Burke, an engineer with Valley Forge, Pa.-headquartered Houghton International, began by asking experts in the auto industry how they determine engine oil drain intervals. Its relatively straightforward, he heard, because engine oil life mainly depends on factors that can be measured, such as engine speed, load, stops and starts, temperature and oil system capacity.

I asked General Motors, how then did you come up with the oil life recommendations that you make? Burke recalled. They told me they added up all the failure mechanisms and assigned points to them. When you reach a certain level of points, its time to change the oil. OK, maybe thats putting it too simplistically. But if you can assign penalty points to the metalworking fluid failure mechanisms, and quantify them and see the interactions, I thought you should be able to create a Fluid Failure Index.

The idea is to spot when an operation is first getting into trouble, and take action before it can progress, Burke told LubesnGreases. That action may be to remove tramp oil or add a dollop of bactericide – or it may require a complete system dump. Understanding the factors that can drive a system to the brink of failure will help users make better choices about their fluid.

For example, is a particularly reactive metal being machined, such as some aluminum alloys? That rates more penalty points in Burkes view. Is the system reservoir the right size to allow the fluid to release its heat after each pass? That means less stress, and earns fewer points. Are bacteria counts kept as low as possible? If they climb, so do the points. Add up the points, and you can gauge how severe the operation is – and choose the fluid accordingly.

Could it Work?

Don Smolenski of GMs Worldwide Facilities Group in Pontiac, Mich., finds the concept plausible, and believes it could gain traction. Here at GM, its fun to see the Oil Life System in someones vehicle, and realize its grown to be in such a huge part of the fleet now in the U.S. and Canada. But what Johns doing is quite complicated compared to the Oil Life System, he said. With engine oil, you have standardized tests and specifications, which have been blessed and signed off on by the key users. With coolants, theres hardly any tests. There are so many processes, load factors, materials – even just within aluminum, youve got a huge number of alloys to consider.

Hes isolated some very good factors, but a whole lot of work still needs to go into it to prove it out, remarked Doug Hunsicker, of Caterpillars Advanced Materials Group in Peoria, Ill. Its something thats done with engine oil, so the idea is good – but metalworking is a wholly different animal.

Hunsicker said Caterpillar already tracks many of the factors highlighted by Burke. We plot everything out on each operating system, and have drawn up a complete history with 10 years of data. This is all gathered into our Chemical Reporting System, and includes all fluid changes and tankside additions.

John is going a few steps further. While Im a little skeptical on the predictive aspect of his proposal, I think you may be able to get relative ratios of how fluids will perform. So this is a good start, and an excellent concept. The next step is to get more users to try the index, put it into effect, and give feedback so it can be improved.

Rick Butler of Chemtool in Crystal Lake, Ill., also sees benefit in a Fluid Failure Index – if it works. Many chemical managers already track key fluid parameters such as concentration, conductivity, pH and bacteria levels, he observes. Trouble is, the report goes into a drawer and nobody acts on it. There seems to be a lot of monitoring, but not much attention being paid. An index that evaluates the entire process, not just the fluid, may help – or not.

A good question is, when to set the alert? Butler said. There are no standards for any aspect of metalworking fluids, its truly the Wild West. I think Johns proposed system would be good for sophisticated users, people with large systems. It may be harder to put into effect in smaller operations.

Butler said end users all know they should pay attention to failure signals, but most dont. More often I find that theyre already past the point of no return before they act, he said. Theyll run to the point of absolute fluid failure – and then theyll run two months more. People dont want to dump fluid, but Id like people to recognize that theyll have more downtime and more bad parts if they dont do it.

Seven Ways to Fail

Burke agreed that the sheer variety of metal removal operations complicates the Fluid Failure Index – but that doesnt makes it impossible. He identified seven main variables that stress the fluid. Of course, he noted, these failure mechanisms dont knock politely one at a time, but often gang up on a system. Failure factors to watch include:

1. Type of metal being machined. Cast iron, certain grades of aluminum and magnesium are among the most destructive; stainless steel is one of the least. How these metals and chips react with the coolant is one of the first things to evaluate.

2. Rate of metal being machined relative to coolant volume. The most destructive processes for fluid life, he notes, are grinding, polishing and ball grinding steps; mildest is single-point turning.

