Pop` wrote:
Prolog:
François Martzloff was the surge guru at the NIST (still thought of by
some of us old people as the National Bureau of Standards). He did
research and wrote many peer-reviewed papers, many of them available
(along with papers from others) at an anthology he maintains at:
http://www.eeel.nist.gov/817/pubs/spd-anthology/
One of them is:
http://www.eeel.nist.gov/817/pubs/spd-anthology/files/Troubleshoot%20PQ.pdf
which is a guide for phone service reps for co-op power utilities. It is
considerably more technical than the IEEE guide, and has a lot of
information on surges caused by powerline switching - may be of interest
to a few people.
He also wrote the NIST guide to surges and surge protection at:
http://www.nist.gov/public_affairs/practiceguides/surgesfnl.pdf
which is aimed at the general public. This is the same link as Charles
Schuler provided.
Post by Pop`...
Post by Bud--Equipment can withstand perhaps 800V. Selecting too low a clamp
voltage results in the suppressor clamping more surges than is
required to protect equipment.
Of course, but you're neglecting the time/power curve of current as it
reaches and passes the "knee" of the conduction cycle.
Post by Bud--Clamping a surge results in power
dissipated in MOVs (the primary protection element), and MOVs
deteriorate and can eventually fail as cumulative dissipation goes
up.
True, but what you're missing is that, nowdays, even the varistor ckts, and
all the solid state ckts, also include a fusing arrangement that completely
opens the ckt. In good equipment, changing the fuse is all that's required.
Cheaper equipment may not allow for the fuse to be changed, but, the fuse
"fixes" the old problem of the MOVs if you insist, going bad without knowing
it.
I did not comment on MOVs going “bad without knowing”.
If a MOV goes bad, changing the fuse won’t help - the new fuse will blow
too. The MOV is now low resistance.
I presume you are not saying a fuse will protect a MOV from failure from
cumulative dissipation - the MOV is already fatally damaged.
Post by Pop`Post by Bud--In addition, low clamp voltage can cause rapid failure if there
is utility overvoltage.
No, not in a properly designed protection ckt. In a misdesign, yes, but few
of those exist on the market due to UL/CSA/EC/NOMs, etc.. If you don't meet
their specs, you don't sell it (or aren't supposed to be able to<g>).
From the Martzloff co-op paper cited at the top - pdf-page 20:
“In Fact, the major cause of TVSS failures is a temporary overvoltage,
rather than an unusually large surge.”
Post by Pop`Post by Bud--The author of the NIST guide on surges says
that overvoltage is the most frequent cause of failure of surge
protectors.
1. MOVs are for the cheapie units, and do actually a pretty fair job of
protection. There is some accuracy in your statement, but ... 800V (peaks =
800 x 2.828) is going to be pretty disastrous to most equipment that sees
it. It looks like you're missing the point of application points for this
equipment.
2.8x gives the peak-to-peak voltage which is of no use.
800V is the peak voltage, not the RMS. Equipment fails on the peak and
lightning surges are more like a pulse making peak appropriate. 800V
withstand came from an old PC magazine evaluation on plug-in surge arrester.
From a Martzloff technical paper - pdf-page 20:
http://www.eeel.nist.gov/817/pubs/spd-anthology/files/Enduser.pdf
“The fact of the matter is that nowadays, most electronic appliances
have an inherent immunity level of at least 600 V to 800 V, so that the
clamping voltages of 330 V widely offered by TVSS manufacturers are
really not necessary. Objective assessment of the situation leads to the
conclusion that the 330 V clamping level, promoted by a few
manufacturers, was encouraged by the promulgation of UL Std 1449,
showing that voltage as the lowest in a series of possible clamping
voltages for 120 V circuits. Thus was created the downward auction of
"lower is better" notwithstanding the objections raised by several
researchers [B8] and well-informed manufacturers. One of the
consequences of this downward auction can be premature ageing of TVSS
that are called upon to carry surge currents as the result of relatively
low transient voltages that would not put equipment in jeopardy.”
