning strike. Just provide a bunch of conductors with lower inductance than your sensitive device.

If you are asking to protect an AC power-line from lightning transients, expect to use a big expensive box to do it. An isolation transformer does pretty well, for instance (the coupling from input to output is frequency limited by the iron). If, on the other hand, you want to ABSORB the transient energy and not just reflect or shunt it, the problem is harder.

Protecting a triac might be as simple as knotting a power wire (to form a crude RF choke) and adding a fuse to the circuit (no, not because the fuse is fast enough; because the fuse has enough resistance to slow the transient and the dI/dt limit of the thyristor might not be exceeded). I've heard of folk using simple spark gaps effectively, too (a suitable

automobile spark plug can do a lot of protection while it melts, and replacements aren't too pricey),

A melted triac might result from other stress than surges, of course. I burned out several power-supply transistors trying to power a little CMOS gate once. It turns out that CMOS at liquid nitrogen temperatures does somewhat higher conductivity during transitions than the data sheet was gonna warn me about.

As previous posts have suggested, adding some series impedance will help. Also, by recording the input transient waveform, and knowing the minimum series impedance, you can calculate the energy in the transient pulse. Another thing you can do is use a gas discharge tube (GTD). These devices can absorb a lot of energy. One caveat about varistors: They degrade as they absorbe more and more hits. You might be better off using a TVS. TVSs are larger than varistos for the same energy spec, but they do not degrade. Regards, Jon