Dear Mariano 12345,

Detecting the cause of an occurring damage to an Aerzen compressor requires normally a thorough investigation.

Compressor A:

I understand that you have found "lead like particles of 2-8mm".

First off all, you should determine the true composition of these particles and search in the system, where they can come from.

You state that this is a N2 looped system. Could the particles come from the N2 generator set?

Particles of this size can damage the rotors before the outlet temperature rise is detected and – in case of a switch off – before the compressor has run out.

Compressor C:

It is unclear whether the PTA powder is; Plasma transferred Arc powder or Pure Terethalic Acid powder.

In a looped system, the filtering of the conveying gas is very important and when this filtering fails, damage is imminent.

Slow wear and tear of rotors and cylinder will cause a compressed gas temperature rise over time and should be detected by the temperature sensor in the outlet (if working).

If a filter bursts and a significant amount of material is entering the compressor at once, the rotors can be stalled and damaged (high rotational energy), causing high rotor and housing temperatures, leading to the disintegration of parts.

The outlet temperature sensor is then too late.

From your report, it does seem that the cause is suspected to be somewhere in the system and not considered to be a malfunctioning of the compressor itself (f.i. failing lubrication or bearings).

The found damaged filters, and the wrongly installed guard filter support this.

In that case, it is understandable that the maintenance company cannot be held accountable.

Helping you to find the cause should be the normal attitude in a supplier/customer relationship.

Success

Teus ■

Hello,

As always, Teus is right on the money with his responses. Loop N2 systems can have some interested quirks. I have been involved in root cause failure investigations where dust particulate went through compressors in significant amounts enough to cause localized hot spot on the rotors that then caused metal to metal contact that in turn then leads to more heat introduced into the hot spots and this does not end well in a machine with such close tolerances. It is feasible to assert that this condition was the most probable cause of failure.

There is still a lot of background information missing. For one, such loop systems are typically equipped with so called start up strainers on the compressor inlet. Often they are left in place for the service like of the compressor. Is this strainer still in place and how clean is it? Furthermore, have the compressor instruments been calibrated for accuracy lately and has the permissive contact been checked recently? Check check for the latter is to stop the machine safeley and to pull the coupling pins. Now start the compressor up and after 10 - 15 seconds the compressor motor should shut down. If it does not do that then the instrument panel's permissive contact is not connected to the main motor starter correctly.

One other question I had was why the second compressor had to be overhauled in the first place.

It is odd that your system has operated for so long to have a such an issue all of a sudden. The incorrectly installed filters seem to be a likely culprit, but if I were stuck in your position I would use a fishbone root cause failure diagram and start a conventional 5-Whys analysis. By the fourth Why a much clearer picture emerges and one or more most likely causes will become known.

Regards,

Ralf Weiser ■

Originally Posted by mariano12345

We had an unfortunate incident where two compressors fail in a short time frame. Only in my nightmares I have seen something like this, related to a very reliable machine like a VM 337, Aerzen compressor.

In one of the cases, compressor A failed, and it looks like some metal particles got into the compressor. That is a pretty clear failure. Not so clear where the particles came from. This is a closed loop N2 system for conveying PTA powder to different locations. There are dust collectors in each of the silos receiving the powder, and there is a guard filter at the suction of the 3 compressors (we have A, B and C). The installation has run very reliably for 15 years, and there were no piping jobs since the start up.

The metal pieces in the compressor A looked like lead, sizes from 2 to 8 mm. Completely irregular. The only place that these particles I can think of coming from is the silencer that the Aerzer compressor has in the suction. Internally (in the silencer) I can see a perforated plate and there is some material packed intended for noise reduction. I do not know what is in that pack, but I could not find any material loose or damaged part when introduced my hand.

Damage was rubbing between both lobes.

About 2 weeks later we had a failure in compressor C. This failure didn't look as obvious as compressor A. The damage was rubbing between the lobes and the high pressure side plate. The cold clearance there is normally 0,12 to 0,19 mm and it is the fixed side. The expansion side is the suction and the cold clearance is 0,6+ mm. Well, the damage was on the compression side.

Two points to consider :

* The compressor was recently overhauled and started in line 4 months before the failure. No problems detected. No problems detected by vibration monitoring.

* There was some PTA powder in the compressor. Further inspection found filters damaged, and the guard filter wrongly installed.

My BIG question here is if powder in the compressor can cause additional and localized temperature rise not detected by the high temp protection and causing contact between lobes and compression end plate.

The company that maintains these compressors says that they would expect wear in the long term caused by the powder, but not damage like described. They say that the metal particles hit both compressors. It is really evident in compressor A in fact, but compressor C looks totally different failure pattern. Obviously they don't want to deal with the responsibility.

OK, guys, the question, one more time is >

can the powder in the compressor coming from a failed filter cause additional and localized temperature rise not detected by the high temp protection and causing contact between lobes and compression end plate, leading to a quick failure.

