A variety of Screw compressors can be used with modern Inverter Drive technology.
This provides effective, reliable and energy efficient water chiller equipment.
Screw compressors provide a long base operating life easily achieving > 50,000 run hours (20 years plus for an average application), but to obtain this length of service a mid-life bearing overhaul is essential.
For larger capacity water chillers twin shaft Screw compressors now dominate a majority of chiller ranges and are proving increasingly popular where energy efficiency is a key consideration (everywhere they are used presumably!). When applied upon the latest chillers use with Inverter drive technology is providing compelling reasons to use Screw compressors as they are resilient, and can match not only energy performance, but offer a more flexible envelope of operation and are generally of lower capital cost than for instance turbo compressors.
A decade ago when this article was first penned, reciprocating compressors were still dominant throughout the chiller world, but with the phase out of R22 so too has our industry seen the virtual phase out of reciprocating compressors, and when the low reliability and poor energy efficiency is considered this is of little surprise. However, the widely adopted attitude to service with the piston machine was to fix at failure. Whereas much could be done to extend compressor longevity, most reciprocating compressors could be repaired, regardless of the actual nature and extent of a particular failure. Generally about 9 of 10 exchange unit returns received were reparable.
This is not applicable to Screw Compressors!
There are two principle designs of screw compressor in common use, twin rotor, and single rotor / star wheel (single or double star). There are only one or two moving rotors associated with star wheels on the latter type. Additionally there is commonly a slide valve device providing capacity variation. Apart from one major manufacturer, who has put their heart and soul into Gate Rotor Screws, the vast majority of Screw compressors in service now are twin shaft.
Critically the operational clearances between the rotors is extremely small, and herein lays a critical problem when failure does occur.
Screw compressors run much faster than equivalent reciprocating type compressors, typically the drive motor runs just below 3,000 rpm on the UK 50 Hz supply, twice the speed of most reciprocating compressors, although with modern use of inverters peak speeds can exceed 4,000 rpm. When a mechanical failure occurs much particulate debris is produced before the machine finally comes to rest when a safety device operates, e.g. overload or seizure. Due to the very effective pumping action of the screw rotor profiles this debris is dragged and gouged through the Screw mechanism, and it’s tight clearances, creates more debris and damage in the process. This effect renders the machine very rapidly beyond effective or economic repair. As many as 3 out of 5 screws we receive that have suffered failure are rendered beyond repair.
With far fewer machines present in the market, most models are not made with redundant castings, thus the only available option is the high expense of a new compressor. Inevitably failure will occur at the most inconvenient moment, leading to an urgent and even more expensive repair solution
What can be done?
Firstly we must consider the nature of the failures experienced. Screw compressors are generally very robust compressors, and will routinely suffer external problems that would destroy the best reciprocating models. Thus it is necessary to look further into the actual construction of the Screw compressor to find the principle problem.
In order to achieve the high operational speed and close running tolerances, plain bearings are simply not suitable, except on larger capacity industrial machines. Additionally there is a heavy bias for the screw compressor rotors to pull toward the direction of the suction gas entry point, thus the designer makes considerable provision to oppose this effect. In virtually all makes and models of twin rotor type this thrust is resisted by “Angular Contact Ball Race Bearings”.
Rolling element bearings are subject to a finite fatigue life measured in billions of rotations, and whereas a billion sounds a lot, when divided by the operational speed of each revolving element within the bearing race, this actually resolves merely into a few tens of thousand hours operation.
Probably the commonest Screw compressor found in the UK air-conditioning water chiller market is the SRM type Screw compressor. These compressors are certainly tough and very well designed. They are also quiet and power efficient. But once they have accumulated ~25,000 run hours the bearings are on borrowed time.
Bearing Service Interval
Most manufacturers whi make service recommendations set down between 20 & 30,000 run hours as bearing overhaul interval for these compressors based upon the running time of the machine, and it’s application. Running time reduces with more arduous usage i.e. higher head operating pressure.
In our experience bearing failure of these compressors is commonly linked to other factors ...
- 1: Copper plating. This is prevalent in over 70 % of compressors we examine. If allowed to build up the generated copper debris interferes with the close running tolerances of the compressor bearings and rotors and leads to seizure. This problem is caused by trace quantities of moisture entrained within the fridge circuit, leading to electrolytic deposition of copper ions from the associated system pipework. In reciprocating compressors copper plate is generally an indication of system malaise, but rarely actually causes compressor failure. However, the copper plating severely impairs the close running tolerances required by the screw compressor, leading to binding or excessive loading upon the rolling element bearings.
