offsite primary injection

Status
Not open for further replies.
ryan_618

ryan_618

Senior Member
Location
Salt Lake City, Utah
On an inspection today I requested the reults of a GFPE test required by 230.95(C). To my surprise, I was told that the testing lab removed the breaker, took it off site, and performed the test in their lab. This was because their test set wouldn't provide enough current for even the lowest setting of the breaker. I rejected this test, as I believe it must be tested while installed, per the "first installed on site" language of 230.95(C). Of course, the testing lab says I am an idiot, and that they have never been made to do that.

Any thoughts?
 
Ryan, was the test "conducted in accordance with the instructions that shall be provided with the equipment"?

I have never had an AHJ ask for the written results of these tests.

What bothers me about this and genny load tests, if I'm not on site when some other party messes with my properly torqued lugs, it may require a building shut down on a weekend for me to check this. PITA!

And now that I think about it, this is a 1000A 480V or higher breaker. Service rated. How did the testing co. remove it without a POCO shut down?
 
chris kennedy said:
And now that I think about it, this is a 1000A 480V or higher breaker. Service rated. How did the testing co. remove it without a POCO shut down?
Click to expand...

I require the test results before I release power clearance to the utility. I found in the past that if I release the power, it was too dificult to get the test results. Also, the hazard that we are trying to address with the GFPE is already there if you energize before testing.
 
ryan_618 said:
I require the test results before I release power clearance to the utility. I found in the past that if I release the power, it was too dificult to get the test results.
Click to expand...
I will be following this thread with great intrest. I did my first one of these mabey 2 years ago. I now have 4 under my belt. At the time the first was wired, upon my request for the test I was told by the EE to adjust the breaker at these levels and the test was not required. When I told this to the owner of my company, he said what are you talking about.

Here in FLA. where we worry about derating NM in attics, No one gives a crap(or just don't understand) how the testing and coordination of these breakers effect the safety of these systems.

Sorry about the rant. Your the first person I've seen that has been concerned about this. My inquiries fell on deaf ears.
 
Thank goodness some took a shot at me. I feel better now.

I reread that and don't see what part you didn't understand.
 
About 6 months ago, I asked to be witness to the test for a "double banger" 4000 amp service test. The technician could not believe that I wanted to witness it, let alone that I was asking for the test. I believe that I as an inspector have the right to not only ask for the test, but to witness it as well.

The technician actually spent quite a bit of time explaining to me what it was all about after he realized I was human just like him.

A couple of years ago, while witnessing a test, the tech had problems with the breaker, and the manufacturer replaced it, of course after much wrangling of who was actually responsible for why the breake did not work properly.
 
Several (many) years ago, I did a lot of ground fault testing so services could be energized. Several local municipalities had required documented testing of ground fault protection prior to the test requirement being added to the NEC. Our NETA company kept track over the years of good/bad test results. If I remember correctly, the failure rate atarted around 75% then dropped and stayed around 40%, even as local electrician's learned to install the systems correctly. Almost half of the ground fault protections were inoperative or operated incorrectly (false trips) as originally installed.

The most common problems were improper installation of the neutral bonding jumper, incorrect wiring or polarity on neutral CT's and multiple grounds on neutrals. Double-ended switchgear always seemed to have something wired or installed incorrectly (most of the time by the factory). Note that none of these problems can be found by pushing the "TEST" pushbutton on the GF relay.

On draw-out switchgear, I have pulled a main breaker and taken it to the shop for full testing of the breaker mounted trip units. But we would also do a low current test of the rest of the installation to make sure the neutral CT's had the right polarity, the neutral was clear of any grounds, the trip circuit power supply was adequate to trip on 277V power and several other checks and tests. (For reference on suggested tests look up the NETA testing specifications for ground fault systems.)

I wouldn't deny a test just because the test lab pulled the breaker and ran it back to the shop. But I would ask to see the documentation that the rest of the system was also verified.
 
The ground fault SYSTEM needs to be tested. Not the just the sensor, not just the relay, not just the shunt trip mechanism but the entire system which includes the presence of any improper neutral to ground bond at the facility.