VM20 741,935 wrapping after receiving the serial number after compressor maintenance circuit;

We are using Aerzen VM20 Compresor and we are feeding Pure Theraftalic Acid by N2.

We live so far about the Aerzen compressor VM20 are four events.

1. In 2002, 18 237 operating hours locked compressor stall. It has been blocked by the compressor discharge.

2. In 2011 Locking during start up after the periodical maintenance. The compressor is blocked by suction side.

3. In 2012 result of operational errors during the operation of the lock compressor. The compressor is blocked by suction side.

4. In 2015, Compressor maintenance after commissioning tested without load, with the release on the work load after a day given to lock the motor overcurrent protection after 30 minutes. It has been blocked by the compressor discharge side.

This is insufficient for setting the suction side of the fault during maintenance space dated 2011 when we evaluate four events, operational errors result of the failure is induced suction dated 2012. I can say that very clearly.

The two are very similar in 2002 and 2015 incidents were caused by rubbing on the discharge side of the rotor t on both issues. For 2002, servise expert told that the space magnitude Marble axial bearings presented a report on the possible occurrence of the failure.

In 2015 the Compressors started after overhoul with firm overseeing the inverter as directional test done without the load (the system inverter no power supply of the main system is entered) .We checked compressor protection has been tested and "Oil Pressure Protection" is the exception that the sensor was confirmed that work actively. While connected to the inverter is working in the oil pressure switch the compressor off-site during commissioning with the removal of the remains was confirmed the next day. (This is seen in the system's alarm recording). Oil Pressure load on the compressor has been observed that the circuit after running low protection. In this way it worked about 30-35 minutes to load and PTA dispatch was stopped while locking.

As mentioned in the report of firm expert "reverse rotation".

But our system requiring a requirement and the data is not available to confirm this.

Namely;

• PTA was found to be shipped when the system is switched on.

• The compressor will not be reversed even if the formed lubricating oil pressure pump to be turned upside down and will remain a very short time after the compressor with low oil pressure alarm is disabled. (It did not make any oil pressure alarm)

We believe that in the light of this assessment can not get the working conditions of temperature for 25-30 minutes covered body temperature, when given the compressor load and formed due to insufficient axial play of axial elongation.

Therefore, this defect is sourced maintenance company must provide a solution to this issue.But they aren't accept. ■

Hello erkaya;

without a detailed failure report there will be very little progress in finding the root cause of your compressor issues. Was the repair shop an authorized Aerzen facility? One way or the other there should be a report with photos of the compressor components be available to you. Message me in the event that you are getting this information. Have a great rest of the week.

Originally Posted by erkaya

VM20 741,935 wrapping after receiving the serial number after compressor maintenance circuit;

We are using Aerzen VM20 Compresor and we are feeding Pure Theraftalic Acid by N2.

We live so far about the Aerzen compressor VM20 are four events.

1. In 2002, 18 237 operating hours locked compressor stall. It has been blocked by the compressor discharge.

2. In 2011 Locking during start up after the periodical maintenance. The compressor is blocked by suction side.

3. In 2012 result of operational errors during the operation of the lock compressor. The compressor is blocked by suction side.

4. In 2015, Compressor maintenance after commissioning tested without load, with the release on the work load after a day given to lock the motor overcurrent protection after 30 minutes. It has been blocked by the compressor discharge side.

This is insufficient for setting the suction side of the fault during maintenance space dated 2011 when we evaluate four events, operational errors result of the failure is induced suction dated 2012. I can say that very clearly.

The two are very similar in 2002 and 2015 incidents were caused by rubbing on the discharge side of the rotor t on both issues. For 2002, servise expert told that the space magnitude Marble axial bearings presented a report on the possible occurrence of the failure.

In 2015 the Compressors started after overhoul with firm overseeing the inverter as directional test done without the load (the system inverter no power supply of the main system is entered) .We checked compressor protection has been tested and "Oil Pressure Protection" is the exception that the sensor was confirmed that work actively. While connected to the inverter is working in the oil pressure switch the compressor off-site during commissioning with the removal of the remains was confirmed the next day. (This is seen in the system's alarm recording). Oil Pressure load on the compressor has been observed that the circuit after running low protection. In this way it worked about 30-35 minutes to load and PTA dispatch was stopped while locking.

As mentioned in the report of firm expert "reverse rotation".

But our system requiring a requirement and the data is not available to confirm this.

Namely;

• PTA was found to be shipped when the system is switched on.

• The compressor will not be reversed even if the formed lubricating oil pressure pump to be turned upside down and will remain a very short time after the compressor with low oil pressure alarm is disabled. (It did not make any oil pressure alarm)

We believe that in the light of this assessment can not get the working conditions of temperature for 25-30 minutes covered body temperature, when given the compressor load and formed due to insufficient axial play of axial elongation.

Therefore, this defect is sourced maintenance company must provide a solution to this issue.But they aren't accept.

■