03.03.2013 Views

2000 Hook-up Book - Spirax Sarco

2000 Hook-up Book - Spirax Sarco

2000 Hook-up Book - Spirax Sarco

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

The importance of effective condensate<br />

removal from steam<br />

spaces has been stressed<br />

throughout this course. If maximum<br />

steam system efficiency is<br />

to be achieved, the best type of<br />

steam trap must be fitted in the<br />

most suitable position for the<br />

application in question, the flash<br />

steam should be utilized, and the<br />

maximum amount of condensate<br />

should be recovered.<br />

There are a number of reasons<br />

why condensate should not<br />

be allowed to discharge to drain.<br />

The most important consideration<br />

is the valuable heat which it contains<br />

even after flash steam has<br />

been recovered. It is possible to<br />

use condensate as hot process<br />

water but the best arrangement is<br />

to return it to the boiler house,<br />

where it can be re-used as boiler<br />

feed water without further treatment,<br />

saving preheating fuel, raw<br />

water and the chemicals needed<br />

for boiler feed treatment. These<br />

savings will be even greater in<br />

cases where effluent charges<br />

have to be paid for the discharge<br />

of valuable hot condensate down<br />

the drain.<br />

Condensate recovery savings<br />

can add <strong>up</strong> to 20 to 25% of<br />

the plant’s steam generating<br />

costs. One justifiable reason for<br />

not returning condensate is the<br />

risk of contamination. Perforated<br />

coils in process vessels and heat<br />

exchangers do exist and the<br />

cross contamination of condensate<br />

and process fluids is always<br />

a danger. If there is any possibility<br />

that the condensate is<br />

contaminated, it must not be<br />

returned to the boiler. These<br />

problems have been lessened by<br />

the application of sensing systems<br />

monitoring the quality of<br />

condensate in different holding<br />

areas of a plant to determine condensate<br />

quality and providing a<br />

means to re-route the condensate<br />

if contaminated.<br />

Condensate Line Sizing<br />

Condensate recovery systems<br />

divide naturally into three sections,<br />

each section requiring<br />

different design considerations.<br />

a. Drain Lines to the traps carry<br />

pressurized high temperature<br />

hot water that moves by gravity.<br />

b. Trap discharge lines that<br />

carry a two-phase mixture of<br />

flash steam and condensate.<br />

c. Pumped return systems utilizing<br />

electric or non-electric<br />

pumps.<br />

Drain Lines To Traps<br />

In the first section, the condensate<br />

has to flow from the<br />

condensing surface to the steam<br />

trap. In most cases this means<br />

that gravity is relied on to induce<br />

flow, since the heat exchanger<br />

steam space and the traps are at<br />

the same pressure. The lines<br />

between the drainage points and<br />

the traps can be laid with a slight<br />

fall, say 1” in 10 feet, and Table 13<br />

shows the water carrying capacities<br />

of the pipes with such a<br />

gradient. It is important to allow<br />

for the passage of incondensibles<br />

to the trap, and for the extra water<br />

to be carried at cold starts. In<br />

most cases, it is sufficient to size<br />

these pipes on twice the full running<br />

load.<br />

Trap Discharge Lines<br />

At the outlet of steam traps, the<br />

condensate return lines must<br />

carry condensate, non-condensible<br />

gases and flash steam<br />

released from the condensate.<br />

Where possible, these lines<br />

should drain by gravity to the condensate<br />

receiver, whether this be<br />

a flash recovery vessel or the<br />

vented receiver of a pump. When<br />

sizing return lines, two important<br />

practical points must be considered.<br />

Condensate Recovery Systems<br />

Table 13: Condensate, lb/h<br />

Steel Approximate Frictional Resistance<br />

Pipe in inches Wg per 100 ft of Travel<br />

Size 1 5 7 10<br />

1/2" 100 240 290 350<br />

3/4" 230 560 680 820<br />

1" 440 1070 1200 1550<br />

1 1 /4" 950 2300 2700 3300<br />

1 1 /2" 1400 3500 4200 5000<br />

2" 2800 6800 8100 9900<br />

2 1 /2" 5700 13800 16500 <strong>2000</strong>0<br />

3" 9000 21500 25800 31000<br />

4" 18600 44000 5<strong>2000</strong> 63400<br />

First, one pound of steam<br />

has a specific volume of 26.8<br />

cubic feet at atmospheric pressure.<br />

It also contains 970 BTU’s<br />

of latent heat energy. This means<br />

that if a trap discharges 100<br />

pounds per hour of condensate<br />

from 100 psig to atmosphere, the<br />

weight of flash steam released<br />

will be 13.3 pounds per hour, having<br />

a total volume of 356.4 cubic<br />

feet. It will also have 12,901<br />

BTU’s of latent heat energy. This<br />

will appear to be a very large<br />

quantity of steam and may well<br />

lead to the erroneous conclusion<br />

that the trap is passing live steam<br />

(failed open).<br />

Another factor to be considered<br />

is that we have just released<br />

13.3 pounds of water to the<br />

atmosphere that should have<br />

gone back to the boiler house for<br />

recycling as boiler feed water.<br />

Since we just wasted it, we now<br />

have to s<strong>up</strong>ply 13.3 pounds of<br />

fresh city water that has been<br />

softened, chemically treated and<br />

preheated to the feedwater system’s<br />

temperature before putting<br />

this new water back into the boiler.<br />

Secondly, the actual formation<br />

of flash steam takes place<br />

within and downstream of the<br />

steam trap orifice where pressure<br />

drop occurs. From this point<br />

onward, the condensate return<br />

system must be capable of carrying<br />

this flash steam, as well as<br />

condensate. Unfortunately, in the<br />

past, condensate return lines<br />

45<br />

SYSTEM DESIGN

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!