|
Pressure Packing leakage:
Pressure leaks from the cylinder end, with the packing, to the atmosphere or collection system. It is like a slight suction valve leak with the following differences. The leakage path can be between the packing rings and compressor cylinder rod or around the packing gland sealing surfaces. If the cylinder internal pressure does not drop below the packing external pressure, it leaks all the time. Even while the suction valve is open. This is also true for a cylinder leaking from the cylinder bore to atmosphere via leaking gasket from the head or crank-ends. It is not a recycle leakage so there is no suction or discharge temperature increase and therefore no increase in the discharge delta temperature. Packing vent temperature measurement or the installation of a packing vent flow measurement device would help to verify packing leakage. When packing leakage is combined with other leakages in the cylinder end, it is nearly impossible to identify the leakage with PV card analysis only.
Rings leaking in one direction in a double acting cylinder:
It is true that piston rings can leak in one direction and not in the other. This occurs when one of the ring side surfaces or one side of the ring land is damaged. The leakage may only be in one direction; however, it affects both ends of the cylinder, but in different ways. If the leakage path is from the head-end to the crank-end, the leakage will only occur while the head-end pressure is greater than the crank-end pressure. For the rest of the time, there is no significant leakage, and the head-end PV card will look normal. While the head-end pressure is higher than the crank-end, the crank-end pressure trace will have the characteristic of a discharge valve leak and there may be extra temperature rise measured on the crank-end discharge valves. I will let you figure that out.
Rings leaking in a double acting cylinder with one end unloaded:
This can take several forms depending on the conditions in the unloaded cylinder end. For our discussion, let’s assume the crank-end is loaded and the head-end is unloaded at suction pressure and the rings are leaking from the crank-end to the head-end. This condition has all the characteristics of a suction valve leak in the crank-end (higher suction VE, lower discharge VE, elevated suction temperature, elevated discharge delta temperature). The elevated suction temperature comes from the higher temperature of the high-pressure gas leaking into the head-end, then being pulled back into the crank end through the suction cavity.
Ring leakage with suction valve leakage:
With what you have learned from previous articles, I bet you can figure this one out on your own.
Ring leakage with discharge valve leakage:
When ring leakage is suspected due to the “fat” suction and discharge toes and the presence of the “hump” on the compression line, but you have a flow balance less than one, suspect that there is discharge valve leakage with the ring leakage. In my experience, pure ring leakage in a double acting cylinder will always result in a flow balance significantly greater than 1.0 in both ends of the cylinder.
Suction and discharge valves leaking in the same end:
Again, several conditions can exist: suction and discharge valves leaking the same, suction leaking more than discharge or discharge leaking more than suction. Let’s assume both are leaking the same. Some say that this is like ring leakage, but remember, to get the traditional ring leakage PV pattern, you need to have rings leaking in both ends of the cylinder. If you have the “fat” toes and distorted compression and expansion events in one end and the other cylinder end looks normal, suspect both suction and discharge leakage in the end with the distorted pattern. The valve with the greatest pressure differential across it will leak the most. This condition will alternate between the suction and discharge valves during the compression and expansion events. The first part of the expansion line looks like suction leakage, and the second half looks like discharge valve leakage. On the compression line, the first half looks like discharge valve leakage, and the second half looks like suction valve leakage. These lines may show curvature, but not the traditional “hump” seen with ring leakage. You will also get a hot suction valve (not a ring leak characteristic).
Conclusion:
Now that we see that there are many different leakage combinations it is easier to understand how the novice compressor analyst may get confused while trying to make a definitive call, and why training and experience are so necessary for effective on-condition and preventative maintenance programs.
Happy monitoring.
Regards,
Warren
| 
