# Hydrocarbon Compression

## Dehydration Capacity VS Temperature

Three process requirements must be met for gas to be dried in a standard glycol dehydration unit:
1. The gas velocity through the contactor tower must not be great enough to entrain glycol into the dried gas. Theoretically, the entrainment of glycol does not interfere with drying. In practice, the continuous loss of glycol will knock a drying plant off-line as the unit’s inventory of glycol disappears. Incidentally, it is not possible to measure the water content of gas containing a glycol mist.
2. The glycol pump must have the capacity to circulate enough glycol to absorb the water vapor contained in the natural gas. Of course, hotter gas can contain more water vapor. Increasing the gas temperature from 80°F to 100°F may double its water content.
3. The glycol reboiler must have a sufficient heat-duty capacity to regenerate the glycol at a high enough temperature to adequately dehydrate the gas.

As the temperature of the gas flowing through a dehydration contactor tower rises, its capacity will decrease as follows:

If a tower temperature increases from 80°F to 120CF, a tower’s capacity will decrease by barely 3V2%.
On the other hand, the amount of glycol circulation may or may not greatly increase as the gas inlet temperature rises. Figure 6-3 clarifies this point. A large booster compressor is serving a concentrated gas field. The gas produced from the wells enters the compressor’s suction scrubber at a temperature independent of seasonal fluctuations. However, the aerial cooler on the compressor’s discharge cools the gas to 80° in the winter versus 120°F in the summer. Question: How much more glycol circulation is required to dry the gas? The requisite data to perform the calculation are given in Figure 6-4.

At first glance, it would appear that three or four times as much glycol circulation is required. But remember that the 120°F compressed gas is not saturated with water vapor; it is really superheated. The compressed gas will have the same water content until it is cooled by the aerial cooler to below its dewpoint, in this case 79°F. If a contactor tower with 10—15 trays were employed, there would likely be no effect at all on glycol circulation requirement. For the typical 6-tray contactor, industry correlations indicate that an additional 10-30% of glycol circulation is needed; that is, far less that the 300—400% required if the gas were saturated with water at the compressor discharge temperature.

Suppose, however, that the gas coming out of the ground is hot, perhaps 110°F. This gas, after compression and cooling to 1,000 psig and 120°F, would be saturated with moisture. Then, during winter operation, when the gas is cooled to 80°F, only one-third as much glycol circulation would be required as in the summertime. The condensed water corresponding to the difference in water content of 110°F, 700 psig gas vs 80°F, 1,000 psig gas would drop out in the bottom section of the contactor tower.

Written by Jack

November 7th, 2009 at 3:36 pm

Posted in Glycol Dehydration

Tagged with

## Flooding Dehydrator Tower Plugged Tray

Drying towers in natural gas service can become rapidly fouled with drilling mud or formation and frac sand. The sand appears in the wellhead gas when the rate of gas production becomes excessive, and the sand is thus sucked out of the formation and into the well’s tubing. Drilling mud is found in natural gas for two reasons:
1. A new well is not properly circulated and flowed-back to clear the drilling mud out of the production tubing prior to commissioning.
2. During the drilling operation, excessive mud pressures are accidentally applied to the well, and the drilling mud is thus inadvertently forced into the producing formation. Some of this mud must eventually reappear in the downhole production tubing.

Not infrequently, a dehydrator loses its ability to dry gas from a field in which a new well has been put on-line. When this occurs, the culprit is invariably drilling mud plugging the contactor internals. For remote locations, one procedure that has proved to work is as follows:
1. A large water truck equipped with a pump to deliver about 50 psig, is sent to the site.
2. The dehydrator tower is blocked in and depressured. Both the tower inlet and outlet are disconnected from the gas piping. A special flange attachment, designed to mate up with a hose connection, is installed on the gas outlet line.
3. A two-inch hose from the discharge of the truck’s pump is connected to the dehydrator tower gas outlet line.
4. The pump is started and adjusted so that the pressure at the top of the tower—i.e., the water inlet—is about 5 psig. It is important not to apply too great a pressure because the trays could collapse.
5. Once the water draining from the bottom of the contactor tower appears clear, switch the water inlet to the bottom gas inlet. Over-flow the tower until the water is again clear. The water overflow rate must be substantially higher than the normal glycol circulation rate to obtain enough liquid traffic to effectively wash the trays.

