Bi-metallic Overload Trip Curve 3

[nightfox1925]

I have at hand a bi-metallic overload relay trip curve of the NEMA Class 10 type. The graph has both time (in seconds) against Multiple of Pickup Current (P.U.)axes. I will trace the curve on a standard log-log paper with Time (seconds) against Amperes axes.

Is this per-unit values based from Motor FLA as base current such that the Multiple of Pickup Current is equal to the O/L Level divided by motor FLA? Is my understanding correct?

Thanks in advance

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata

 

[gepman]

The bimetallic OL's that I am familiar with have a dial setting. The pickup value would be a multiplier of the dial setting. The OL relay has no way of knowing what the FLA of the motor is, only the setting on the dial of the OL. You can check this on the curve since a Class 10 OL should trip at 10 seconds or less with a 600% value (6.0 multiplier) of the OL rating or setting.

 

[dpc]

There's a specific ampere rating for the heater element in the overload relay. The setting of the overload relay is a percentage of the current rating of the heater element.

The heater is selected based on the motor full load amps or service factor amps, but will generally not be exactly the same.

When plotting TCCs, we generally plot in terms of motor FLA since the exact heater element size is often not known. It's a simple matter to convert from one "base" to the other.

 

[oldfieldguy]

If it's bimetallic, then the heater element is selected based on the FLA of the motor. The manufacturer will have tables to cross reference the heater size to the proper heater element. The dial adjustment allows you to select a point in the range of currents for a specific heater element. The tables usually reference direct correlations for 115% service factors and will tell you to choose the next lower size for 1.0 service factors.

The "Class 10" refers to the seconds to trip at 6X the current rating. You can plug in the time and current of this point into your time-current plot. The "curve" is generally I^2*T, so the 300% trip time will be approximately 40 seconds.

Remember that these are NOT precision devices.

old field guy

 

[waross]

Old field Guy,
"Remember that these are NOT precision devices"
I wish that I had said that. And if the installer bent the heater when installing it,- you get the idea.
respectfully

 

[oldfieldguy]

waross--

I have been amused over the years when some of my clients 'upgraded' medium voltage motor protection from thermal overload blocks and CT's to the latest, bright, shiny microprocessor-based motor protection relays.

Whereas before, with the old protection, everybody was happy if the motor tripped within ten seconds of the expected point, they were suddenly greatly concerned with trip times that didn't match the overload curve within a tenth of a second. And suddenly there's this "current unbalance" thing that had been ignored for thirty years prior. And dozens of other settings that they never needed before, but NOW we had to address them.

That fluttering sound was good common sense flying out the window...

old field guy

 

[ScottyUK]

The current imbalance is a measure of the negative sequence heating taking place in the rotor. It might not matter too much on little motors but it can kill big expensive ones. Just because the old technology didn't detect it doesn't mean it wasn't there.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[burnt2x]

oldfieldguy,

Maybe it's good to qualify when you posted:

And suddenly there's this "current unbalance" thing that had been ignored for thirty years prior.

Some manufacturers have produced thermal overload relays with innovative trip bars that actually takes care of "current imbalance". The bi-metallic elements are arranged to flex and push a common trip bar that if a current imbalance is present, the unequal travel of the bi-metallic strips creates a faster trip! These TOL's I have observed way back 70's, so it's not true that "current imbalance" thing has been ignored for 3 decades!

 

[oldfieldguy]

guys--

Thanks for reading...

I worked with motors from 480-13,800 volts, from fractional to 24,000 HP.

I was referring to medium voltage motors in the <1000 horsepower range, typically protected by thermal overload blocks of the simplest varieties up into the 1980's. Much past that horsepower range, the users went with current balance and negative sequence relays for the protection of the much more expensive motors.

Then along came the 80's and microprocessor relays hit the market. And things got fun...

Don't get me wrong. I'm not discounting the capabilities or needs for advanced protection now that it's available.

old field guy

 

[slavag]

Old field guy.
If I understand you right.
Problem is not new relays, problem is right setting calculation.
Regards.
Slava

 

[oldfieldguy]

Slava--

You're right. When the protection as a thermal relay, it was simple for people who were not necessarily versed in the various protection elements of a motor. They had ONE, choose the right thermal element heater for the motor current. When I upgraded these, then I had wo work with them on many other elements like unbalance, acceleration time, etc.

It's easy for you and I who deal with these things often, but when we're talking about a "maintenance engineer" for a samll petrochemical facility, and he's usually an ME or CE, there's more things to get confused over and to be frightened by.

I had to do a lot of hand-holding and comforting in those cases.

old field guy

 

[slavag]

Olf field guy.
It's common problem.
We use some EXCELL files with MACRO's for thermal image and unbalunce current calculation and with graph's. In many cases it's help also to ME and CE.
Regards.
Slava

 

[nightfox1925]

Thanks for all your inputs. The curve at hand provided to me seems to be a per-unit wherein the slope is the same regardless of heater rating. My motor FLA is 29.7A, S.F=1.15, then by Code the O/L setting max = 1.25 x 29.7= 37.125A and say I set that O/L at 36A. This current is I perceived to be the tip of my O/L curve where it starts to pick up.

In P.U. terms or Multiples 35A is 35/29.7 = 1.178 or 117.8%. Using the O/L P.U. curve for NEMA class 10 O/L relay, I started with 1.78, then 1.25, 1.4, 1.5, 2.0, 2.5, 3.0...8.0 in 1.0 P.U. intervals of O/L (with FLA as base) and get the individual time (seconds) equivalent. In my Log-log paper in terms of current, I simply multiply the P.U. O/L to the FLA and using the equivalent time as calculated earlier, I plotted the curve.

I have no software here to simulate easily, so everything i s done manually...I happen to get a hold of blank log-log graph only. Its like John Rambo in the middile of the forrest.

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata

 

[dpc]

Welcome to the dark ages. We used to have light tables to allow us to stack the vellum TCCs on top of each other and select the best settings.

 

[nightfox1925]

Oh yeah DPC. I'm here in a plant in a middle of nowhere. The poor contractor is setting up the Overload relay which the plant people just decided on site to add with their new motor. The contractor provided the works for them and set the relay to comply with code requirements...then alas, out from nowhere they suddenly told him to make a coordination plot for a 25HP motor. Poor contractor did not have the facility nor a sofware like ETAP to do one. I just made it manually with the motor curve, O/L, MCP, short circuit value, contactor maximum rated breaking current, make current and upstream relay curve. The O/L curve was not the same scale with the loglog paper so, I have to do it the hard way.

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata