Dry Gas seal Buffer Gas low dew point 2

[Midan]

Hi there,
Anyone faced a problem in over dryness of Nitrogen as a buffer gas for Dry gas seal?
I have an application of Centrifugal compressors fitted with tandem dry gas seal with intermediate labyrinth, where the buffer gas used is Nitrogen with 95% purity.
While the Nitrogen supplier warned me of that the Nitrogen might be over dry as its dew point reaches -50 deg. C.(sub zero)
The compressor supplier did not stop for long on this warning while I feel that the Seal supplier may have an issue about it.

Your contribution is appreciated.

 

[nickelkid]

The following comment is from a Flowserve Gaspac Tandem Dry Gas Seal that was provided for a recent project:

"Additional gas conditioning may be required to maintain seal supply gas at a minimum of 20° C (36° F) above dew point at the seal faces."

Our source of N2 is cryogenic and we are not intentionally increasing the dew point. For other reasons moisture is not desirable and seal life is not as critical.

 

[TPL]

Quote "to maintain seal supply gas at a minimum of 20° C (36° F) above dew point at the seal faces"

This requirement is prevent liquid drop-out.

Rule # 1 for Dry Gas Seals: They need CLEAN DRY GAS -this applies to the sealing gas and any buffer gas

I fail to see how providing 'over dry' gas can be a bad thing

 

[nickelkid]

TPL, you are correct. I have yet to see moisture added or maintained...

 

[nfinit]

some manufacturers put a requirement in for a specific dew point in the buffer gas supply. The moisture in the buffer gas is intended to lubricate the seal faces. Great in theory, I have never seen it applied and would not recommend that it would be applied.

 

[Compositepro]

I've never heard of this particular problem, but perhaps it is related to why graphite is not used as a lubricant in space applications. The adsorbed gases on graphite are an important part of it lubricant effect.

 

[TPL]

"The moisture in the buffer gas is intended to lubricate the seal"

"The adsorbed gases on graphite are an important part of it lubricant effect."

Sorry guys - the working gap in a typical dry gas seal is 3 to 5 microns and you want absolutely nothing in that gap except CLEAN, DRY gas - no lubricants, no liquids, no particles, no graphite: NOTHING but CLEAN, DRY gas - that's the key to reliable operation of DRY GAS seals

 

[DLite30]

^True.

 

[nfinit]

TPL - it's simply a statement from one of the major compressor OEMs. The actual design doesn't follow the recommendation to maintain a certain dew point.

 

[TPL]

Somewhere in the O+M literature there will be a statement (most likely from the seal vendor but endorsed by the package supplier) along the lines of 'supply gas temperature must be at least 20 deg C above its dew point': this will ensure that there is no liquid dropout within the seal, to meet the essential requirement that the seal only ever sees CLEAN DRY GAS

 

[Midan]

Thank you Guys for all of your replies.
I can conclude that no one had ever seen a problem of over dry buffer gas.
I assume it was only a concern of one of Nitrogen package supplier's clients, but it might never happened in real life.
The most convincing experience is using cryogenic nitrogen as in RobinHandy case.
Anyway, thank you all for all your valuable advises.

 

[GreatHarry]

Hi Midan,

Problems with over dry gas in a DGS? No, never heard of this being a problem. My experiance of failure of the DGS is when the seal gas is supplied with a dewpoint > -55'C. We need N2 supplied at -55'C, to ensure no vapour/ice crystalisation. The application is on a turbo expander and because of the very cold temperatures (-170'C) we experiance ice crystals forming in the DGS and consequently failure. Conversely, in turbo compressors, this is not a problem, since the temperatures of the compressed gas, in this application N2, is from 36'C to 110'C. Not sure why the nitrogen supplier would suggest this problem.
Trust this is of some use.

BRG Great Harry

 

[JamesDUK]

Hi,
Just to say I have heard of this problem, I think it was to do with lifting carbon seal rings on a specific John crane design, since rectified. If you check out John crane literature on the type 83 for example it specifically states it's good for cryogenically dry N2.
Cheers
JD

 

[CaracasEC]

For a compressor driven by a steam turbine with a turning gear motor, this require slow roll start-up and during shutdown to prevent rotor bowing below the DGS lift-off speed. In the absence of water vapor, carbon has a poor lubricating properties and can wear rapidly.

 

[dmertz]

I did some follow up after reading this post. I think there is some confusion on the terminology used. I will attempt to clarify below.

1. Buffer gas or separation gas is for the barrier seal portion of the DGS. This separates the lube oil from the secondary seal chamber on a tandem labyrinth dry gas seal.

2. Primary seal gas is used for the first or primary dry gas seal.

3. Secondary seal gas (typically nitrogen) supplies the secondary seal portion of the dry gas seal.

I spoke with Burgmann and there is no issue with low dewpoint nitrogen to the dry gas seals. However, there IS an issue on certain types of barrier seal designs. The barrier seal is part of the dry gas seal pack in the compressor so that is probably why there is some confusion.

The barrier seal comes in various styles:

A) Labyrinth
B) Carbon ring (static)
C) Carbon ring (dynamic)

The difference between static and dynamic carbon rings are the static have a fixed gap and the dynamic actually lift off after startup. During standstill, the dynamic seals close the clearance and thus there is less consumption.

The issue with the dynamic type of carbon ring is that it can be damaged by excessive dryness. So if nitrogen is used as buffer/separation gas and the dewpoint is very low, then there is a potential to damage the seal.