LED indicator for bad MOVs

[stefanv]

Does anyone know how to switch a LED to indicate if a varistor has gon
bad? I suppose if it shorted out it will be obvious but in case of a
open MOV, how can you connect a led as an indicator that it's still good?

Stefan V

 

[Ian Stirling]

Does anyone know how to switch a LED to indicate if a varistor has gone
bad? I suppose if it shorted out it will be obvious but in case of an
open MOV, how can you connect a led as an indicator that it's still good?

Unfortunately, the only reliable way is to determine if the voltage has
ever exceeded what the clamping voltage should be.
Actually measuring the clamp voltage can be done in som ecases.

 

[Don Klipstein]

Unfortunately, the only reliable way is to determine if the voltage has
ever exceeded what the clamping voltage should be.
Actually measuring the clamp voltage can be done in som ecases.

I have seen how this is done:

The product with the MOV has a fuse that is assumed to blow if the MOV
fails. Maybe the fuse is a thermal one placed against the MOV - I don't
remember that well. Supposedly maybe, not many MOVs fail open except in
response to something that would blow that fuse.

The indicator lamp goes across the fuse. If the MOV is open rather than
shorted, then you need some load in parallel with the MOV to draw current
through the indicator lamp.

- Don Klipstein ([email protected])

 

[Joerg]

Hello Stefan,

Does anyone know how to switch a LED to indicate if a varistor has gone
bad? I suppose if it shorted out it will be obvious but in case of an
open MOV, how can you connect a led as an indicator that it's still good?

With MOVs it isn't this easy. When it has gone open that means the
damage may have already been done, that it let some surge pass through.

MOVs are like bank accounts. They offer x many Joules of 'protective
power absorption'. But you never know how much is left in the account.
Will it be enough for one more surge? That's one reason why I don't use
them in my designs, usually.

A LED could tell you whether it failed. But that may be too late for
whatever was connected.

Regards, Joerg

 

[Don Klipstein]

Hello Stefan,


With MOVs it isn't this easy. When it has gone open that means the
damage may have already been done, that it let some surge pass through.

MOVs are like bank accounts. They offer x many Joules of 'protective
power absorption'. But you never know how much is left in the account.
Will it be enough for one more surge? That's one reason why I don't use
them in my designs, usually.

A LED could tell you whether it failed. But that may be too late for
whatever was connected.

I heard that the joule limit of an MOV is not lifetime, but closer to
per-incident. Short of bad incidents severe enough to make the thing hot
enough to degrade, they don't degrade from absorbing surges.

I try Google:

http://www.iaei.org/magazine/04_b/04_b_brown.htm

Says MOVs do not degrade if they don't get heated to the point of a
thermal runaway condition or to temperatures at which cause a significant
rate of degradation.

Although this article does mention surge history as a factor in
degradation of MOVs, the same sentence emphasizes surges that are in
excess of ratings. The paragraph with that sentence makes it appear to me
that the degradation is from temperature-related diffusion.

However, there is still the problem of an MOV experiencing a surge
failing open before any fuse upstream of it opens and before the surge is
over. That should only be a problem with especially catastrophic surges.

- Don Klipstein ([email protected])

 

[Joerg]

Hello Don,

I heard that the joule limit of an MOV is not lifetime, but closer to
per-incident. Short of bad incidents severe enough to make the thing hot
enough to degrade, they don't degrade from absorbing surges.

It can be either and the data sheet should state whether that is a 'per
event' or 'over the lifetime' rating, and under which test scenario it
was derived. According to most documents I read they do degrade even
when not exposed all the way into the thermal run-away region.

http://www.brickwall.com/nofail.htm
http://www.surgex.com/library/42001-2.html

Although this article does mention surge history as a factor in
degradation of MOVs, the same sentence emphasizes surges that are in
excess of ratings. The paragraph with that sentence makes it appear to me
that the degradation is from temperature-related diffusion.