Post by Pop`2. Your complaint about 130V clamping voltage is incorrect. The "knee" of
the operating impedance waveform initiates at 130V. It begins passing
measurable current, usually still in milliamps, at about 150V. That will
NOT burn out the varistors and other solid state devices used. Actually,
varistors have given way to better components these days but are still used
in some designs, especially inside the protected equipments.
As the energy content passes the "knee" voltage, current pass increases
exponentially until it is essentially a short circuit.
I didn’t talk about a 130V [RMS] clamping voltage. I am not talking
about 130V, but overvoltage which is significantly over the normal line
voltage that puts lower UL voltage classes of MOVs above the “knee”.
I have not seen anything that indicates MOVs aren’t the dominant element
that actually provides surge protection for power circuits. From the
IEEE guide [published 2005] - guide-page 37:
“The vast majority (>90%) of both hard-wired and plug-in protectors use
MOVs to perform the voltage–limiting function. In most AC protectors,
they are the only significant voltage limiters.”
Post by Pop`Post by Bud--A major reason manufacturers have low clamp voltage ratings on their
surge suppressors is that it "sounds better". Higher voltages provide
better overall protection.
No, it's because it's good design these days, and/or the not so often
mentioned idea is to sacrifice the protector rather than the protected
equipment. Such vague terms as you are using are misleading and
misinformation at best.
Repeating from the Enduser.pdf paper above:
“Objective assessment of the situation leads to the conclusion that the
330 V clamping level, promoted by a few manufacturers, was encouraged by
the promulgation of UL Std 1449, showing that voltage as the lowest in a
series of possible clamping voltages for 120 V circuits. Thus was
created the downward auction of "lower is better" notwithstanding the
objections raised by several researchers [B8] and well-informed
manufacturers.”
Post by Pop`Post by Bud--http://www.mikeholt.com/files/PDF/LightningGuide_FINALpublishedversion_May051.pdf
says Joule rating is not reliable for _comparison_ because there is no
standard for how the rating is determined, and comparisons are thus
apples and oranges. The guide recommends current ratings, IIRC, and
has guidelines on adequate ratings.
The above is not what the IEEE guide says; you need to go read it again. It
points out the misuse of the term, and possible problems with it, but does
not recommend current ratings; it recommends well beyond just the current
ratings. When you read such documents, you have to take the entire document
into consideration, not just the paragraphs that seem to support your own
stance, because there is often a lot more to it than the one paragraph.
I read it - all of it.
The IEEE guide gives guidance on values for service panel current and
clamp-voltage ratings on guide-pages 18-19. It gives guidance on plug-in
protector clamp-voltage ratings and discusses UL test currents
guide-page 37-38. It gives warnings on using Joule ratings pages 25 and
40. It gives no Joule ratings. I have provided more endorsement of Joule
ratings than the IEEE guide.
Post by Pop`Post by Bud--But energy ratings are still important. MOVs deteriorate with surge
hits.
Yes, that's true. Also time is extremely important. As is total energy.
The unfortunate part of all this is that such information as in the IEEE
goes well beyond the use of typical lay people; it's not targetted at, nor
meant for, them. When the discussions go in this direction, the layperson
is left out completely.
The IEEE guide is targeted at people ”with some technical background”
and is well within the capabilities of many here, but I usually include
a link to the NIST guide (provided above and by Charles) which is aimed
at the unwashed masses.
Post by Pop`Post by Bud--With a higher Joule rating a suppressor can withstand larger
hits. And the total cumulative energy rating of a MOV will be much
higher than the stated energy rating
No, it would be much LESS!! It becomes less effective also under such
circumstances. This is one of the main reasons fusing technology came into
being for surge protection. A fuse has its own time/heat/melt/etc ratings
which works in conjunction with the specs on the device doing the energy
detouring.