Ground fault trip settings cannot exceed 1200A, most portable primary injection testors can do that level.
 
We test 100's of GFPE systems every year, I have never had to remove a CB or Bolted pressure switch to test off site (FOR GFP). We do remove CBs to primary injection test at our shop. It is easier to move a few CBs that our 1,200 LB test set.

The typical test set can deliver 3000 amps with bolted connection SOLID bus 6' long. Did one last week we had a current issue I ended up using 8 pieces of 535 KCMIL to pass current through a CT mounted in a NASTY location.

We supply detailed reports for all test we perform.

Current pick up at all factory presets for current.
Timed operation at as found or engineered recommended setting for current and time.
If there is more than one CT, Test all CT's at at least one current setting.
Megger line load bus. line to load and neutral to ground with disconnect link removed.
Micro-ohm/ductor/DLOR the power contacts.
Note any ANY code issues.
Any one that wants to watch is allowed, just do not hold me up. But the inspection department should have a decent relationship. understanding or trust with the testing companies. IMO (and this benefits me of course) a DETAILED test report should suffice.


As for Chris's complaint, Please think to call the testing company prior energizing (you may not do this but LOADS OF CONTRACTORS DO, some areas give a TPF without the testing being completed A MISTAKE IMO) as for loosening of connections this is VERY RARE.
 
Last edited:
What 230.95(C) requires is a ?performance test.? It does not tell us what constitutes that type of test. Rather, it defers to the manufacturer, and requires that the manufacturer?s instructions be followed in performing the performance test.

My view is that the ability of a breaker to trip at its design setpoint(s) is something to be tested in the lab. But I also believe that the manufacturer is responsible for creating a ?performance test? that can be conducted on site, a test that verifies the performance of the entire SYSTEM, as Jim pointed out.

So yes, Ryan, I agree with your call. The test performed by the lab is not a bad thing. But it is not the ?performance test? that is called for in 230.95(C). That brings us back to Chris? first comment.
chris kennedy said:
Ryan, was the test "conducted in accordance with the instructions that shall be provided with the equipment"?
Click to expand...
 
The instructions are somewhat limited in scope, I have had several defective GFP relays and due to warranty issues I had to talk with manufactures engineers. MOST HAD NO CLUE. Have also had several on site to witness or testing methods...


IMO If you do not pass current through the CT's, and do not trip the Switch or CB, you are missing the point of testing. Simply pushing the TEST button does prove it will trip, but not necessarily at the value supplied in a coordination study or at recommended values of the engineer.
 
ryan_618 said:
On an inspection today I requested the reults of a GFPE test required by 230.95(C). To my surprise, I was told that the testing lab removed the breaker, took it off site, and performed the test in their lab. This was because their test set wouldn't provide enough current for even the lowest setting of the breaker. I rejected this test, as I believe it must be tested while installed, per the "first installed on site" language of 230.95(C). Of course, the testing lab says I am an idiot, and that they have never been made to do that.

Any thoughts?
Click to expand...

Primary injection testing is a remnant of electromechanical relays.

With current generation electronic relays from a reputable manufacturer the primary injection test is an overkill and a waste of money and resources. If you are not intimately familar with the relay, consult witrh the manufacturer as to what self diagnostics do they perform, where is the break in the digital/analog circuit, etc. Of course check the CT and it's proper rating. Ask the factory for certified tested CT's and PT's and you will not have to worry about calibrating those, ever again. You electronic realy should detect an open sensor. The analog to digital conversion usually occur right at the sensor input, and after that point the circuit is completely digital and self-diagnosing. Earlier generations of electronic protective relays have contained much more extensive analog ciruits that were subject to thermal aging, therefore calibration would "drift".

When your hired testing company insist in PI testing of your electronic protective relays, they ususally just trying to make money. Hav'em splain themselves....

