CORROSION ALLOWANCE FOR TUBES IN SHELL AND TUBE HEAT EXCHANGERS

[brkmech1234]

I have done many heat ex-changer where there is no corrosion allowance specified by client even if the material is CS . This is also supported by API 660. I would like to why there is corrosion allowance required for tubes.

 

[rakuday]

Usually, corrosion allowance (CA) is not specified for tubes. Even TEMA does not specify CA for the tubes. During the design stage, fouling factor will be used, but this is NOT to be mixed up with CA.

 

[SJones]

Could it be that corrosion allowance is a contrary concept for heat transfer properties? How would one design the heat transfer characteristics: with the corrosion allowance; or assuming the allowance has been completely consumed? Is it also the case that there is some potential for 'by product' corrosion allowance if the standard tube thickness selected is thicker than required for pressure containment?

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[TD2K]

I've been curious about this myself but have never seen a good explanation. I think it's because tube thicknesses, relative to the typical corrosion allowances you see specified, are much less than the main pressure components. In addition, while a pressure vessel based on expected corrosion rates is typically designed to last between 10 to 20 years, it's not uncommon to retube heat exchangers, tubes are more of a disposable item.

Obviously, you select tubes that have a wall thickness greater than that required for the design conditions so there is a 'corrosion allowance' provided of sorts. I've also seen customer specifications that set tube wall thicknesses as a function of materials, if you want to use CS tubes the minimum thickness per the customer's specs is greater than say chrome alloy tubes or stainless steel which I've taken is to provide additional capacity to handle corrosion.

I doubt the heat transfer with 'uncorroded tubes' or 'hot and corroded' would vary much. The thermal conductivity of the tube wall material is so much greater than the inside and outside film coefficients and fouling factors (which are estimates in themselves) that the effect of tube wall thickness is minor. There are some exceptions, checking the performance of a 'clean' heat transfer with high heat transfer fluids, saying boiling water by condensing steam tube wall thickness would impact your results.

 

[SnTMan]

I've not seen anything forma on this myself, but TD2K has it about right. There are occasions where I have seen something like 0.010" CA on tubes.

Regards,

Mike

 

[rmw]

And not only corrosion allowance, but erosion allowance as well is accounted for in the tube actual wall thickness vs. the calculated required wall thickness.

Soft metal tubes in dirty fluid service often erode away faster than they corrode.

In most Hx's, the tube material is chosen to resist corroding effects of the fluid, but shells and channels aren't as sophisticated due to cost, hence a corrosion allowance is added to account for future wastage.

rmw

 

[rhg]

This is only an economical and financial decision.

Regards

rhg

 

[SnTMan]

Which is why metallurgies and CA's are most properly specified by the plant owners / operators, rather than equipment fabricators.

Regards,

Mike