CS vs SS corrosion mechanisms

[jalvarez]

Colleagues:
I need to define the material for a pump; the fluid is a brackish solution, NaCl content of 1%. Some other salts like sulfates. The pH is neutral. The temperature is 95°F.
Looking at a corrosion design curve for brackish waters, I verified that for Carbon Steel, a straight line is defined containing the following two points
20 mpy at 81°F
45 mpy at 137ºF
Further, this curve is defined at a velocity of 5 ft/sec. And a comment in the paper adds that the corrosion rate would be lower, 5 mpy at ambient temperature, let’s say 81ºF, if there is no flow.
Now, from other thread (thread338-100636) I posted on Aug 4th, all the comments are oriented to avoid stagnant conditions of the water when speaking about stainless steels.
This is an interesting difference. The corrosion mechanisms must be different.
Comments??
I'm oriented to cast iron as the selected material for the pump, accepting some corrosion. From the mentioned thread, SS is not a correct option for 1% NaCl.
Have a safe day
J.Alvarez

 

[EdStainless]

The sensitivity of a material to flow related effects is a function of the strength of the protective passive surface films. In materials where these films are weak or soluble you see the corrosion rate have a great dependence on velocity. In materials where the surface films are very tough and stable (stainless) you are more worried then about local corrosion issues such as fouling and stagnation leading to a galvanic condition that would break the passive layer.

A duplex stainless would be a good solution for your pump. They will handle the Cl, are strong enough to offer some erosion resitance and not very suseptable to stress cracking.
If you want a more traditional cast alloy try NiResist. This austenitic cast iron has been used in brackish water pumps for decades. It does not have great erosion resistance, but it has good corrosion resistance.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm