PART LOAD OPERATION

PART LOAD OPERATION ISSUES:

One of the very strong features of the Turbocor compressor is its ability to operate at part load conditions. This is particularly true in multiple compressor units where, as the load reduces, one or more compressors can be staged off.

It is always preferable for the Turbocor compressors to run part loaded as much as possible as opposed to cycling on and off. When reviewing operating logs, anytime it is noticed that one or more compressors are cycling more than once per hour, attention should be given to what is causing this cycling.

Also, in multiple compressor units, attention should be given as to the part load operating characteristics of the various compressors in the system such that they all operate the same under part load conditions..

The most difficult situation for part load operation is when the system has high discharge pressures. Normally, part load operation in strict A/C applications will be during night time hours when the discharge pressure will be lower, due to lower outside ambient temperatures.

In cases where a low load condition exists on a single compressor unit at the same time that discharge ´pressures are high, the compressor will be typically near the surge limit of the compressor and, particularly if cycling occurs, may enter into surge while cycling off.

Their is also the circumstance where load on the chilled water system is suddenly reduced and the chiller controller holds the compressor on at very low load/ chilled water temperature due to delays built into the chiller control algorithm. During this time, the compressor may enter surge.

The parameters of the chiller control algorithm are defaulted to values that serve for the majority of installations. However, in the case of an installations subject to sudden changes in the chilled water loop and/or part load operation combined with high discharge pressure attention must be given to the modifying these parameters to prevent the compressor from going into surge during these conditions. There also may the possibility to put an offset into the controls such that the minimum rpm of the compressor is set at a fixed offset from the calculated minimum rpm (surge limit) for the compressors

In multiple compressor units, particularly an older unit where changes have been made to one or more compressors, it is necessary to look at the BMCC program revision level. Different BMCC programs have different part load operating characteristics.

One compressor may run fine at part load conditions for that particular installation, while another tends to cycle.

In the case of multiple compressor units the optimal solution is to make sure all compressors have the same BMCC program revision. If this is not practical, it may be possible to make the compressor that operates well at part load conditions, the permanent lead compressor. This means that when the chiller unloads, this particular compressor will always be the last compressor running.

Given that, mechanically, the Turbocor compressors do not experience ¨wear¨, there is no need to be overly concerned about equalizing run hours between compressors. In fact, it would be better to run the compressor that does not cycle, always as the last compressor running rather than put a different compressor having lower run hours online when that compressor is going to cycle.

In the case of multiple compressor units with all compressors having the same BMCC revision level and one of the compressors runs differently under part load conditions, investigation must be made as to why this difference exists.

SURGE - WHAT CAUSES IT - HOW TO PREVENT

PREVENTING SURGE:

Surge typically occurs when the compressor is operating with a high pressure ratio (PR), or when either the suction or discharge pressure change more rapidly than the compressor controls can adjust the compressor speed (RPM´s) and/ or inlet guide vane (IGV) position.

Pressure Ratio (PR) is a number that is calculated based on the relationship between the discharge pressure and the suction pressure. Pressure Ratio for water cooled chillers typically will be on the order of 2.0 to 2.5. Due to the higher discharge pressure experienced by Air Cooled units, Pressure Ratio for Air cooled chillers will typically be on the order of 2.5 to 3.0. Higher condenser water temperature or higher ambient temperatures will result in higher Pressure Ratio.

It is important to recognize that the compressor is experiencing surge and take action to prevent this from occurring. This action may be the repair of a defective component of the chiller or an adjustment to the chiller control loop algorithm parameters.

Routine review of the chiller controller logs and the compressor histories is important to assist in recognizing that the compressor is entering surge.

Anytime there is a Front Radial Bearing Displacement Fault, the cause must be investigated. Particularly if this Fault is displayed repeatedly.

As an overall rule, the compressor should not cycle more than three to four times per hour under normal operating conditions.