Of course all the above presumes you know what you're doing....:grin:
 
jim dungar said:
The ground fault SYSTEM needs to be tested. Not the just the sensor, not just the relay, not just the shunt trip mechanism but the entire system which includes the presence of any improper neutral to ground bond at the facility.

Ground fault trip settings cannot exceed 1200A, most portable primary injection testors can do that level.
Click to expand...

Thank you Jim.

We require testing here too and so do most of the local jurisdictions. It should not be tested until all of the electrical has been completed, though some are and the test report should indicate that.

I also agree with Charlie that 230.95(C) is pretty clear.
 
Last edited:
Ryan, your title of this thread mentions a ?Primary Injection? test. The code article you cited does not. Is there a reason to believe that the required ?Performance Test? can only be achieved by running a ?Primary Injection Test?? What is it that you want to see, in the report of the on-site, after installation test?
 
weressl said:
Primary injection testing is a remnant of electromechanical relays.

With current generation electronic relays from a reputable manufacturer the primary injection test is an overkill and a waste of money and resources. If you are not intimately familar with the relay, consult witrh the manufacturer as to what self diagnostics do they perform, where is the break in the digital/analog circuit, etc. Of course check the CT and it's proper rating. Ask the factory for certified tested CT's and PT's and you will not have to worry about calibrating those, ever again. You electronic realy should detect an open sensor. The analog to digital conversion usually occur right at the sensor input, and after that point the circuit is completely digital and self-diagnosing. Earlier generations of electronic protective relays have contained much more extensive analog ciruits that were subject to thermal aging, therefore calibration would "drift".

When your hired testing company insist in PI testing of your electronic protective relays, they ususally just trying to make money. Hav'em splain themselves....

Of course all the above presumes you know what you're doing....:grin:
Click to expand...

Secondary injection testing of sensor driven electronic relays is not the same as doing performance testing of GF systems per 230.95(C).

230.95(C) requires the ground fault SYSTEM to be performance tested when first installed ON SITE. FPN No.2 reminds you that interconnects and transfer equipment may affect the sensing of a ground fault.

An improper neutral-ground connection downstream of the GF sensor can prevent the GF system from operating correctly. while these improper connections can be found using a megger, they will not be detected by a secondary injection test only. And I would never commission a Main-Tie-Main scheme without testing all possible current paths through all of the multiple sensors.
 
weressl:

Sir you are wrong and you do not know what you are talking about in REGARDS to GFP relays and sensors.

GFP relays DO NOT NECESSARILY DETECT OPEN SENSORS or improper sensors. In the process of replaciing a bad sensor at present all fuunctioned adequatly with the test button with injection testing the relay tripped at a low of 10 amps and a high of 400 amps (factory setting 100-1200)

These relays fail regularly, additionally they do not test with in their parameters, but may function with the TEST BUTTON.

So lets assume we do as you suggest and assume electronic equipment works adequately..

Yeah my PC functions like a top, never had an issue with my I-Pod, haven't seen one person complaining about their I-phone. IF ANYTHING the electronic relays have a equally or higher failure rate that the electro-mechanical relays we worked on in the past (and there were few electro-mechanical relays installed for 480/277 wye GFP protection, some of the early systems utilized ground return method and dash pots to operate bolted pressure switches) .


Of course all the above presumes you know what you're doing
Click to expand...

Besides I have 4 kids two in college, two more on the way, which is the only reason I promote this.
 
Last edited:
charlie b said:
Ryan, your title of this thread mentions a ?Primary Injection? test. The code article you cited does not. Is there a reason to believe that the required ?Performance Test? can only be achieved by running a ?Primary Injection Test?? What is it that you want to see, in the report of the on-site, after installation test?
Click to expand...

Charlie, the title of my thread was "injection" because that is what was performed. I failed the installation because the test does not satisfy 230.95(C).
 
ryan_618 said:
Charlie, the title of my thread was "injection" because that is what was performed. I failed the installation because the test does not satisfy 230.95(C).
Click to expand...
In that case, I agree with your action. Don't expect to get that for free anymore, however. :wink: :grin:
 
Status
Not open for further replies.
Top