There will always be local heating even with a massively parallel
structure. It's a matter of how much and not linear. But AFAIK they will
still age with non-severe surges that are significant enough.

However, there is still the problem of an MOV experiencing a surge
failing open before any fuse upstream of it opens and before the surge is
over. That should only be a problem with especially catastrophic surges.

Yes, it's the same problem as with the oil pressure light in a car. When
that light comes on it is often too late for the engine.

Regards, Joerg

 

[w_tom]

Going to Brickwall or Surgex for a discussion about MOVs is like
going to Rush Limbaugh for an honest analysis of Hilary. Above is this
idea that MOVs absorb surges. Not true. They are called shunt mode
devices for good electrical reasons. MOV is rated in joules much like
wire is rated in amps. Joules determine how much surge gets shunted
(diverted) just like wire. Unlike wire, MOVs degrade. And as joules
increase, the degradation decreases exponentially. Notice from here on
the numbers that series mode protectors forget to provide. No number
is how one 'lies by telling half truths'. It's called spin.

Some discuss life expectancy of MOVs is because many plug-in
protectors are so grossly undersized. If any MOV protector vaporizes,
then two things have happened: 1) the protector was grossly undersized.
2) MOVs operated in an unacceptable mode that even its own
manufacturer does not define. Charts for life expectancy rate 1, 10,
100, 1000, and 10000 transients. This is life expectancy where
threshold voltage eventually charges by 10% - no vaporization. There
is no curve for zero transients - which are what happens when MOVs
vaporize.

Brickwall and Surgex also forget to mention a wire that carries
destructive surges completely around -bypasses - their protectors.
Again, Rush Limbaugh type spin where they conveniently forget to
mention certain critical facts - assume a reader is that naive.

Brickwall and Surgex series mode protector are supplemental
protection. To be effective, a building must have shunt mode
protection. What is shunt mode protection? Often using MOVs because
MOVs are so effective. How can this be if MOVs stop or absorb surges?
They don't.

MOVs fail when a protector is not properly installed - undersized by
a human. This problem is common with plug-in protectors that don't
claim effective protection anyway. When MOVs vaporize or blow that
thermal fuse - that causes the light to change - then you know that
protector was grossly undersized, and failed leaving electronics
completely exposed to that last surge.

However when selling plug-in protectors to the naive, then profits
are higher by grossly undersizing those MOVs. The naive human will
foolishly assume, "The protector sacrificed itself to save my
computer". Reality: computer's internal protection saved it.

A quote directly from an MOV datasheet:

The change of Vb shall be measured after the impulse listed
below is applied 10,000 times continuously with the interval
of ten seconds at room temperature.

The series 14 MOV is rated for 150A standard transient pulses for
10,000 times. So where is this degradation? It exists. And
degradation is irrelevant when properly sized.

We install 'whole house' protectors for typically destructive
transients that occur once every eight years. Smaller transients (that
are often nothing more than noise) are made irrelevant by protection
already inside every appliance.

Numbers and facts that series mode protectors forgot to mention.

 

[ehsjr]

Going to Brickwall or Surgex for a discussion about MOVs is like
going to Rush Limbaugh for an honest analysis of Hilary. Above is this
idea that MOVs absorb surges. Not true.

They are (variable) resistors. P = I^2R. That's heat, created
as electrical energy is *absorbed*. When more energy is absorbed
by the MOV than it can dissipate, the MOV fails, sometimes
spectacularly. Stating that MOV's do not absorb surge energy
is an untenable position.

Ed

 

[stefanv]

thanks for the input!

Somehow this thread makes me feel more comfortable to work with MOV’
considering I use quality MOV’s at a high rating.
To protect the MOV’s somewhat, different stages are mentioned to reduc
the size of the surge by limiting it’s speed using HF inductors or L
snubbers. Any idea on values to get a better 117AC line protection? Pu
these after the Mov’s as 2nd stage or make them 1st stage protecting th
MOV’s also?

I feel that with $40 well spend on parts I’ll get a much better protectio
than these $100.00 devices. I’m note looking for fine noise reduction, jus
keeping the big bang out of my electronics.

