when we define a reciprocating compressor as “lubricated”, we mean that it has oil lubrication in its

cylinders. All reciprocating compressors – by inherent necessity – have to be oil lubricated in their
“power train”. Here, the power train is defined as all the running gear inside the crankcase (crankshaft,
main bearings, crosshead, connecting rod, etc.). Normally the power train is lubricated by a lubrication
pump that recirculates the required lubrication oil to the lubricated points, and to a cooler while
pressurizing it.

That leaves three other critical sites that, in the past, always required oil lubrication of some sort:

1. The piston rod packing;

2. The cylinder valves; and,
3. The piston rings within the cylinder.

.All three sites pose potential oil contamination of the compressed gas and, therefore, have been the
subject of scrutiny and safety engineers (due to the contamination and potential for explosion in the
case of oxidizing gases). The application of special materials to the piston rod packing and the cylinder
valves has virtually eliminated the need for lubrication at those sites – if one so desires it. However, the
need for furnishing a non-lubricated piston has not been resolved 100% in some cases. Two types of
resolution are presently available:
1. Use of a “labyrinth seal” design on the piston; and
2. Use of a “soft”, self-lubricated material instead of the customary cast iron piston ring.

Sulzer Brothers in Switzerland was the innovator of the labyrinth seal piston in their vertical,
reciprocating compressors and this has met mixed reviews. As you can imagine, the machining of the
piston in conjunction with the cylinder has to be super-accurate and precise. The clearances between
the piston and the cylinder are microscopic and the piston virtually rides on a “gas” bearing. Any lateral
movement or mis-alignment in a vertical machine causes a potential meeting of both piston metal and
cylinder metal – with bad results. Additionally, as can be expected, there is always some “blow-by” of
gas in the piston. In a double-acting cylinder, this means an inherent loss of capacity or efficiency.\

The use of re-inforced Teflon piston rings (and similar materials) has been more popular in non-
lubricated compressor cylinders. The pistons actually have to be re-designed by using lighter metal
(such as aluminum) for the piston and incorporating “rider bands” on the piston edges. The lighter
piston makes it easer to support it in a true alignment and the rider bands serve as bearings to take some
of the lateral forces away from the Teflon piston rings and allows these to act just as sealers and not as
supports for the piston. This type of design has been successful but is subject to a range of compression
ratios and discharge temperatures. Additionally, there is also more inherent piston ring leakage than in
lubricated rings. Some applications have resorted to using a mixture of non-lubricated design together
with some oil lubrication in the cylinder. This is called “mini-lube” and is what I furnished in the
natural gas compressors mention above.

One elastomer material that has been introduced into reciprocating compressor design of late is the
material Polyetheretherketone (“Peek”), also referred to as a polyketone. This material is now being
used in compressor valves and has given excellent service in giving long working life and no need for
lubrication at that point. PEEK is very expensive.
Piston speed in a reciprocating compressor is a subject that evokes a lot of emotion and opinions. This
is so because manufacturers are always trying to lower the selling price of their compressors in order to
remain competitive in the marketplace. To do this, they can increase the speed of a machine and obtain
more capacity for less space occupied. However, the wear and tear of a reciprocating machine (in my
opinion and also that of other experienced engineers) is directly related to the speed of the pistons.
Therefore, an experienced engineer uses historical, empirical results to select an appropriate piston
speed for a given application. In the industry, reciprocating compressors are rated as “slow speed”,
“medium speed”, and “fast speed” machines. The actual, real values (in ft/min or meter/min) given to
each of these classifications is rather clouded and not defined by anyone. It is a matter of subjective
opinion on anyone’s part. I personally have always tried to limit my reciprocating compressor speeds to
a maximum of 350 – 700 ft/min. I have found the machines that I have operated work better, longer,
and more economically at that speed than others. That is my experience. Even the API 618 Standard,
“Reciprocating Compressors for Petroleum, Chemical, and Gas Industry Services” fails to pin point a
definite piston speed range limit for dependable service and wear. They tacitly admit to there being a
practical limit to the maximum allowable piston speed – but don’t specifically identify it. By the fact
that all in the industry know that API 11P standard is for “fast” (and quickly consumable) machines
and API 618 is for more durable and reliable machines, we know that there is a difference in price and
in service performance between the fast and the slower machines. Therefore, you must allow for this
effect as an engineer.

Lubrication or non-lubrication depends on a variety of things:

 Lubricated cylinders tend to run cooler in the actual compression chambers. The oil not only
“lubricates”, it also seals and it provides cooling for the affected surfaces.
 Sometimes you simply can’t inject oil into a cylinder. The case of Oxygen compression is an
obvious example of this. Here, you are forced to use something like a soap solution instead.
Breathing air is another example where you should use a non-lube cylinder – as well as
instrument air.
 Non-lubricated cylinders represent less compressor effective life. They require more
maintenance and periodic total replacement of rings, rider bands and sometimes rectification of
pistons. They also are less efficient – capacity and power-wise.
 Compresion ratios are much more critical than the absolute discharge pressure as to whether
you can use non-lubricated cylinders or not. The discharge temperature must be held to a
relative lower value in order to protect the elastomer materials inside the cylinder. PEEK is an
exception here, since it is more temperature resistant than Teflon. However, PEEK costs a
relative fortune as compared to Teflon.
 Control of discharge temperatures is easily done by multiple staging on a machine. However,
this means more capital costs – which is to be expected with non-lubricated features.

For a single-acting cylinder compressing at the outer end of the cylinder,

Pd = (St) (N) ( Π D2/4)....................3.1

Pd = (St) (N) [ Π (D2 –d2)/4)............................3.2

Where,

Pd = Piston displacement, cubic inches

St = Piston stroke, inches
N = Compressor rotative speed, rpm
D = Cylinder internal diameter, inches
d = Piston rod diameter, inches

Piston Speed

Another value to be determined is piston speed, PS. The average piston speed may be calculated by:

PS = 2 x St x N .....................................3.8

To complicate the subject of piston speed, look at Equations 3.1 and 3.8. Note the term St (stroke). The
piston speed can be controlled by a shorter stroke, but because of loss of displacement, the diameter
and/or the speed must be increased, If only speed is increased, the whole exercise is academic as the
piston speed will be back up to the original value.

If, however, diameter alone or both diameter and speed are increased, the net result can be a lower
piston speed. This is another factor that comes to bear at this point because valve life decreases with the
increase in the number of strokes and this can negate the apparent gain in maintenance cost by
shortening the stroke (but increasing the RPM). It would appear that the engineer trying to evaluate a
compressor bid just can't win. The various points are not tendered just to frustrate the user but rather
are given to help show that this is another area that must have a complete evaluation. All facets of a
problem must be considered before an intelligent evaluation can be made.

After all the previous statements, it would seem that it is very difficult to select piston speed. For
someone without direct experience, the following guidelines can be used as a starting point. Actual gas
compressing experience should be solicited when a new compressor for the same gas is being
considered. These values will apply to the industrial process type of compressor with a double-acting
cylinder construction. For horizontal compressors with lubricated cylinders, use 3.5 m/s and for non-
lubricated cylinders use 3 m/s. For vertical compressors with lubricated cylinders, use 4 m/s and for
non-lubricated cylinders use 3.5 m/s. Another factor to consider is the compressor rotative speed
relative to valve wear. The lower the speed, the fewer the valve cycles, which contribute to longer
valve life. A desirable speed range is 300 to 600 rpm.