
YORK INTERNATIONAL
FORM 160.73-O2 (605)
26
VACUUM TESTING
After the pressure test has been completed, the vac u um
test should be conducted as follows:
1. Connect a high capacity vacuum pump, with in di -
ca tor, to the system charging valve as shown in
Fig. 11 and start the pump. (See “Vacuum De hy -
dra tion”.)
2. Open wide all system valves. Be sure all valves to
the at mo sphere are closed.
3. Operate the vacuum pump in accordance with
VAC U UM DEHYDRATION
until a wet bulb
tem per a ture of +32°F or a pressure of 5 mm Hg is
reached. See Table 2 for corresponding values of
pres sure.
4. To improve evacuation circulate hot water (not to
exceed 125°F, 51.7ºC) through the evaporator and
condenser tubes to thoroughly dehydrate the shells.
If a source of hot water is not readily available, a
portable wa ter heater should be employed.
DO NOT
USE STEAM
. A suggested method is to connect a
hose between the source of hot water under pres-
sure and the evaporator head drain connection, out
the evaporator vent connection, into the condenser
head drain and out the condenser vent. To avoid the
possibility of caus ing leaks, the temperature should
be brought up slow ly so that the tubes and shell are
heated evenly.
5. Close the system charging valve and the stop valve
between the vacuum indicator and the vacuum pump.
Then disconnect the vacuum pump leaving the vac-
u um indicator in place.
6. Hold the vacuum obtained in Step 3 in the system for
8 hours; the slightest rise in pressure indicates a leak
or the presence of moisture, or both. If, after 24 hours
the wet bulb temperature in the vacuum in di ca tor has
not risen above 40°F (4.4°C) or a pres sure of 6.3 mm
Hg, the system may be considered tight.
Be sure the vacuum indicator is valved
off while holding the system vacuum
and be sure to open the valve between
the vac u um indicator and the system
when checking the vacuum after the
8 hour period.
7. If the vacuum does not hold for 8 hours within the
limits specifi ed in Step 6 above, the leak must be
found and repaired.
TABLE 2 –
SYSTEM PRESSURES
*GAUGE
ABSOLUTE
BOILING
INCHES
OF
TEMPERATURES
MERCURY
(HG)
MILLIMETERS
OF
BELOW ONE
PSIA
OF MERCURY
MICRONS
WATER
STANDARD
(HG)
°F
ATMOSPHERE
0
14.696
760.
760,000
212
10.24"
9.629
500.
500,000
192
22.05"
3.865
200.
200,000
151
25.98"
1.935
100.
100,000
124
27.95"
.968
50.
50,000
101
28.94"
.481
25.
25,000
78
29.53"
.192
10.
10,000
52
29.67"
.122
6.3
6,300
40
29.72"
.099
5.
5,000
35
29.842"
.039
2.
2,000
15
29.882"
.019
1.0
1,000
+1
29.901"
.010
.5
500
–11
29.917"
.002
.1
100
–38
29.919"
.001
.05
50
–50
29.9206"
.0002
.01
10
–70
29.921"
0
0
0
*One standard atmosphere
= 14.696 PSIA
= 760 mm Hg. absolute pressure at 32°F
= 29.921 inches Hg. absolute at 32°F
NOTES: PSIA = Lbs. per sq. in. gauge pressure
= Pressure above atmosphere
PSIA = Lbs. per sq. in. absolute pressure
= Sum of gauge plus atmospheric pressure
Maintenance