Why, you might ask, it is necessary to initially wash a badly fouled tower from the top, down? A tray plugged with mud will severely restrict the flow of water. The resulting pressure drop may be sustained by the tray when it is pressed down onto the tray support ring when applied from the bottom of the tray.

In more accessible locations, it is a good practice to acidize a contactor tower after water washing. Acidizing consists of circulating an inhibited hydrochloric acid solution (typically 5% HCI) to the bottom of the tower with an acid truck. This is an effective method to clean contactors without promoting channeling of the gas flow through the trays. Acidizing is especially effective when iron scale deposits make up a portion of the fouling deposits. Including the acid disposal expense, acidizing a drying tower can cost between \$20,000-50,000. When hydrocarbon deposits consisting largely of polymers formed in the glycol reboiler are the major fouling component, a caustic wash, as opposed to acidizing, is in order. In the caustic washing procedure, a degreaser is also employed. A more elaborate, but thorough, procedure is summarized in Table 6-1.

Written by Jack

November 1st, 2009 at 3:21 pm

## Flooding Dehydrator Tower Fouling vs Flooding

A distillation column can flood due to dry damage, undersized liquid downcomers, high liquid level in the bottom of the tower, foulcommonly encountered in natural gas conditioning. The troubleshooter should first check for flooding due to excessive vapor velocities. The following correlation may be used for trayed columns 2 feet or more in diameter with a standard 2—foot tray spacing:

This equation is not intended for design purposes; rather it is based on field observations for towers exhibiting noticeable but tollerable glycol looses. These towers had been in service for some time and had been exposed to a moderate amount of fouling. If the actual volume of gas exceeds the allowable volume as calculated above, you may be confident that an intolerable glycol loss is due to an excessive  vapor velocity. Note that for sizing a new contactor tower, a coefficient of 2.0 in the above equation would be suitable.

Written by Jack

November 1st, 2009 at 3:16 pm

## Flooding Dehydrator Tower

The field supervisor’s first indication of a flooded contactor tower is usually a report of excessive glycol loss. A check of a lowpoint bleeder on the gas pipeline downstream of the tower will show glycol. After refilling the glycol reboiler, the level in the reboiler gauge glass noticeably decreases after a few hours. This is a further indication of flooding. Of course, a dehydration system loosing glycol this fast cannot dry natural gas on a continuous basis.

One simple explanation of such glycol losses is a leaking dry gas to dry glycol heat exchanger (Figure 6-2). Note that the glycol pressure in this heat exchanger will be slightly higher than the gas pressure. To check for leakage, shut off and block in the glycol pump, block in the dry glycol at the contactor tower, and open an intervening bleeder between the pump and the tower. If gas does not blow out of the bleeder, the exchanger is not leaking.

Written by Jack

November 1st, 2009 at 2:34 pm

Posted in Glycol Dehydration

Tagged with

## Leaking Feed-Effluent Exchanger

The hot glycol from the reboiler is cooled by heat exchange with the wet glycol from the contactor. This heat transfer typically takes place in a double-pipe or plate-type exchanger. On one of the double-pipe heat exchangers, I noticed that the reboiled glycol was being cooled to a rather low temperature. I suspected that this could be an indication of a leaking feed-effluent exchanger. That is, cooler (120°F) wet glycol might be leaking into warmer (165°F) dry glycol. To verify my suspicions, I blocked in the dry glycol at the reboiler and at the suction to the pump. The appearance of a steady stream of liquid at an intervening bleeder confirmed that the feed-effluent exchanger was leaking, hi effect, wet glycol was bypassing the reboiler and flowing straight back to the contactor tower.

After fixing the leak, this reboiler and the units that had suffered from an inefficient pump and a faulty temperature controller were put back on-line. The treated natural gas was checked and found to meet pipeline moisture specifications.

