leak checking

[captainleaky]

Our company leak checks our refrigeration packages by using a combination of a digital pressure gauge and digital thermometer. our procedure goes as follows...
1. fill package with 300 psi nitrogen
2. let sit for two hours to equalize.
3. take pressure and temperature reading every hour for two hours.
4. enter the first and last readings of pressure and temp into a program that was developed by our refrigeration engineering dept. This program will tell you if the pressure drops more than the temperature which indicates a leak...

the problem is that during certain times the test quite often shows a leak when there is not one.

my feeling is that it is caused by atmospheric pressure such as a barometeric drop.

does anyone out there have any thoughts on this??

[lowcool]

yep it can drive you crazy even pulling full vac on a system then hitting it with nitro,keeping all non condensables out of system it still varies,its a bugger if you have to go through certified pressure tests,pays to keep a screwdriver handy for a quick clock adjustment.thing that gets me i cannot remember this happening 15 years ago,pressure variances are dependent on temperature ive found,not bad for an inert gas.

[gwynn jones]

The firm I work for are big into pressure testing.
On a new system, or one with new components, firstly has a strength test of 1.1 x maximum allowable pressure [PRVs have to come out for this] just to make sure nothing is going to explode [I've pressured things over 38 Bar, buttock clenching time...] Then that is dropped to the pressure/leak test, which is 0.95 x maximum allowable pressure, which it must hold.
We work on a pressure rise/fall of approx 0.34% per every 10K rise/fall of temp.
Gas used is OFN, hoses steel braided or steel tubing and we use a special charging manifold with an adjustable pressure relief valve built into it, just in case someone accidentally over pressures the system. No manifold gauges!! Especially those with a sightglass built in.
Area has to be roped off and only trained people allowed that side.
Some guy died here in the UK a few years ago, pressure testing. Got his head too close to the compressor body just as his mate over opened the OFN bottle - can't happen with our pressure rig!
It's a real mission pressure testing for our lot but it works.