How to detect leaks on Steam Pressure Safety Valves 3

[Jimmyvanskip]

I am currently conducting a survey of all the Steam PSVs in our plant to determine which ones are leaking.
I have tried using a infra-red thermometer and/or thermocouple to measure the pipe temperature under the PSV while also noting the PSV vent pipe temperature. I then use steam tables to convert to gauge pressure and compare this to the pressure set point of the PSV and that of the upstream Pressure Reducing Valve.

My theory is that if the line pressure under the PSV is less than the upstream PRV setpoint then the PSV is passing. If the line pressure is the same as the PSV, then the upstream PRV is passing and the PSV is relieving.

Problem is: in steam lines which have existing pressure gauges the pipe temperature reading is significantly less than the corresponding pressure from the steam tables.

My questions are:
1. Is measuring the pipe temperature an accurate way to determine steam pressure in the pipe or is the pipe temperature always less due to losses, specific heat capacity of pipe etc...

2. What is a typical pipe vent temperature for a PSV which is not passing? Is the conductivity of the pipe such that the temperature of the vent pipe will always be high?

3. Any other methodology/experience regarding determining if PSVs are passing would be appreciated.

 

[TD2K]

I would have thought that if I had a steam PSV leaking very much steam at all I should have a plume of steam coming off the tailpipe or at the very least, condensate dripping out of the low point weephole and it would be pretty obvious I had a passing PSV.

Where I've used infrared guns, I seem to recall a reasonable temperature offset over what the process fluid should be, 50F or more? It's been a while since I did this. Have you tried taking a portable thermocouple out to the piping and then put insulation over the thermouple end so the pipe temperature should be closer to steam temperature?

 

[Duwe6]

You look for a different - higher - temperature on the suspect PSV's discharge pipe, as far away from the PSV as possibly.

I like the Three-Mile Island method of having a thermocouple on each discharge pipe, well away from its PSV. If three read 240° and the fourth reads 310° you have found the steam 'whisper'. If the TMI idiot operator had looked at his thermocouples, he would've known that he had a leaking PSV.

Temperature/Pressure table cannot help - the pressure in the tailpipe is zero; open to atmosphere. You have to look for differences in temperature in your PSV tailpipes, reading them well away from their valves, and at close to the same distance downstream from each PSV. You won't get a quantity, but you will get an indication of 'leaker' vs. 'tight'

 

[sheiko]

Jimmy,

In the steam lines, there is a single component (H2O), a single-phase (vapor), so the degree of freedom would be equal to 1-1+2 = 2. That is the temperature and the pressure in a line is independant.

Maybe a noninvasive external flow measurement device strapped onto a PSV outlet line could help.

"We don't believe things because they are true, things are true because we believe them."

 

[rustbuster]

I would use an ultrasonic instrument. Go to

http://www.uesystems.com

for details on the equipment.

 

[Jimmyvanskip]

Thanks Rustbuster,
I actually have a UE Ultraprobe, would you listen to the pipe and check for a "whistle" similar to doing steam trap testing?
Jimmyvanskip

 

[rustbuster]

Jimmy,

If a valve is paaing you should be able to hear the vibration with the contact probe.