There are many possible causes of surge, however, following are some of the more common ones encountered:

Chilled Water Flow Switch inoperable in the closed position or jumped out – This condition is particularly dangerous for the compressor. If the compressor starts with no chilled water flow or the chilled water flow is reduced/ eliminated during compressor operation, the suction pressure will rapidly drop. The compressor controls can´t adjust rapidly enough to prevent the compressor entering into a surge condition. Eventually, the compressor shaft displacement will exceed the safety limits of the controls due to the surge and the compressor will shut down, typically with a Front Radial Bearing Displacement fault.

The continued operation of the chiller under this condition, with repeated Front Radial Bearing Displacement faults has a high probability of eventually resulting in damage to the touchdown bearing.

Lack of Water Flow proof or inoperative flow proof on condenser of water cooled unit – If there is no water flow proof installed to protect the chiller against operating with low water flow, a condenser pump becoming inoperative will force the compressor into a surge condition.

High Ambient Conditions in combination with lower chilled water setpoint – In cases where the client is operating the chiller at low load with a chilled water setpoint below 4.5 degrees C (40 degrees F.), during high ambient conditions, it is imperative that proper adjustments be made to the chiller control parameters to prevent the compressor entering in surge.

Under these conditions, the compressor will likely be operating close to the surge rpm limit. As the the chiller controls attempt to reduce compressor capacity to meet the low load at the same time maintaining enough compressor rpm to overcome the difference in suction and discharge pressures, surge is a risk.

High Evaporator Approach – High evaporator approach (more than 8 degrees C or 14.4 degrees F.) will result in a lower than normal suction pressure. This, in combination with higher ambient conditions can also result in the compressor operating close to and occassionally entering into surge condition.

Failure of Inverter Temperature Sensor or Motor Cavity Sensor – If either of these sensors fails such that it is giving a higher than normal reading, the compressor controls will automatically reduce compressor rpm in order to limit the amount of heat being generated by the inverter or the compressor motor itself and forcing the chiller into a surge condition.

Chiller Operation with rapidly changing load – In cases where the chilled water loop is extremely small or the load on the chilled water loop suddenly drops off, the compressor will tend to cycle frequently and has risk of entering surge as the load disappears and the compressor attempts to reduce compressor capacity in response to the changing load.

In this case, the compressor may enter surge before the chiller control algorithm reaches the point where the compressor will be cycled off. Repeatedly cycling the compressor under these conditions can result in surge and compressor damage.

Failure of the Compressor Suction Pressure Sensor – The compressor is equipped with a pressure/ temperature sensor on the compressor suction and discharge. If the suction pressure sensor begins to read high, the compressor controls can drive the compressor into surge repeatedly due to the fact that the compressor does not really know where it is operating on it internal compressor map.

It is important to routinely verify the suction pressure sensor accuracy. There is no calibration possible on this sensor. Sensors reading improperly must be replaced.

SURGE - WHAT IS IT?

Historically, the author has worked with technicians from Danfoss Turbocor, Multistack and Smardt, Canada, Smardt Australia and Smardt, Germany. The author is aware that there is a higher than normal percentage of compressor failures at certain sites and has discussed these failures with technicians and engineers from various organizations.

The conclusion drawn from this set of information is that allowing Turbocor to routinely enter Surge condition on a continuous basis will eventually lead to a bearing failure. Occasional Compressor surges can be expected during chilled water pump exchange or unusual operating conditions. However, the routine occurrence of surge is dangerous to the long term health of the compressor.

