Dry face seal

The second option, dry face or gas seals, was developed in the early 1970s. In dry face seals, the hydrodynamic pressure that develops between the rotating and stationary seal faces produces positive sealing with minimal seal gas leakage. 

Dry Face Seal Application, Atmospheric Bearing Housing (Onshore Application)... 

Dry face seal leakage test under static and dynamic conditions... 

Dry Face Seal Leakage Test Under Static and Dynamic Conditions... 

To this day, balanced mechanical seals using o-rings are the standard in industry. Meehanical seals eontinue to evolve in sealing face technology, computer design, finite element analysis, cartridge designs, split seals, double or dual seals, and dry gas seals. 

The first documented use of dry gas seals in turboexpanders was in 1989. At that time, an ethylene plant in Scotland worked with a dry gas seal manufacturer to retrofit a turboexpander-integral gear-generator package. The partial success of that project was sufficiently encouraging for both user plant and turboexpander manufacturer to undertake a redesign of all three expander stages. Dry face or gas seals were installed at that time. 

The dry gas seal is a variation of the mechanical contact seal. It differs in that it uses a microscopically thin layer of gas to separate and lubricate the faces. The seal is configured in a tandem or double-opposed seal arrangement. More complete details are covered in Chapter 5 under Dry Gas Seals. 

Dry gas seals are in the positive seal class and have the same basic design features as mechanical face seals with one significant difference. The dry gas seal has shallow grooves cut in the rotating seal face located part way across the face. The grooves may be in a spiral pattern the exact location and pattern vary from one manufacturer to another. Lubrication and separation is effected by a microscopically thin film of gas. This implies some finite amount of leakage, which is quite small but must be accounted for in the design. 

As stated, the dry gas seal does come with its own set of disadvantages. The biggest of these is that the buffer gas must be reliable. Loss oi buffer gas in some cases will reverse the differentia] pressure across (iie seal faces, which will damage the seal in short order. The seals will opei ate at a zero differential pressure level, but when possible, even a small differential in the proper direction is recommended by the manufacturers. Another disadvantage is the requirement for clean and dry gas at the seal... 

When starting equipment with mechanical seals, make sure the seal faces are immersed in liquid from the beginning so they will not be damaged from dry operation. The following recommendations for seal startup apply to most types of seal installations and will improve seal life if followed ... 

Caution the electrician not to run the equipment dry while checking motor rotation. A slight turnover will not hurt the seal, but operating full speed for several minutes under dry conditions will destroy or severely damage the rubbing faces. 

The purified commercial di-n-butyl <2-tartrate, m.p. 22% may be used. It may be prepaid by using the procedure described under iso-propyl lactate (Section 111,102). Place a mixture of 75 g. of d-tartaric acid, 10 g. of Zeo-Karb 225/H, 110 g. (136 ml.) of redistilled n-butyl alcohol and 150 ml. of sodium-dried benzene in a 1-litre three-necked fiask equipped with a mercury-sealed stirrer, a double svu-face condenser and an automatic water separator (see Fig. Ill, 726,1). Refiux the mixture with stirring for 10 hours about 21 ml. of water collect in the water separator. Filter off the ion-exchange resin at the pump and wash it with two 30-40 ml. portions of hot benzene. Wash the combined filtrate and washings with two 75 ml. portions of saturated sodium bicarbonate solution, followed by lOo ml. of water, and dry over anhydrous magnesium sulphate. Remove the benzene by distillation under reduced pressure (water pump) and finally distil the residue. Collect the di-n-butyl d-tartrate at 150°/l-5 mm. The yield is 90 g. 

Fig. 3.4.8 One-dimensional SPI drying profiles drying are indicated by the symbols ( ) and of concrete moist-cured for 28 days and of a (O), respectively. The measurement para-0.6 water-cement ratio [9]. The specimen was meters were field of view (FOV) 150 mm, sealed except for one face and exposed to a acquisition points 64, tp = (55 - 300 ps, 8 val-drying regime at 38 °C and 40% relative humi- ues), a = 6°, TR = 100 ms, acquisition time dity for 28 days. The spatial moisture content 3.5 min per encoding time, after 28 days of moist curing and 28 days of...

The sealed tube is placed in the tube furnace, and the temperature is raised slowly over a period of about 2 days. The end containing the reactants should be maintained at 1100-1200°C., and the other end at about 700°C. After 2 or 3 days in these conditions, black, shiny crystals up to 1 mm. on an edge form in the tube at a region of about 800°C. The tube may be cooled in the furnace by turning off the power. When the tube has cooled to room temperature, it may be removed and opened. Owing to the possible presence of excess phosphorus, the tube should be opened in a hood and the superfluous phosphorus allowed to bum off. In addition, the precaution of gloves and a face shield should be taken. The crystals are washed in water and acetone and allowed to dry. Anal. Calcd. for SiP2 Si, 68.8 P, 31.2. Found Si, 68.51 P, 31.33. 

A sample with a w/c ratio of 0.4 by weight (which is at the threshold water content required for complete hydration of OPQ was cured for 28 days in a dry desiccator with a relative humidity of <5%. All surfaces of the sample except the top face were sealed with Parafilm, to promote evaporation of water through this surface only. This was done to disrupt the hydration process in this part of the sample and hence introduce some inhomogeneity into the cured sample. Following the cure period, the Parafilm was removed and the sample was dried in an oven for 48 h at 105°C in order to remove all evaporable water. Water penetration studies were then carried out as described below. 

In order to verify the dynamic concept of the behavior of gas throughout sedimentary basins, we first report the results of an empirical study on The Netherlands offshore gas fields (Broad Fourteens Basin, Fig. 1 shadowed area). This basin is part of the European Upper Carboniferous Basin. The gas fields are sourced from Upper Carboniferous coals and shales, reservoired in Upper Permian Zechstein carbonates (Plattendolomit) and sealed by Zechstein rock salt (Na2-4). The problem facing exploration was that the rate of dry holes was very high even if structural closure and favorable reservoir properties were confirmed. In order to find a solution and to offer a tool for the prediction of fill or non-fill of structures before drilling, we examined 29 case histories (fields, finds, shows and dry holes) under the aspect of dynamics of gas flow. 

---