1
$\begingroup$

I was visiting an industrial site not too long ago and I noticed an interesting phenomenon involving one of their steam condensate pipes. This (insulated) pipe was suspended from the ceiling. It hung down about 20 feet, supported by a series of supports. Each support comprised of a roller (positioned under the pipe) and a rod, which ran from the roller to the ceiling directly over head.

enter image description here

It ran roughly 100 yards from one end of the facility to the other. This pipe also included a U shaped expansion loop.

enter image description here

As I watched it, I noticed that it was swaying back and forth, with period of about 2 seconds, with an amplitude of a couple of feet. It followed the description of a second harmonic standing wave in a string with fixed ends, where there is a stationary point in the middle that does not move.

enter image description here

I was (and am) still very curious about this. Is this often seen in a steam system? Is this ok? What is this called? What causes it? Should it be dampened? Will it hurt the life of the pipe? It may be ok, but it really reminded me of the tacoma bridge.

I was assuming that the oscillation was caused by water flowing around the expansion loop, imparting momentum to it as it encountered the turns, which then caused the pipe to oscillate at its resonance frequency.

$\endgroup$
  • $\begingroup$ This condensate pipe is probably being driven at its natural frequency, which isn't good. A mechanical engineer or civil engineer should take a look at it to ensure that it is safe to operate. $\endgroup$ – David White Mar 2 at 0:13
1
$\begingroup$

This is common in piping systems carrying steam and connected to big boilers and turbines, for the following reasons.

A boiler generates a strong random "rumble" while operating, which comes from the boiling process inside it. When connected to piping systems which possess compliance and inertia, those pipe runs are driven with that random spectrum and resonate at their natural frequencies. The amplitudes that result are sufficient to abrade the pipe joint seals to the point where they develop significant leaks.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ I've never seen this kind of swing in a pipe before. Does it have a name? Where can I read more about it? $\endgroup$ – Hoytman Mar 3 at 0:41
  • 1
    $\begingroup$ Don't know, I did a failure analysis of this very problem once in a chicken feed processing plant some years ago but never found out if it has a special name. It had catastrophic consequences: metal fatigue leading to stress corrosion cracking in superheated steam of a stainless steel pressure flange which then exploded violently. They were lucky no one was in the boiler room at that moment- pieces of broken metal were stuck into the walls! $\endgroup$ – niels nielsen Mar 3 at 2:39
  • $\begingroup$ I've been thinking about the role that the expansion loop plays in this situation. it's designed to absorb thermal expansion. In this situation, I think it acts as a spring (or like a capacitor in an LC circuit) and actually creates a situation is which a resonance can exist (especially a second order harmonic resonance.) It adds elasticity, which the oscillation requires. $\endgroup$ – Hoytman Mar 5 at 3:19
  • $\begingroup$ I agree. Typically, someone laying out a piping run is not going to take that into account. $\endgroup$ – niels nielsen Mar 5 at 4:14
  • $\begingroup$ They're creating a little steam engine that just might take flight. $\endgroup$ – Hoytman Mar 5 at 15:29
1
$\begingroup$

I have worked with boilers and steam equipment for some years and have noticed some movement in condensate return lines, although usually not nearly as much as you describe.

Condensate lines are fed from steam traps that tend to be very cyclic in operation, so hot water (and a small amount of steam) is normally discharged in bursts a few seconds apart into the line.

This causes the water in the line to accelerate and decelerate as the bursts start and stop. The discharged steam will sometimes cause a bit of "banging" in the line due to rapid expansion followed by rapid condensation back to water, exacerbating the sudden water movements. This is common in steam plants but not normally a problem if the pipe is fairly thick walled and well supported so that the movement is not severe.

The line will react to the forces exerted on it by this water acceleration (F=ma) and will move if it is not well attached to something solid. A line that is long and/or inadequately supported will tend to oscillate or ring at its natural resonant frequency for a moment at each discharge, much like a guitar string that is plucked.

If the burst repeat rate (frequency) is near the natural resonant frequency of the line (or a multiple of it) then the amplitude of the oscillation can become very large.

Movement will cause stresses in the pipe and its supports at some points which can cause changes in the localised steel crystal structure resulting in stress fractures and failure over time if the movement is repetitive and severe enough.

Catastrophic failure (like the bridge) is usually caused by multiple failures, where something fractures or fails and the magnitude of the resulting stress exceeds the design strength of nearby structure elements, which then fail to take their share of the load and this then overloads the next element, etc.

Steel pipes are very tough so quite a lot of movement over a long time is usually needed, but it sounds like support failure might be possible in this case.

Damping the resonance (e.g. more/ better supports) will decrease the movement and therefore the stress peaks, probably worth doing. The line will not be under much pressure so a failure will not be particularly dangerous unless it or the (near boiling) water lands on someone.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.