Zero Sequence CT

Status
Not open for further replies.
B

buffalonymann

Senior Member
Location
NC
I am designing a ground fault interruption system for our process; the load is strictly heating elements. Our control system employs shunt trip circuit breakers to supply energy to the heaters. My design is to use the ZCT, and when there is a ground fault, I will trip the shunt to open the CB. I'd like to trip the CB at no more than 100 milliamp ground fault so with a 50:5 ZCT my secondary current will be 10 milliamp. I plan on using a Modicon 140HLI34000 high speed interrupt module to receive the 10 milliamp signal and then trip the shunt of the breaker.

Any thoughts?
 
ron

ron

Senior Member
Location
New York, 40.7514,-73.9925
Depending on the environment, it might be worth wrapping a CT around the conductors that you plan to do this zero sequence measurement and see if making mA measurements is practical. You need a pretty sterile environment to not false trip at 10mA or even 100mA. Industrial environments often have lots of stray currents running around.
 
B

buffalonymann

Senior Member
Location
NC
ron said:
Depending on the environment, it might be worth wrapping a CT around the conductors that you plan to do this zero sequence measurement and see if making mA measurements is practical. You need a pretty sterile environment to not false trip at 10mA or even 100mA. Industrial environments often have lots of stray currents running around.
Click to expand...

The stray currents would be ground faults, and that's what I'm seeking to detect
 
I

Ingenieur

Senior Member
Location
Earth
it'll be more than 100k with Benders
better to spend more on something that works and is rated for such use

the burden on your input is too high for a conventional ct
how do you planninputting it? i, v, with a resistor to convert to voltage
the equipment is not rated as an epd

where did you read a 50:5 0 seq ct is accurate to <0.2% x fs?
the response time will never be fast enough
it will trip no where near 100 mA
the Bender will trip everytime at ~85-90 mA
I assume you do not have an impedance grounded system
a zero seq is not well suited for a solid grounded system
better to wrap the egc
 
I

Ingenieur

Senior Member
Location
Earth
this may help
https://w3.usa.siemens.com/us/inter...ocs_MV/TechTopics/ANSI_MV_TechTopics98_EN.pdf

100 mA prim (10 sec) will produce no output on a typ 0 seq relay


In the ungrounded system, ground current is limited to the system capacitive charging current of the cables, commonly called 3 I0, where the capacitive charging current of each phase is typically 1-2 A and total ground current typically 4-5 A. This current is too low to be sensed with current transformers, so a system using voltage transformers is needed.
 
ATSman

ATSman

ATSman
Location
San Francisco Bay Area
Occupation
Electrical Engineer/ Electrical Testing & Controls
Why re-invent the wheel? Why can't you use a GF system made by any of the major manufacturers?
They contain all the necessary components, test panel, GFR, ZCT and have the necessary listings and warranty.
 
B

buffalonymann

Senior Member
Location
NC
Ingenieur said:
it'll be more than 100k with Benders
better to spend more on something that works and is rated for such use

the burden on your input is too high for a conventional ct
how do you planninputting it? i, v, with a resistor to convert to voltage
the equipment is not rated as an epd

where did you read a 50:5 0 seq ct is accurate to <0.2% x fs?
the response time will never be fast enough
it will trip no where near 100 mA
the Bender will trip everytime at ~85-90 mA
I assume you do not have an impedance grounded system
a zero seq is not well suited for a solid grounded system
better to wrap the egc
Click to expand...

epd?

According to the literature the Modicon module turns on at 2 to 8 mA within 30uS

fs?

am I mistaken to understand that a 100mA primary will produce a 10mA secondary current?

solid grounded wye

tell me why the ZCT is not well suited for solid grounded system
 
B

buffalonymann

Senior Member
Location
NC
Ingenieur said:
this may help
https://w3.usa.siemens.com/us/inter...ocs_MV/TechTopics/ANSI_MV_TechTopics98_EN.pdf

100 mA prim (10 sec) will produce no output on a typ 0 seq relay


In the ungrounded system, ground current is limited to the system capacitive charging current of the cables, commonly called 3 I0, where the capacitive charging current of each phase is typically 1-2 A and total ground current typically 4-5 A. This current is too low to be sensed with current transformers, so a system using voltage transformers is needed.
Click to expand...

Trying to understand what you're telling me here -

Are you saying 100mA current in the primary of a ZCT will not produce 10mA in the secondary?
I read that in low voltage system, the charging current need not be considered.
 