Stefan V

 

[Joerg]

Hello w_tom,

Going to Brickwall or Surgex for a discussion about MOVs is like
going to Rush Limbaugh for an honest analysis of Hilary. ...

ROFL! That was a good one. Almost spilled my coffee when I read it. Once
I had to do some FFT algorithm tests and I just used the radio as an
arbitrary source. It happened to be Rush. Sometimes you could see the
spectral average of his voice slowly drift upwards. That was a sure sign
that he was slowly getting upset about a topic he just started.

Above is this idea that MOVs absorb surges. Not true. They are called shunt mode
devices for good electrical reasons. ...

With all due respect, shunt mode devices are called shunt mode because
the do dissipate energy. Except when it's a wire of infinite
conductivity. MOV transfer spike energy into heat.

MOVs fail when a protector is not properly installed - undersized by
a human. This problem is common with plug-in protectors that don't
claim effective protection anyway. When MOVs vaporize or blow that
thermal fuse - that causes the light to change - then you know that
protector was grossly undersized, and failed leaving electronics
completely exposed to that last surge.

Well, yes, but that's the same with almost every product. My SUV came
with truck tires so they have lasted 55000 miles by now. Still lots of
thread left on them. The tires on our sedan were long gone by then
because they were smaller and less durable car tires. Of course, the
truck tires cost a lot more.

I have nothing against MOV if they are large enough (they usually
aren't). But relying on them to prevent over-voltage isn't my cuppa tea.
You cannot select them too close to the rated rail or line voltage since
they aren't as precise as a voltage reference, meaning they could
'seep', get hot, age faster and then one day, 'Kaboom'. But if you have
to protect against some substantial surge as a first line of defense
they may be the ticket.

Regards, Joerg

 

[w_tom]

MOV connection to earth determines quality of that protection. Why
are so many MOV based protectors so undersized? Because plug-in
protectors don't have an effective earthing connection. No earth
ground means no effective protection. So why installed enough joules?
Effective MOV based protector that is also sufficiently sized is 'whole
house' protectors.

'Whole house' protectors being so effective and so inexpensive that
a telco installs one, for free, on your phone service. But again,
protector is only as effective as its earth ground. Earthing is
another essential part of protection that those application notes
forgot to discuss.

To appreciate the most critical component of a protection system,
learn about that most essential component - single point earth ground.

Numerous industry professional citations, concepts, experience,
numbers - about one days worth of technical reading describe protection
in hyperlinks previously posted in alt.comp.periphs.mainboard.asus on
30 Mar 2005 entitled "UPS unit needed for the P4C800E-Deluxe" at:
http://makeashorterlink.com/?X61C23DCA

Meanwhile, if concerned about catastrophic MOV failure, then consider
MOVs manufactured by Littelfuse that include thermal protection as part
of the component.

 

[Sjouke Burry]

MOV connection to earth determines quality of that protection. Why
are so many MOV based protectors so undersized? Because plug-in
protectors don't have an effective earthing connection. No earth
ground means no effective protection. So why installed enough joules?
Effective MOV based protector that is also sufficiently sized is 'whole
house' protectors.

'Whole house' protectors being so effective and so inexpensive that
a telco installs one, for free, on your phone service. But again, Hm.. Top post...

Why not improve your outside protection first?? A simple airgap with a
solid ground connection maybe. If that does not survive, what else wil??
Or else the gas discharge tubes,both make a nice path to ground,without
to much raise in line voltage.
Just blocking with an outside inductor will only help for the small
discharges,the bigger ones wil just jump across the inductor.
Anything getting past the tube or gap, might be stopped by the
fuse/inductor/MOV combination inside the building.
Even if the MOV shorts, at least it protects by blowing the fuse.
Of course fuse and MOV must be replaced after a heavy hit,or a number
of smaller ones.
There is no way you can protect against a full hit,but at least
such a combination should give you the best protection.