I suspect I am reading you wrong. Fuses aren't of use on lightning
induced surges. A fast fuse can open in 1/4 cycle, about 4ms. Surges
from lightning are over in 20-100 microseconds. If a fuse opened that
fast it would have to survive the thousands of amps of surge current and
the thousands of volts that would develop across the fuse.
Post by Pop`Post by Bud--if the stated energy rating is
much higher that the single event hits. Higher energy ratings do not
just raise the simple sum of energy hits that can be absorbed.
I don't know what the rest of that means.
The idea is hard to convey in just a few words. Let me try again in
excessive gory detail.
MOVs have a MCOV voltage rating (maximum continuous operating voltage)
which is the voltage at which the MOVs conduct 1mA (you referred to
above). When a MOV clamps the voltage across it, it has a current
through it and a voltage across it, which means it absorbs energy. Thus
MOVs also have an energy (Joule) rating. If the MOV takes a single hit
that is equal to its Joule rating, the MCOV will be lowered by 10%. This
is considered the end of life. As it is used further the MCOV will
continue to lower and eventually it will conduct at normal circuit
voltages resulting in thermal runaway and high current failure. Surge
protectors have thermal protectors or fuses to disconnect in this
situation.
Values taken from an actual 330Joule MOV:
* with 330J hits the MOV can withstand 1 hit - cumulative energy
dissipated 330Joules
* with 90J hits the MOV can withstand 10 hits - cumulative energy
dissipation 900Joules
* with 24J hits the MOV can withstand 100 hits – cumulative energy
dissipation 2400Joules
* with 13.5J hits the MOV can withstand 1000 hits - cumulative energy
dissipation 13,500Joules.
Note the cumulative energy rising as the energy of the hits goes down.
The point I was trying to make is that if you get a surge protector that
is rated far above the energy it is likely to dissipate in a single
event, it will be able to take far more hits (and last far longer) than
the simple sum of the hits would predict. ‘Oversizing’ the Joule rating
has definite advantages.
Post by Pop`Post by Bud--Also very important is a single point ground - the phone, CATV, ...
protector earthing wire connecting with a short wire to the earthing
conductor from the panel, close to the panel. Surge currents on the
earthing wires can produce large voltage drops which can appear
between power and phone wires.
No, the difference in potentials is between the ground references for the
stated systems, not all the wires. All grounds are not equal.
If the ground reference for the CATV shifts away from the ground
reference for power, that voltage will appear at anything connected to
both. An example is in the IEEE guide starting guide-page 30 where a
CATV ground block is distant from the N-G bond at the power service by
30 feet. A 3000A surge on the CATV lead-in produces a voltage difference
of 10,000V between the CATV ground block and the power system N-G. That
voltage appears at TVs connected to power and CATV, and they won’t like it.
With a “single point ground” the CATV ground block is located adjacent
to the power service and connected with a short wire to the power
earthing conductor at the power service. If the resistance to earth is a
low 5 ohms and the same 3000A surge, the power and CATV grounds will
lift from ‘absolute earth’ by 15,000V. But because of the “single point
ground”, the CATV ground and power system grounds will be together,
protecting connected equipment. “Single point ground” is one of the most
important surge protection tools. [In the previous case above, the power
ground will lift 15,000V above ‘absolute ground’ and the CATV will lift
25,00V.]
Post by Pop`Post by Bud--Single point ground minimizes th
difference between power and phone (CATV, ...) wires.
Not sure what that's about; guess you're still reading the IEEE specs maybe.
I'm not trying to fly in your face here; simply to keep things straight and
nothing more. I'm about out of time now so you can say pretty much what
you'd like.
If your'e interested in some good links on grounding techniques with respect
to sensitive equipment, let me know and I can look a few of them up for you.
IEEE is pretty good, and so are some of the folks who interpret it, but
there are some better links for the layperson that will make things easier
to understand. I'd simply have to relocate them again; I don't see them in
my Favorites - damn that last rebuild<G>.
Always interested in surge info. Technical papers are good. If you are
looking for ways to kill an evening, week, ... the anthology at the top
has numerous papers on surges.
--
bud--