The Case of the Gummy Gears

A failure in a military gear pump underscores why two soft materials should never be rubbed against one another

Ken Russell, Contributing Writer

Human society uses a variety of pumps. These include reciprocating pumps that raise water and oil from underground, rotary-vane coolant pumps used for automobile engines and gear pumps shown schematically in the figure, above right. Each pump type has its own characteristic strengths and weaknesses with regard to such things as size, power consumption, maximum pressure difference and output.
The gear pump in the current case pumped a solution of ethylene glycol and water (automobile coolant) to cool a laser in a space application. I suspect the application was military, but was never told.
The gears in the pump rub against one another during operation. The art and science of such rubbing comes under the term “tribology.” Designing two parts to rub against one another without galling, excessive wear or some other problem is a daunting task.
One should never rub two soft materials against one another. An earlier Design News article (”The Case of the Croupy Carburetor“) illustrates this rule. A no-name kit was used in the rebuilding of the carburetor for a small airplane. The original equipment was a hard stainless-steel shaft in a soft stainless bushing. The kit provided a soft shaft and soft bushing. The shaft and bushing soon galled and seized. The resulting loss of power led to a crash and a lost life.
A further example is the mostly (totally?) unsuccessful attempts by auto manufacturers to run aluminum alloy pistons against aluminum alloy blocks. (The piston rings are steel or cast iron.) The aluminum alloys are fairly soft and the result has been a series of failures. The usual solution is to fit a cast-iron sleeve into the aluminum block. The aluminum piston-cast iron cylinder combination has been working fine since at least the 1940s, when a Big Three automobile manufacturer advertised that its “four ring aluminum piston saves oil and saves gas and gives power to pass.”
In this case the gears and case were made of a medium hardness stainless steel. To reduce galling, the surface was implanted with nitrogen which combined with the chromium in the stainless steel to form very fine chromium nitride particles and give a very hard layer. The hard surfacing prevented galling, but opened the door for corrosion. Storing the pumps with the coolant in place resulted in heavy pitting and corrosion of the tooth surfaces and premature failure.
Stainless steel is stainless because of 12-plus percent chromium dissolved in the mostly iron matrix. Unfortunately, formation of the chromium nitride particles effectively gettered the matrix surface of chromium. The resulting matrix corroded like mild steel, which is to say, rapidly. As in the old painkiller ads, the nitriding traded a headache for an upset stomach.
Corrosion did not occur if the pump was kept running and the fluid de-ionized. Keeping the pumps running from the introduction of coolant until the start of operation was not an attractive solution. The pumps simply had to withstand a static coolant environment.
After poring through the material in the case file, I opined that the steel in use was not satisfactory and could not be made to be satisfactory. I suggested a high-carbon stainless of the sort widely used for bearings and cutlery. This stuff is hard enough to resist galling and is adequately corrosion resistant for the pump application. I do not know whether my advice was followed. I suspect not, as the people involved were in an awful hurry.