B

buffalonymann

Senior Member
Location
NC
ATSman said:
Why re-invent the wheel? Why can't you use a GF system made by any of the major manufacturers?
They contain all the necessary components, test panel, GFR, ZCT and have the necessary listings and warranty.
Click to expand...


$$$$ is the only reason.

Not reinventing - I researched ground fault apparatus and learned they make CTs specifically made for sensing ground faults. I already have shunt trip breakers connected to a PLC, seems like a no-brainer to use the ZCT to sense ground faults and with the right interface trip the CB
 
ron

ron

Senior Member
Location
New York, 40.7514,-73.9925
buffalonymann said:
$$$$ is the only reason.

Not reinventing - I researched ground fault apparatus and learned they make CTs specifically made for sensing ground faults. I already have shunt trip breakers connected to a PLC, seems like a no-brainer to use the ZCT to sense ground faults and with the right interface trip the CB
Click to expand...
It is only a problem if you buy a bunch of these things and it doesn't work.

A 50:5A CT isn't accurate at that very low % of full scale.
 
I

Ingenieur

Senior Member
Location
Earth
this is why this place gets weird
you ask for feedback (when what you really want is validation)
you get it contrary to your method and then you argue
if you knew you would not ask
it takes the same skill to recognize the correct answer as to derive it

over the last 3 years I've spent considerable time researching this very subject
gf protection is required by law on all mining circuits
std was 4-6 A using 50:5 0 seq ct and relay
I rewrote the law to a x00 mA level, usually 100-125
to save cost we looked at reusing the 50:5 with a new lower value relay
the ct would not function at that level
needed to get to >2-3 A, and then poor accuracy
one mfg took the ct and input it into a xfmr to increase i
some tried a 5:5 ct, we use a 15 A ngr
most just use the Bender or Littelfuse

do a test
set up one ct and run test i thru it, start at 0 and ramp up to 100 mA
won't be totally valid since in use you may have 10-100's of amps flowing when the fault occurs
 
kingpb

kingpb

Senior Member
Location
SE USA as far as you can go
Occupation
Engineer, Registered
Ingenieur said:
this is why this place gets weird you ask for feedback (when what you really want is validation) you get it contrary to your method and then you argue, if you knew you would not ask
it takes the same skill to recognize the correct answer as to derive it
Click to expand...

I was thinking the same thing. :thumbsup:
 
I

Ingenieur

Senior Member
Location
Earth
ron said:
It is only a problem if you buy a bunch of these things and it doesn't work.

A 50:5A CT isn't accurate at that very low % of full scale.
Click to expand...

bingo

the plc is not rated as an epd
ct into a plc input, sw crunching, plc output to shuntrip
sounds like a liability if something goes awry, if it worked
 
B

buffalonymann

Senior Member
Location
NC
I asked for feedback and I have to weed out the feedback from people who have no clue what they're talking about. Its been my experience those who get upset because somebody ask them to explain their position usually get upset because they don't really understand the subject matter.

Of course I realize fault currents may exceed 100mA.

If you can provide feedback without complaining that I'm questioning your responses; I'm interested. If your just trying to sound intelligent and really have no clue regarding this subject, then I'm not interested.

I'm seeking somebody's real experience in lab work or field work dealing with this subject
 
I

Ingenieur

Senior Member
Location
Earth
buffalonymann said:
I asked for feedback and I have to weed out the feedback from people who have no clue what they're talking about. Its been my experience those who get upset because somebody ask them to explain their position usually get upset because they don't really understand the subject matter.

Of course I realize fault currents may exceed 100mA.

If you can provide feedback without complaining that I'm questioning your responses; I'm interested. If your just trying to sound intelligent and really have no clue regarding this subject, then I'm not interested.

I'm seeking somebody's real experience in lab work or field work dealing with this subject
Click to expand...

no one is upset, just amused
you ask, we respond, we are wrong and you stubbornly defend your position
and you toss in passive agressive back handed insults for good measure
do as you like, it will work fine, seriously, let us know how it works out
it is you who do not understand the subject matter, if you did you would not ask
do you know what symmetrical components are? the 0 seq is one of them

I have 30 years experience dealing with sgf (sensitive and nml gnd fault protection)
been required in mining and oil and gas forever
just wrote a federal law on it

these guys are industry leaders in the subject
see if they make a 2 wire passive ct rated for 0.1 A
http://m.littelfuse.com/~/media/fil...es/111-groundfault-ct-selection-flowchart.pdf

I strongly suggest you hire an engineer in this matter
200 ckts
material/install
$250k-$300k or more
the ct needs wired to a relay and the relay to the cb
forget the plc, there is a reason it's never done that way
 
Status
Not open for further replies.
Top