RCD Tripping times question

[Moortown89]

A few 30mA RCBO's that i have tested recently are tripping lower at x1 than they are at x5, e.g 13.1 x1 13.8 x5, I know both these results comply but im curious does any one know why the 1 times trip is lower than 5?

 

[Electro-tech]

Are they AC or Type A RCBOs ? Are you testing at the Distribution panel?

 

[Moortown89]

Are they AC or Type A RCBOs ? Are you testing at the Distribution panel?

Type A RCBO, Yes being tested at the DB

 

[newfutile]

This is within error of measurement or maybe the inductance adds a tiny bit that is usually cancelled out by the increased current . In any case we don’t test at 5 x any more.

 

[Moortown89]

This is within error of measurement or maybe the inductance adds a tiny bit that is usually cancelled out by the increased current . In any case we don’t test at 5 x any more.

Sorry should have stated it was for job done in the south of ireland were they still record times 1 and 5 on the test sheet

 

[newfutile]

I take it they don’t use BS7671 there ?
in any case i would guess coil rise time is providing a slight delay which is not offset by the extra current flow , its only a guess though only the manufacturer would really know why, are you testing just the device as a type AC with all loads disconnected? , possible load capacitance or inductance might play a part.

 

[Risteard]

This is within error of measurement or maybe the inductance adds a tiny bit that is usually cancelled out by the increased current . In any case we don’t test at 5 x any more.

Just to be pedantic I would suggest that we don't record the 5* test anymore, but aren't precluded from carrying it out. Although if it doesn't disconnect within 40mS @ 5 I delta n this may not necessarily be a failure.

 

[newfutile]

GN3 says that too many test will shorten the life span of the device and to counter that a 5 times test is discouraged,however I’m not sure I believe that ,most rcds I come across suffer from stiction .

 

[pc1966]

I can't accept the "reduces life" argument either. I'm not sure what the official standards say, but the Hager catalogue mentions "Mechanical life: 20,000 operations" for one of their RCBOs. Sure it will be slightly less on-load, but when testing RCD trip times that is usually on no (or very little) load.

If you assumed el-crapo brand has a life of 1000 operations, and your full test has 5 trips (one on test button, then four covering In and 5*In on 0/180 deg) that is 200 tests. How many times per year do you check things again?

As @newfutile says, I am more worried about devices sticking because they never get cycled!

 

[newfutile]

We have a rcd test socket for our monthly calibrations ,we have 2 electricians doing at least 4 trips each per month for the last 5 years , it still trips at 19.6 ms every time.
that’s still only 480 operations.
if we assume 1 test button test every 6 months ,plus 4 every 5 years (in out tenanted properties here) 14 operations every 5 years don’t seem that excessive to me.

 

[timhoward]

It also seems illogical that a device that can routinely handle many amps would actually care about 150mA going somewhere else and I agree that this can't really make a material difference to device life.
Surely the only metric that is relevant is the total number of operations.

I did however watch a video suggesting that the 1x and 5x at 0 and 180 degrees were basically manufacturing design standards and these got adopted for routine in-service testing without much consideration.
At the end of the day the device will trip at a certain threshold or not, and the required time for this incorporates several waves at 50Hz so maybe it is fair to say that testing at both 0 and 180 degrees has always been a bit over-zealous.

It's certainly quicker to go along a row at x1/2 and check they don't trip, and go along at x1 and check they do, and then check the test buttons, and the lesser disruption during commercial EICRs can be helpful!

 

[JulianO]

I agree with that but I found as well the reason behind tests that can be done on a specific RCD and results could give you a better resolution of the behavior while the RCD is tripping. I do not think testing an RCD tripping time should be based on simplest (as simple as possible..) test; all this probably with the rational of saving time. The tester (the electrician) himself should have his own INPUT on the test results that appears in the front of his eyes as different values within factory specs will make him judge the behavior of the RCD..
The tester in your hand it is just a calibrated tool; you as a tester should appreciate whether or not the RCD is still within reliable disconnection times..

 

[pc1966]

My MFT auto mode will do the 5*In anyway so easier to just keep resetting until test is done...

As an aside, I have seen different trip times with 0/180 deg, and most oddly was some RCD FCU that on type A waveform were tripping in one direction at 0.5*In, though I suspect they were really type AC as pretty old.