30A fuse trip, compressor starter 2

[JLuc]

Hi all,

we've been asked to investigate fuse tripping on a compressor starter:

Compressor nameplate:
RLA: 16.8A
LRA: 84 A

Fuse:
Low-Peak
class CC
current limiting
LP-CC-30 (30A)

Contactor:
Siemens
42BF35AJ
FLA:30 A
LRA:120 A
15HP 3ph


First thing we did is hook up a power recorder to the starter, here's what we found (see atached file, with max&min rms current for 1 minute intervals);

-compressor seems to start at least 30 times per hour, for 12 consecutives hours (starting 2008-10-19 @ 17:04:33).

-see blown fuse seems to happen : 2008-10-20 @ 04:44:33

-average RLA: 12 A
-average LRA: 60 A

MY QUESTION IS: can fuse operation be caused by too many consecutives starts? I2t making fuse operate maybe?

We also observed contactor wearing very fast...any comments on that?

Thank you

 

[itsmoked]

Yes fuse failure due to consecutive starts can occur. In your case it is not the problem.

The most likely problem with a compressor is the unloader malfunctioning. This is about 100x more likely than anything else.

Next is short cycling. That appears to be the problem you are facing. Every 2 minutes is BAD verging on the ridiculous.

The air consumption has exceeded the start duty cycle of the compressor. Has a new process been added or has a leak developed? One of the two is causing it.

If there is absolutely nothing wrong other than a viable increase in air demand, then you need to do one of two things.

1) Add a large receiver tank. This causes the compressor to run longer between shut-offs dropping the start duty dramatically.

2) Switch the unloader to a constant speed unloader. This causes the compressor to stop compressing air but allows the motor to continuously run. The motor will love this as it will cool down during the unloaded phase.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[dpc]

JLuc,

Yes, pre-loading of the fuse could be an issue when there are so many starts per hour. Basically the published fuse curves are based on an assumed starting temperature for the fuse. If the fuse starts warmer, it will melt faster.

You also have to wonder how hot the rotor is getting inside the motor if it is not running long enough to dissipate the heat generated in the rotor during each start. That is a lot of hot starts.

Contact wear could definitely be an issue with a large number of starts per hour. Is that an IEC or NEMA-rated contactor?

 

[JLuc]

See attached file:

after 2008-10-20 @ 04:42:33, the current in phase A goes up to 75A when compressor attempt to start.

Question is: is it normal to draw more current on two other phases when attempting to start a motor with a blown fuse on one phase?

Thx

 

[dpc]

If one fuse is blown, a three-phase motor will not start since there is no starting torque. If it blows while the motor is running, the motor will continue to run until the overload trips or the motor burns up.

 

[jraef]

PLS for Keith, I think he summed up the answer to the question well.

HOWEVER, look at the time point just before the fuse cleared, 4:42:33 AM (line 15281). The starting currents up to that point had been very symmetrical, but right before the fuse blew, there was a severe imbalance; 81A on Phase A and 38A on Phase C. I would say your line supply may have had a voltage drop on A phase. The only other possibility I can think of is that this was recorded at the exact moment that the fuse was clearing and picked up the peak current as the arc blew out inside the fuse.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[JLuc]

jraef,

in fact, the recorded data are the max & min value for the entire minute interval.

For each 1 minute interval max & min values for the current, during this minute, are recorded.

JL

 

[burnt2x]

"compressor seems to start at least 30 times per hour, for 12 consecutives hours (starting 2008-10-19 @ 17:04:33)."

That's a hell lot of starts! The compressor I know compressor runs continuously. Only the Loader/Unloader valve controls whether the compressor needs to load up the pressure vessel. That way you do away with high starting currents.

Is this an operational decision decided over by "bean-counters"? (let's save energy by putting-off equipment if not delivering work).

 

[JLuc]

I'm not familiar with the control system at all.

I will certainly point out that fact, that the start frequency is way too high.

From a power quality stand point, everything looks Ok.

I will ask them to modify the system and see if fuse still operates and also monitor contactor wear.

JL

 

[jraef]

To get an interesting perspective on the start frequency issue, highlight all rows in the spreadsheet with starts (current spikes over 40A) in red, then set your Excel zoom level to 10%. Scroll up and down, you can see these periods of gross activity separated by long periods of relative smooth operation. I would investigate what transpires in your production floor at those times.

The long periods of steady state operation would indicate a relatively stable system, so there is some machine or process that seems to happen which puts a severe demand on your little compressor for short periods. Right before the fuse blew though, the high demand was almost continuous.

Also interesting to note: AFTER the fuse blew, something was attempting to re-start that compressor exactly every 10 minutes (after a few attempts every 3-5 minutes). That would indicate to me that the compressor control is not just on a pressure demand system, there is some sort of automation involved. Not only that, but whatever it was did not apprently know that the compressor was off line and kept right on trying to start it! That was extremely bad for your motor, because every one of those attempts was a single phasing condition, the motor was heating up very severely for no reason.

You need a thorough review of everything controlling this compressor.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

Ah.. Nice sleuthing Jeff!

Keith Cress
kcress -

http://www.flaminsystems.com