Tube and Shell HX - Solution to Tube Vibration 5

[TurbineBlade]

Hi,

I have a split flow tube and shell HX that is subject to potential vibration problem due to design deficiency. The secondary side flow is designed to 600 L/s. Right now, the plant is operating at much lower flow to avoid the vibration problem. What are some of the potential solutions to allow the HX to accept full flow?

 

[MikeHalloran]

Ask the HX supplier to fix it or replace it?



Mike Halloran
Pembroke Pines, FL, USA

 

[rmw]

Give us some details on the Hx. Are we talking about a FWH, a MSR, a gland exhaust condenser, or what?

Is the tube bundle removable?

Help us help you.

rmw

 

[StoneCold]

If the tube bundle is removable you might be able to add rod baffles to stop the vibration. You might also be able to put inpingment plates in the inlet nozzles for the shell to stop the vibration.
We need more details. How big is it?
Cost to replace? Would a tube failure be catastrophic? Is it a high pressure unit? etc.

Regards
StoneCold

 

[TurbineBlade]

Hi,

This is a split flow 4-passes cooling heat exchanger. (22 ft tall, 80 inch dia.) with water in both sides for low pressure and low temperature application.
Primary side Operating Condition:
P=0.7 Mpa(g), Tinlet = 66C, Primary Flowrate 516.4 kg/s.the Secondary side Operating Condition:
P=1.03 MPa(g), Tinlet=19C, Shellside Flow Rate: 470 L/s

The tubes bundles are removable. A tube failure is undesirable as it will cause a lost of plant production time.

 

[Peregrino7]

I hope this information will help find the right solution:


When a heat exchanger has a flow-induced vibration problem, some corrective actions are available to reduce or eliminate the problem. First, verify that the vibration arises from shellside flow, not from some other external source. After you ascertain that the vibration is flow-induced, consider the following actions:

i Plug leaking tubes Often, one of the first indications of a tube vibration problem is that some of the tubes are leaking. The expedient response is to plug the leaking tubes and/or to weld the seal in place. Replacing tubes in the field is generally not practical. While this solution does not address the problem of flow-induced vibration, it can permit continued operation of the heat exchanger until the next maintenance turnaround.

i Remove tubes to create bypass lanes For a heat exchanger with a tube vibration problem caused by fluidelastic instability, creating artificial bypass lanes in the window area is an effective temporary remedy. First, remove tubes in the window areas to create a lane from the baffle tip to the shell parallel to the direction of flow. Then plug the tube sheet where tubes were removed. Allow additional bypassing; do not seal the holes in the baffles. This temporary solution, which lowers shellside pressure drop and reduces shellside thermal performance, is particularly attractive when the heat exchanger is tubeside heat transfer limited.

i Reduce shellside flow rate Because flow-induced vibration depends strongly upon velocity, reducing the shellside flow rate temporarily minimizes the vibration problem. This solution is acceptable only if the reduction can be tolerated within plant operating requirements.

i Stiffen the bundle Increase natural frequencies of an existing tube bundle by inserting lacing or driving wedges between tubes, thus restricting tube motion and preventing wear. Lacing and wedging are often used in the U-bend of U-tube bundles, which can be particularly susceptible to tube vibration problems. Thermal and pressure drop performance can be significantly altered, depending upon the way bundle stiffening is accomplished.

i Roll tubes in baffles near nozzles Rolling tubes in the region where they penetrate baffles is a tricky and expensive operation. However, it has successfully eliminated tube vibration problems originating from nozzle flow in end zones.

i Add deresonating baffles When the problem is acoustic vibration, pull the bundle, remove selected tubes, and add deresonating baffles. If resonant frequency measurements have been made, the deresonating baffles can be positioned precisely to correct the problem. Loss of tubes somewhat reduces thermal performance of the heat exchanger.

i Remove tubes to control acoustic vibration Another approach is selective removal of tubes from the bundle. Measure acoustic vibration characteristics of the heat exchanger before designing the solution. The tubes to be removed depend upon the bundle’s geometric arrangement. This approach has been particularly successful with double-segmental baffled heat exchangers where tubes have been removed in the tube overlap region. Because the number of tubes removed is small, the loss in thermal performance is small.

i Replace tube bundle At times, the only solution to a flow-induced vibration problem is to redesign and install a new bundle. Because maintaining the same shell and external piping is often desirable, the design configuration is limited to the existing envelope; however, considerably different internal features can be involved. If the replacement bundle is to use a different tube material and/or baffle arrangement, performing a vibration analysis ensures no new vibration problem


GH

 

[TurbineBlade]

Hi Peregrino7,

Thanks for the response.

The HX is subjected to flow induced vibration. The tubes in the vicinity of the tube free lane area are particularly susceptible to FIV. It looks like inserting lacing or driving wedges between tubes, or remove tubes are some of the solutions. Are you aware of any reputable companies in Canada/U.S. that perform FIV analysis and can provide design solutions?

Patrick



How to you detect if the retorfit

 

[ggrindle]

There are several companies that offer vibration mitigation productions. The Atlantic Group for one. My understanding is that they essentially jam a piece of plastic or foam or brass, depending on the application and thereby damper the vibration by holding the tube in certain modes.

Temperature?

 

[EdStainless]

The Atlantic Cradle-lock does work. It is better than traditional stakes, they stay in place better.

http://www.atlanticgroup.com/anti_vibes.htm

Is there anything that you can do with baffles to redirect some flow? You must either have very wide support spacing or real high cross flow velocity.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[TurbineBlade]

HI,

This is a D2O-to-H2O HX. It is a 4-passes, split flow HX with U-tube bundles. The HX has been operating for 23 yrs with the shell side flow at around 420 L/s. The shell side outlet temperature is 40C. The tube side outlet temperature is ~40C. We want to shell side to accept ~630 L/s (i.e., 100% of the design flow).

Modification using sealing strips/rods to divert flow into the periphery of the HX was done in the past. The original HX already has an impact plate at the shell inlet nozzles with anit-vibration baffle. Despite that, past analyses showed that the HX will subject to FIV if operates in excess of 550 L/s for long period of time. It appears that only the tubes in the vincity of the tube free lanes are the ones that suspectible to FIV.

There are 11 baffles and each is 16" apart.


Thanks

 

[EdStainless]

What is the tube size and material? How loose are the baffel holes? I don't doubt you, but it really sound fishey. Maybe they haven't told you everything.

If you rebuild the bundle look seriously at rod baffle designs. You have to pay a premium, but you get good support with less negitive impact on the shell side flow.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[TurbineBlade]

Hi,

The tube O.D. is 0.625in and is made of Sandvik 3RE-60 (ASME-A669). I need to check the baffle hole size. Pls elobrate on the fishey part?

Can't rebuild the bundle, too expansive.

 

[EdStainless]

Liquid flow may have high density, but the velocities are low. For someone to have missed the design that much doesn't sound right. Are there crossflow vibration calcs in the original design package?
I see vibration in HX with high velocity gas cross flow, but rarely in liquids. They dampen so much more.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm