Cross Country Faults?
Rowbotth
for what is called "Cross-country Faults". If I've been told the truth,
this would be for when there are two circuits on the same double circuit
tower, and the wind gets high enough to let both conductors of the same
phase slap together. This "Cross-country Fault" is supposed to know
what his is, and it is supposed to do a single poe reclose instead of a
3 phase trip.
Questions:
1) Any familiarity with this feature in use? (Does it work as well as
the mfr says it should?)
2) Any thoughts on whether the "Cross Country fault feature shold be
applied universally throughout the North American continent? Or is
there a high wind speed location where it would be most benefitial?
3) I'm told that this was applied years ago, but abandoned because it
didn't work in practice as well as in theory. Is this true? Is it
xpected to work better now?
Thanks, folks.
HR.
EDSEL ATLENZA
I'm assuming you're referring to distance protection in the SEL-421.
This logic is described in the SEL-421 Applications Handbook, Section 1,
Page 43, available here:
http://www.selinc.com/sel-421.htm
For those not familiar with the SEL relays, here is a brief description:
This scheme involves the use of 2 permissive signals for each line in a POTT
(Permissive Overreach Trip Transfer) scheme: one to key single pole trip and
the other to key a three phase trip.
Cross country faults involving the same phases on parallel lines should be
easy for any relay with single pole tripping. Relays from both lines on
both ends should detect single phase to ground faults, send permissive
signals for single pole tripping, trip single phase on both lines, and
attempt reclosing.
Cross country faults involving different phases close to one end of the line
becomes more complicated. It is possible that the relays on the end
furthest from the fault will see the fault as a phase to phase fault within
their overreaching zone 2 and send permissives for both single phase and
three phase tripping. The relays on the end closest to the fault should see
the fault as a phase to ground fault on each line simultaneously within zone
1, trip instantaneously, and send permissives for single phase tripping to
the relays at the end furthest from the fault. The relays on the end
furthest from the fault will receive the permissive signals to trip single
phase. The relays furthest from the fault would correctly trip after the
relays closest to the fault trip and the relays furthest from the fault
detect only single phase to ground faults. Relays on both ends would
attempt to reclose.
If you have parallel lines, need single pole tripping for cross country
faults, and have the communications support to provide 2 permissive signals
per line in each direction, use the additional logic. This logic does not
cost extra in the SEL-421. Be careful setting the reach on short lines.
Single pole tripping line current differential
(http://www.selinc.com/sel-311l.htm) would be faster and more secure but may
require greater communications support. Configuring a line current
differential relay is much simpler than configuring a distance relay
especially on short lines.
I hope this helps.
EA
"Rowbotth" <rowb...@telusplanet.net> wrote in message
news:rowbotth-847390...@news.telus.net...
Eur Ing John Rye
I have been following this thread with interest, and was surprised to find a
different meaning being given to "cross country fault" to the one to which I
am accustomed.
To me the term applies to the situation where one has earth faults on two
different feeders out of a substation, where the substation earthing is
resistance, reactance, or especially arc suppression coil.
Under these condition the fault current leaves the substation on one phase of
one feeder, flows cross country and returns on a different phase of a
different feeder.
They are significant because they produce a higher fault current in the
ground path than normal, and can have quite interesting implications for
protection schemes.
I should point out that this applies to the 3 wire MV distribution schemes
that we use in the UK, and I am not sure that an equivalent problem can arise
with the 4 wire schemes used in the USA.
John
--
EurIng J Rye CEng FIEE Electrical Engineering Consultant
18 Wentworth Close Hadleigh IPSWICH IP7 5SA England
Tel No 01473 827126 <http://web.ukonline.co.uk/jrye/index.html>
---< On Line using an Acorn StrongArm RiscPC >---
Ed Atienza
as the terms broken delta, open delta, and V connected voltage transformers.
It gets even more complicated when trying to translate into another
language. Our meanings for "cross country fault" are similar in that they
both involve two different lines or feeders, but we have two very different
applications. The distribution application seems much more complicated to
detect because of the wide range of fault resistance due to the variation in
distances between the faults.
Impedence grounded systems, especially systems grounded with an arc
suppression coil, limit fault current during single phase to ground faults
allowing continued operation of a faulted feeder. Aside from industrial
applications, the USA primarily uses solidly grounded systems resulting in
high fault currents for single phase to ground faults. This wouldn't allow
the possibility of operating during any fault conditions making the
likelyhood of having two different phases on two different feeders faulted
to ground simultaneously very unlikely.
On systems with arc suppression coils, how does the fault current magnitude
of cross country faults compare to the load current?
Would the wattmetric method still report directionality correctly for all of
the feeders out of the substation during a cross country fault?
Ed
"Eur Ing John Rye" <jr...@ukonline.co.uk> wrote in message
news:4c8646a...@ukonline.co.uk...
Eur Ing John Rye
In article <kRf%b.14550$fL4....@nwrddc01.gnilink.net>,
Ed Atienza <edat...@verizon.net> wrote:
> It's amazing how we define some terms and some concepts so differently such
> as the terms broken delta, open delta, and V connected voltage
> transformers. It gets even more complicated when trying to translate into
> another language.
Agreed
> Our meanings for "cross country fault" are similar in that they both
> involve two different lines or feeders, but we have two very different
> applications. The distribution application seems much more complicated to
> detect because of the wide range of fault resistance due to the variation
> in distances between the faults.
Agreed
> Impedence grounded systems, especially systems grounded with an arc
> suppression coil, limit fault current during single phase to ground faults
> allowing continued operation of a faulted feeder. Aside from industrial
> applications, the USA primarily uses solidly grounded systems resulting in
> high fault currents for single phase to ground faults. This wouldn't allow
> the possibility of operating during any fault conditions making the
> likelyhood of having two different phases on two different feeders faulted
> to ground simultaneously very unlikely.
> On systems with arc suppression coils, how does the fault current magnitude
> of cross country faults compare to the load current?
Can be several times the value, but clearly depends on how good the earth
connections at the two faults. The fun scenario is that both feeders see the
fault either as overcurrent or fault, so that if the condition remained
stable both feeder circuit-breakers would trip. However one will inevitably
beat the other, and as soon as one circuit-breaker opens the fault current
disappears and one is left with an ASC fault on the other feeder. After its
set time delay the first circuit-breaker recloses and the fault current may
or may not reappear depending on what caused the fault on this feeder. The
worst picture is that it breaks down again after a minute or two, and you go
all round the sequence again. It is not too serious a problem providing ones
operators are well trained and can appreciate what might be happening.
> Would the wattmetric method still report directionality correctly for all
> of the feeders out of the substation during a cross country fault?
I do not think so. I did not have much experience with wattmetric systems.
The utility I worked for had a couple of systems but they were not well
though of. Our approach for ASC systems looked at zero sequence harmonic
current flows.