MOVs

[Cursitor Doom]

Hi,

I've noticed a high failure rate in appliances built-in mains filters.
These are the filters that are typically found at the point the power
lead enters the appliance. They're silver coloured things about the size
of a matchbox and they tend to fail short and blow the fuse. The ones
I've seen are usually rated for 1A since the only stuff I do is low
power. However, some of that low power stuff is test equipment with SMPSs
that initially look like a short circuit at power-on. I think this is
what blows those filters. Early SMPS designs didn't seem to pay much
attention to surge prevention so I'm wondering if it would pay to
retrospectively fit MOVs at the beginning of the SMPS section to prolong
filter and power supply life (hopefully).
Any thoughts on the advisability of doing this and which MOVs are best
suited to this purpose?
Thanks.

CD.

 

[Jeff Liebermann]

On Fri, 22 Apr 2016 08:11:21 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

I've noticed a high failure rate in appliances built-in mains filters.
These are the filters that are typically found at the point the power
lead enters the appliance. They're silver coloured things about the size
of a matchbox and they tend to fail short and blow the fuse. The ones
I've seen are usually rated for 1A since the only stuff I do is low
power. However, some of that low power stuff is test equipment with SMPSs
that initially look like a short circuit at power-on. I think this is
what blows those filters. Early SMPS designs didn't seem to pay much
attention to surge prevention so I'm wondering if it would pay to
retrospectively fit MOVs at the beginning of the SMPS section to prolong
filter and power supply life (hopefully).
Any thoughts on the advisability of doing this and which MOVs are best
suited to this purpose?
Thanks.

I think you're mixing up MOV and NTC Thermistor (surgistor or inrush
current limiter). The MOV is an overvoltage limiter and goes ACROSS
the power line. The NTC thermistor goes in SERIES with the AC power
and limits the initial inrush current. You'll need to trace out the
schematic of the power line filter to see which one (or both) you're
dealing with.

In normal use, I've never seen MOV's or surgistors blow. I've seen
pleny of them blow up when the AC line conditions are all wrong, such
as lightning hits, unbalanced phases causing the voltage to increase,
power supply overload causing the surgistor to draw too much current,
and such. I've also seen a few design errors, where the surgistor
specs were inproperly selected.

I can guess my way through your problem, but I'm busy/lazy and would
prefer you to do the dirty work. Do some measurements such as a scope
picture of the inrush current, normal operation, extreme operation,
etc with a line current sensor. Get the numbers off the MOV and
surgistor and I'll see if they were properly selected. Read some
design articles on selecting inrush current limiters.
<http://powerelectronics.com/community/how-do-you-choose-right-type-ntc-thermistor-limit-inrush-current-capacitive-applications>

Incidentally, most (not all) of the devices I've seen that have inrush
current problems also use slow blow fuses.

Good luck.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Cursitor Doom]

Thanks for that correction, Jeff. Yes, in my haste I got the wrong term
for the component in question. Must be my advancing age.

 

rosoft Help



got to www.cccam-host.net

 

[Phil Allison]

Cursitor Doom wrote:

I've noticed a high failure rate in appliances built-in mains filters.
These are the filters that are typically found at the point the power
lead enters the appliance. They're silver coloured things about the size
of a matchbox and they tend to fail short and blow the fuse. The ones
I've seen are usually rated for 1A since the only stuff I do is low
power.

** Going short circuit is likely the result of one of the capacitors across the AC line failing - generally these are "class X" types which have a limited life expectancy. The other possibility is the "common mode" choke but that is very unlikely.

However, some of that low power stuff is test equipment with SMPSs
that initially look like a short circuit at power-on. I think this is
what blows those filters.

** Very unlikely too. The surge current may be quite large but also very brief, so not enough energy to damage copper winding wire which in any case would blow open.

Early SMPS designs didn't seem to pay much
attention to surge prevention

** So the equipment IS rather old ? Like 20 years +

so I'm wondering if it would pay to
retrospectively fit MOVs at the beginning of the SMPS section to prolong
filter and power supply life (hopefully).

** Small SMPSs can use a resistor in series with the supply to limit surges - say 5 or 10 ohms, wire-wound. Otherwise it is normal to use an NTC thermistor with similar cold resistance values.

The resistor or NTC limits the surge current to a value that is tolerated indefinitely by fuses, diodes and switches used in the PSU - ie something like 20amps peak for a couple of milliseconds.

PSU makers also have to consider the situation where multiple supplies are on the same AC circuit and all switched together creating a very large surge that may trip the breaker.


..... Phil

Any thoughts on the advisability of doing this and which MOVs are best
suited to this purpose?
Thanks.

CD.

 

[Cursitor Doom]

On Fri, 22 Apr 2016 18:44:11 -0700, Phil Allison wrote:

** Going short circuit is likely the result of one of the capacitors
across the AC line failing - generally these are "class X" types which
have a limited life expectancy. The other possibility is the "common
mode" choke but that is very unlikely.

Yes, most like a cap.

> ** So the equipment IS rather old ? Like 20 years +

Almost always. Typically 70s & 80s. I can't deal with SMDs anyway.

** Small SMPSs can use a resistor in series with the supply to limit
surges - say 5 or 10 ohms, wire-wound. Otherwise it is normal to use an
NTC thermistor with similar cold resistance values.

Noted, thank you.

 

<The resistor or NTC limits the surge current to a value that is tolerated <indefinitely by fuses, diodes and switches used in the PSU - ie something like <20amps peak for a couple of milliseconds.

I think that since the surge current occurs during the peak of the 120V 60Hz power waveform, I think the timing is more like several milliseconds or a few 10's of milliseconds at most, but not just a couple of milliseconds.

 

[Phil Allison]

hrho...@att.net wrote:


" The resistor or NTC limits the surge current to a value that is tolerated indefinitely by fuses, diodes and switches used in the PSU - ie something like 20amps peak for a couple of milliseconds. "

I think that since the surge current occurs during the peak of the 120V
60Hz power waveform, I think the timing is more like several milliseconds
or a few 10's of milliseconds at most, but not just a couple of
milliseconds.

** The surge occurs soon as the PSU is switched on - the magnitude depends on the timing relative to the AC wave. My comments referred to the worst case surge, when the switch on is at or near an AC voltage peak.

When this happens and if the total series resistance is say 5 ohms and the main electro is 220uF, the time constant is 1mS.

Remember, the context was all about *SMALL* SMPSs.


..... Phil

 

[Cursitor Doom]

I suppose the advantage of the resistor solution - apart from the fact
that I already have lots of suitable ones in my junk box - is that the PSU
remains protected even when the power supply is temporarily interrupted,
say when someone turns it off and on again over a short interval.