- 2: Light load running. Whereas these compressors will tire more quickly at high head pressures, compressors that run consistently at low (slide valve) load settings generally display more internal wear than fully loaded machines, thus we recommend staging of these compressors to allow full load running wherever feasible, obviously without pushing the head pressure through the ceiling.In this regard use of Inverters improves machine longevity, because the lubrication becomes more effective at lower operating frequencies, and over a typical 25,000 run cycle the machine speed will average out below the standard 50 hz to which a non inverter compressor is assigned.
- 3: Spalling of the bearing. Rolling element bearings do not actually wear in the conventional sense. Instead bearing failure is caused by a fatigue condition known as spalling. The rolling element is basically a malleable core with a case hardened shell. As each element (ball or roller) rotates under load it is microscopically squeezed out of shape by the applied loading. This effectively flexes the inner material of the rolling element. Like all metal materials under a flexing condition, below it’s ultimate tensile strength, will be subject to fatigue, and after many millions of revolutions small fissures form beneath the hardened surface at the interface with the more malleable core. Eventually these fissures allow a piece of the surface material to flake off, and this is referred to as a spall. Once the spall has occurred two immediate effects occur. The element can no longer roll smoothly and this causes massive build up of friction heat, accelerating the failure, plus the generated particles foul the close operating clearance of the bearing, leading to complete failure of the bearing.
Critical to Screw compressors are the thrust bearings which bear the relatively high load caused by the natural tendency for the screw rotors to pull towards the direction of the incoming suction gases. If this control of thrust is lost the immediate effect is a catastrophic failure as the screw rotor effectively collides with the stationary housing at the suction end, leading to secondary seizure and overheat damage. It is this secondary damage that renders the compressor irreparable. In many cases it is not feasible even to dismantle the machine for detailed inspection.
The key point is that had the machine been stopped at any point prior to bearing failure for routine renewal of the bearings, the secondary damage will be avoided. The actual time period between initial spall and total failure can also be very brief and upon machinery expected to operate in a remote condition there is little chance for an operator to stop the machine. No conventional safety device exists to detect the actual failure in the early stages, thus the machine relies upon other safety devices to operate, e.g. fuses, thermal overloads, or circuit breaker.
Of the above High Hours running as the principle attributable problem to Spalling is probably the most common and certainly the most preventable failure. Thus periodic replacement of the rolling element bearings will extend life and whilst the compressor is dismantled any other gathering problems can be identified and corrected before further problems develop.
In Situ Overhaul Method
Replacing the rolling element bearings requires a good deal of direct experience, and a high standard of factory cleanliness, but is not a technically complex task. Providing the work is instigated prior to any signs of distress, some smaller machines can be overhauled without removing the compressor from site, or even in many cases from it’s bedplate.
It is most important to note that site inconvenience by using the ‘In situ’ method for Screw compressors is minimised, and providing no other damage is found within the compressor then complete removal is obviated. This aspect is a major advantage for the ‘In situ’ method and offers additionally substantial economy in the job overall, particularly when the compressor is sited on a roof or in an awkward plant room.
Not all Screw models are suited to the “In Situ” method, but even so prudent maintenance planning to include provision of replacement bearings either on or off site cannot be understated in terms of cost effectiveness and longevity of the machine itself.
Inverter & R134a Conversion
Inverter conversion has also been extensively proven to be effective for improved reliability and control and in particular improved energy efficiency. Inverter control allows capacity cvariation by simply changing the running speed of the rotors, rather than the complex hydraulic control of a slide valve. Slide valves can be troublesome, and all offer efficiency reductions when not at the 100% load point – with inverters they are held at this load point.
With the recent F Gas Directive update all HFCs are now subject to a gradual majority phase out over the next 15 years, and the impact will be highest upon those refrigerants with a higher Global Warming Potential (GWP). For water chillers this means many thousands of water chillers designed for (and in many cases still in use upon) R22. R407C was the natural successor to R22 as a straight application replacement, but is less energy efficient than R134a. R134a is the only HFC likely to remain commercially available muc after 2025, and the only where effective non HFC replacements gases will be available and proven for field applications.
R134a requires a higher specific volume of refrigerant to be displaced per unit of cooling energy transfer, but conversely requires less energy per same volume. With Slide valve control this meant larger compressors, but once an inverter is applied compressor size does not have to increase, as it can be sped up as well as slowed down, typically to ~70 hz or 4,000 rpm.
Market presence of Screw Compressors
From 10% in 2003, Screw Compressors now represent approximately 75% of the compressor throughput at ThermaCom Ltd. Makes commonly overhauled include: Bitzer, Hanbell, Hitachi, Trane, Kobe, Daikin, Stal, Refcomp, Hall, McQuay, Carrier, Sabroe and York.
If you require further information upon the techniques available to reduce maintenance and renewal costs upon these machines then why not give our technical staff a call.
©Trevor Dann - ThermOzone Ltd