Written by Jack

November 1st, 2009 at 2:31 pm

Posted in Glycol Dehydration

Tagged with

## Glycol Regeration Temperature

The gas exiting the top of the contactor in Figure 6-1 can be assumed to be in equilibrium with the reboiled—i.e., dry—glycol. The higher the glycol reboiler temperature, the dryer the glycol. The dryer the glycol, the dryer the treated natural gas. For most of the year in El Gringo, critical control of the glycol reboiler temperature gas was not vital. Relatively cool ambient temperatures maintained the top temperature of the contactor towers below 110°F. But now, in mid-July, this temperature was peaking at 122°F every afternoon. I checked my gas purification data book1 and calculated that, for the 1,020 psig operating perssure of the contactors, it should be possible to meet the required moisture specification. My calculations were based on a reboiler temperature at 375°F. For triethylene glycol, which is the work horse of the gas drying industry, the maximum recommended reboiler temperature to prevent thermal degradation of the glycol is 400°F. The six El Graingo dehydrator reboilers were all set to hold 375°F. But by checking the actual reboiler temperatures with a calibrated thermometer, I determined that one of the reboilers was actually operating at 350°F as opposed to 375°F. This reduced temperature was sufficient to greatly increase the water concentration of the “dry” glycol, so that the moisture content of gas treated with this glycol stream was doubled.

A simple recalibration of the reboiler temperature controller rectified this problem. Incidentally, operating a triethylene glycol reboiler at 375°F-400°F does not necessarily result in a noticeable increase in glycol degradation. The trick is to keep the glycol filters in good repair. Dirty glycol fouls the reboiler heat-transfer tube. This in turn causes hot spots on the heat-transfer surface, which accelerates thermal decomposition.

Written by Jack

November 1st, 2009 at 2:25 pm

Posted in Glycol Dehydration

Tagged with

## Indication of Reduced Glycol Circulation

The first oddity I noticed was the noise from the vents associated with the individual reboilers. As Figure 6-1 shows, the expanding gas, used to drive the glycol pumps is also used as fuel to reboil the glycol. The excess gas not burned in the reboiler is vented under pressure control to the atmosphere. When the efficiency of the glycol pump is reduced due to mechanical problems, two factors act to increase excess gas venting:

• The reboiler firing rate drops because less glycol must be reheated.
• The amount of gas flowing from the tower to the glycol pump increases because there is less glycol liquid to restrict the flow of gas.

Hence, the net result of a reduction in glycol circulation rate due to reduced pumping efficiency is increased venting of excess natural gas. Of the six vents (one for each reboiler), only one was blowing hard. I also observed that the main burner on this particular reboiler was rarely on. Note: Temperature control on glycol reboilers works hke your home heater—either full on or full off. Lack of firing on a glycol reboiler—that is, low reboiler heat duty—is another indication of a low glycol circulation rate.

The usual cause of glycol pump failure is deterioration of the O ring seals. Next morning, I requested that the suspect pump be overhauled. While this work proceeded, I continued my investigation.

Written by Jack

October 25th, 2009 at 8:38 pm

Posted in Glycol Dehydration

Tagged with

## Dehydration and Compression Station Troubleshooting

Natural gas transported through common carrier pipelines must meet a moisture specification of 7 pounds of water per MMscf. Gas is usually dried to meet this requirement by scrubbing with a concentrated glycol solution. Figure 6-1 shows a standard glycol contactor tower, regenerator, and pump.

Gas flows into the bottom of this tower where entrained water and naphtha drop out and are withdrawn under level control. The upflowing gas is contacted with the circulating glycol and dried. The glycol is pressured from the contractor to the regenerator, where it is heated to its boiling point to drive off water. Typically, 100 pounds of circulating glycol absorbs 3—4 pounds of water. After cooling, the reboiled glycol is pumped back to the contractor tower.

On the surface it would not seem possible that much could go awry with such a simple system. But, of course, the experienced process operator knows that it is only a matter of time for anything that can go wrong to go wrong. As a case in point, consider the operation of the glycol circulating pump.

This ingenious positive displacement pump is driven by expanding gas withdrawn along with the wet glycol, from the contactor tower (see Figure 6-1). The speed of this pump is set by a small valve that controls the amount of expanding gas emitted into the pump. An operator judges the amount of glycol circulation based on the audible strokes made by the pumps internals. The quicker the strokes, the greater the glycol circulation.

But suppose the pump has developed mechanical problems that reduce the volume of glycol normally pumped per stroke? Or perhaps the pump internals have deteriorated to the point that glycol circulation has stopped. Since glycol drying units are not normally equipped with flow meters on the circulating glycol, how can the process operator of the troubleshooting engineer recognize the problem.

Written by Jack

October 25th, 2009 at 8:28 pm

Posted in Glycol Dehydration