Timer Current Curve pickup

[mivey]

Is this the equation you have in mind or am I way off?

That's it. You will see the equation is infinite at exactly 1X.

The logic diagram will show you do not pick up until amps>1X. I'm not sure how many decimal places that is as I've never had a need for that. The rest of the circuit (CTs, etc.) is not exact so it does not concern me if the relay picks up at 1.009X instead of 1.008X, etc.

 

[mbrooke]

That's it. You will see the equation is infinite at exactly 1X.

The logic diagram will show you do not pick up until amps>1X. I'm not sure how many decimal places that is as I've never had a need for that. The rest of the circuit (CTs, etc.) is not exact so it does not concern me if the relay picks up at 1.009X instead of 1.008X, etc.

One amp over is enough for me, especially considering normal load current should not even approach that pickup. Awesome and much thanks for your help

 

[mivey]

Ok- that would make more sense then. It was those curves that were throwing me off. So when I hit 801 amps (primary) from an 800amp (primary) set point, the relay will indeed start timing, albeit it will take some time to produce a trip output.

Assuming the primary CTs and circuitry feeding the SEL faithfully reproduce the primary current: yes.

 

[mivey]

One amp over is enough for me, especially considering normal load current should not even approach that pickup. Awesome and much thanks for your help

No problem.

 

[mbrooke]

The timing starts at the zero (0) second reference. It is not shown. The chart is a 'window' plot, i.e. zero is not depicted one or both axes. For example, you could have an even smaller plot window by hiding the part below .09 seconds and to the left of 1.5x pickup and still have the same data plot.

I owe you an apology, and a standing ovation- you were the first to say what took me 35 posts to grasp Not your fault however, nor have I taken you for granted, rather I am slow to grasp the painfully obvious :ashamed1:

 

[Ingenieur]

Easier said then done. Knowing ahead of time is key before building or commissioning anything.

that is obvious
knowing or spec'ing the equipment is paramount
that is why we have manuals

 

[mbrooke]

that is why we have manuals

Which aren't always clear and concise as they need to be. Nor do they cover all real world applications. Ie, a relay engineer figured out SEL4XX are a perfect solution to the age old main and transfer bus- despite the manual making not one mention of M&T.

 

[Ingenieur]

Which aren't always clear and concise as they need to be. Nor do they cover all real world applications. Ie, a relay engineer figured out SEL4XX are a perfect solution to the age old main and transfer bus- despite the manual making not one mention of M&T.

they tell you the functionality
the engineer determines suuitability for the application

aren't sel manuals like 700 pages?

 

[rian0201]

Easier said then done. Knowing ahead of time is key before building or commissioning anything.

i agree, but one can learn a lot during commissioning or just like the SEL manual says - test if one is unfamiliar with the theory or operation of the relay.

 

[mbrooke]

they tell you the functionality
the engineer determines suuitability for the application

aren't sel manuals like 700 pages?

Yes- the current SEL-421 is 1452 pages and makes not one mention of main and transfer besides a few graphical interface display options. Yet tons of application, examples, and logic programing for ring bus and breaker-and-a half.

But going back to my original incertitude the 351 manual threw me off when the curves stopped right at 125-150%. Which made me question what was really going on as I assumed that they reflected the 51 equations 100%.

 

[rian0201]

Hmm i understand you, but in my experience we normally test the relay upon receipt to verify the declared specifications. Although it has been tested by SEL.

But that would be difficult if you are limited only to design.

Normally the graphs are shown at that values since during the time of electromechanical relays, errors at these values are huge.

Even at some digital relays they recommend to test the OC function at these levels for greater accuracy.

But still, testing should be done if one is unfamiliar with it.

Sent from my vivo 1606 using Tapatalk

 

[mbrooke]

Hmm i understand you, but in my experience we normally test the relay upon receipt to verify the declared specifications. Although it has been tested by SEL.

But that would be difficult if you are limited only to design.

Normally the graphs are shown at that values since during the time of electromechanical relays, errors at these values are huge.

Even at some digital relays they recommend to test the OC function at these levels for greater accuracy.

But still, testing should be done if one is unfamiliar with it.

Sent from my vivo 1606 using Tapatalk

I know- putting a relay in service without testing is asking for outages and equipment damage. Especially when you are using complex Boolean.

As is I can't think of an application where the settings will not be changed past the factory default.

 

[mivey]

Because that is how the old EM curves were drawn and they were duplicating the historic curves.

The EM relays were not considered consistent below 1.5X (even though you test them to within 3% of pickup). The digitals are computer consistent down to the pickup setting.

Looked at some old relay data and some had accuracy ranges lower than 1.5X (one Basler I looked at had stated accuracy starting at 1.3X).

For some ABB relays the time curve test is performed at 1.2X, others at 2X, 5X, etc. So the 1.3X-1.5X is for the low range of the stated accuracy.

The actual pickup is usually within a few percent of 1X (ranges from about 1-2% on low time dials and may be up to like 6% for higher TDs).

So for inverse EM relays you pick up within a few percent of 1X but don't have accuracy guarantees until 1.3X-1.5X. Good for coordination at many times p.u. but not so much near p.u..

The digitals can have logic to make it useful for overload and fault monitoring and don't have the same accuracy issue as EM relays.

 

[mivey]

I know- putting a relay in service without testing is asking for outages and equipment damage. Especially when you are using complex Boolean.

As is I can't think of an application where the settings will not be changed past the factory default.

Integrated wiring and programming is usually where we find problems. Very few failures with the digital relay hardware itself.

 

[mbrooke]

Integrated wiring and programming is usually where we find problems. Very few failures with the digital relay hardware itself.

Yup- hence my point about testing. Most misoperations be it under trip, over trip or wrong output are caused by improper settings and often the reason why its best to have similar relays instead of two separate by two separate manufactures.