“we have a screw comp. with a dry gas seal (John Crane) Type Tandom , we used N2 as a buffer gas for this seal. suddenly we've found a leakege from the seal discharge side for the female rotor.”
First of all, being a tandem seal, the primary seal is the one that performs the main sealing function – the secondary seal is only there to maintain containment in the event of a primary seal failure, allowing the machine to be safely shutdown (tripped) without loss of process gas to the local environment.
The buffer gas seal complements the dry gas seal but has a completely separate function. It is usually either a labyrinth or carbon ring and its purpose is to 'protect' the dry gas seal by preventing lube oil from reaching the secondary (or outboard) seal. Buffer gas is injected into the seal to enhance its sealing capability. You can replace ‘buffer’ by ‘separation’ to describe the function of the seal in a different way. In this case, the mention of N2 is most likely irrelevant, but it is important to ensure that the buffer gas supply is turned on before the lube oil pumps/machine is started.
Far and away the most common cause of dry gas seal failure/excessive leakage is contamination of the working faces – the working gap is only about 3 to 5 microns (a human hair is about 125 to 150 microns in diameter) so it doesn’t take much in the way of particulant contamination (think talcum powder) to cause a seal failure. You should check your seal filtration system for damaged/blown filters and have any contaminants analysed (likewise later on if you have the seal removed and overhauled – you will need to identify any contaminants that are found and then figure out their likely source).
With such a small working gap, it is possible that the seal face can be ‘cocked’ and fail to seal – a shock (such as a trip) can often remedy this issue.
I have a couple of compressors with dry gas seals that often leak after startup, but the leakage rates gradually drop off to normal between 6 and 12 hours of operation. I believe that this is due to fine particles being drawn into the seal across the seal between the compressor and the dry gas seal (when the seal gas pressures are in a transient mode during startup and shutdown conditions) and gradually being blown out following a period of operation.
Is this a new machine or has it been operating for a while? Do you get many failures of these seals? Do failures occur more at the suction or discharge end? What is the typical MTBF of the seals? What is the current leak rate compared to alarm and trip settings?
Was the excessive leakage noted immediately after a startup or did it develop over a period of time when the machine was operating normally?
Typically, seal faces are tungsten or silicon based and you should check for compatibility with the sealing gas to ensure that there is no corrosion taking place
If there is high leakage at all times, then you have little choice but to replace the seal. Make sure you get a report describing the reason for failure.