What is Surge? 
This is the condition in which the compressor enters an operating condition where the differential pressure generated by the centrifugal action of the compressor impeller is not sufficient to overcome the difference between the suction pressure and the compressor discharge pressure. When this occurs, there is an instantaneous reversal of refrigerant flow through the compressor. In the case of Turbocor application, there is a discharge check valve. When this instantaneous inversion occurs, the check valve closes abruptly. When this occurs, the pressure differential between suction and compressor discharge disappears. Flow through the compressor returns to normal, the check valve opens and the original pressure differential returns, causing the cycle to repeat. All of this happens very quickly. This phenomenon is accompanied by a ¨clacking¨ from the check valve in the compressor discharge and vibrations in the discharge line of the compressor. Normally, this Surge condition will last 5 to 10 seconds as the compressor controls adjust compressor speed to get the compressor out of the Surge condition.
Why is this bad for the compressor? When the pulse occurs, there are rapid pressure changes within the compression area on the compressor. Typically, the compressor shaft rotates between 20,000 and 30,000 revolutions per minute. The compressor bearing control acts to keep the shaft precisely located radially and axially. Movement of the shaft at a distance less than the thickness of a sheet of paper during operation may result in the shaft entering contact with the contact touchdown bearings at a high rotational speed. Touchdown bearings are designed to support the shaft once it is completely stopped. The shaft that contacts the landing bearings at normal operating speed has a high chance to destroy the touchdown bearings. The rapid changes in pressure that occur during surge act to shift the axis from its normal operating position. Magnetic bearing control measures shaft position 10,000 times per second and adjusts for any shaft movement. However, under surge conditions, with rapidly changing pressures and related forces on the shaft, there is an internal safety limit to the ―orbital displacement‖ of the shaft. Under Surge conditions, when these internal forces cause shaft displacement from its normal position, if the displacement exceeds the ―orbital displacement‖ limit, the compressor will shut down automatically and normally this will be accompanied by a message ¨FRONT RADIAL BEARING DISPLACEMENT FAULT¨. In certain instances, the internal forces are such that the shaft enters in contact with the touchdown bearing at a high rotation speed, possibly causing and damage the touchdown bearing.

The touchdown bearing is a standard type roller bearing which is part of the magnetic bearing assembly, specifically designed to support the shaft when the shaft is de-levitated. The touchdown bearing is not designed to support contact with the shaft at operating speeds.

Under these circumstances (damaged touchdown bearing), the compressor must be returned to the compressor manufacturer (Danfoss Turbocor) for repair. Repair of the touchdown bearing locally is expressly prohibited by the compressor manufacturer.

PURPOSE OF BLOG

PURPOSE OF BLOG -

As technicians that work on a highly sophisticated and complex compressor, in the opinion of this blogger, we are severely handicapped by the lack of really in depth information in the area of determining causes of failures and indicators of potential failures.

The Danfoss Turbocor training course is aimed at how to locate, isolate and repair a compressor failure. The Turbocor OEM manufacturer´s training courses (such as Multistck, Smardt and Arctic) essentially copy the Danfoss Turbocor training course (in fact, they use many of the same visuals) with added information regarding each manufacturer´s chiller control.

There are two very important areas not addressed by any of these training courses.

First of all, and most important, any time there is a compressor failure, all possible efforts must be made to determine why the failure occurred. This is absolutely critical to prevent repeat failures. These efforts include detailed review of the chiller controller historical logs as well as the faults and events logs from the compressor itself. I personally know of various installations with multiple compressor failures where the cause of the failures has never been addressed.

Secondly, any time we have contractual responsibility for maintenance on a chiller, we, as the responsible technicians, must continually review the operation of the chiller using the same information indicated above, looking for potential causes of future failures. This is extremely important to protect the client against a potential compressor failure.

In my experience, the application engineers and technical support people from Danfoss Turbocor and the various OEM manufacturers have extensive levels of experience and are good reference sources. However, they typically have a great deal of priorities occupying their time and it is difficult to get real time help when you are in the field working on a unit and need assistance. Added to this is the fact that the most experienced engineers are the Danfoss Turbocor application engineers and they are not directly available to the technicians/ engineers working directly with the units. They are only available to the application engineers of the OEMs.

The information we need to properly support the Tubocor Compressor based chillers in the field is only available through experience. There is no reference manual that contains this information.

The purpose of this blog is to begin to record this body of experience in a manner that it is readily available directly to personnel working directly with the Turbocor technology.