3. Type of water entering the system. Water quality is a near-mania for Burke, based on firsthand experience in the years he spent at Eaton Corp. before joining Houghton. Most money going through a metal cutting plant is spent on tools, chips, boiler chemistry and cooling tower chemistries, he said, but very few pay atention to the water chemistry going into their metalworking fluid. Its so basic, and so ignored. Water hardness and pH both earn prime spots in his rating system.

4. Amount of water entering the system per day. This is another indicator of stress. If 10 percent or more of the system volume is being lost daily through evaporation and usage, that signals a lot of energy and stress in the operation, which can affect even premium emulsified oils.

5. Tramp Oil. Users need to know the type and amount of hydraulic and way oils entering the system. These oils and also their additives – ZDDP, corrosion inhibitors, viscosity index improvers and more – build up in the metalworking fluid, to harmful effect.

6. Failure to control pH. The ideal range for most metal removal fluids is 9.0 to 9.2, Burke said. If it dips below that, or climbs too high, the system can be in trouble.

7. Biological control. Burke advocates the controlled and proper application of tankside microbicides, and insists that these require skill and training to use effectively. Under-dosing and over-application are both causes of failure.

Once these key stressers are rated by end users, they then can select the appropriate metalworking fluid for their operations best performance. Fluids are rated on their own scale, according to how stable and durable they are (see chart below). Correctly matched, this should let end users boost productivity or achieve longer fluid life.

Next Stop: Reality

Burkes Fluid Failure Index was described in part of a white paper on metalworking fluids, written by Neil Canter of Chemical Solutions and published this spring by the Society of Tribologists & Lubrication Engineers. It also was the subject of a presentation Burke made during STLEs annual meeting in Cleveland in May.

Selim Erhan of fluid and additive maker Polartech Inc., in Bedford Park, Ill., favors the idea – with a few caveats. The more calculations that are done up-front and the more customers are made aware of their choices, the better they can make real decisions about the cost of the fluid and overall operation, he said.

But its important, Erhan observed, to not single out the fluids role. Many failure factors are in the users hands, and not due to the fluid selection. We try to formulate with as much backup potential as you can, to handle all the water, speeds, temperatures, etc. Formulators build in reserve alkalinity. It also helps to choose the right additives, such as branched esters that are less appealing to bacteria than straight esters. That makes the additive last because it doesnt get consumed as easily by bacteria. Keeping the pH over 9 in the concentrate also helps with resistance.

There are so many human factors too, Erhan continued, including cigarettes being thrown into the sumps, and workers not making it to the bathroom so they use the return drains instead. All these contribute to bacteria growth and acidity. And if the user dilutes the concentration below a certain level, nothing works. In fact, that will be more expensive.

For example, we expect a fluid to be used at 5 percent concentration, so we may add corrosion inhibitors at a level that will work down to 3 percent, for safety. But if the operator dilutes it down to 2 percent, theyre going to get corrosion like crazy and bacteria counts will go through the roof. We have guidelines and they work well, but if they dont follow them, thats a problem.

Some have ideas for testing or improving the Fluid Failure Index. GMs Smolenski suggested, The analysis will need to be very sophisticated to be useful. Perhaps it could be illustrated as a spider chart? Then you could graphically see where one factor – the alloy – may increase machining difficulty, or how the stability of the oil may have an effect.

He also warned that a high number on the fluid stability chart isnt the answer to every problem. You could have a coolant thats rated with a high number for your type of machining operation, but it ranks poorly on water hardness tolerance, for example. Or a high-rated fluid may contain ingredients that a factorys waste-treatment plant rejects.

Disposal is a very important issue, Smolenski emphasized. It could be the factor that blows the whole chart out. For example, you may suggest using a biocide based on phenolics, because its good against a broad spectrum, but the waste treatment people dont want it so you cant even consider it. There could be some show-stoppers like that, where nine things on the chart are right, but number 10 runs into your local regulations.

Overall, I like the whole idea, and hope to see it move along, Smolenski reiterated.

Hunsicker, too, would like to see the Fluid Failure Index given time to evolve. What fluid is best, how to maintain it, how to improve the operation? Its all those things, he said. You need to know how severe the cut is, how severe the stresses on the fluid are, and everything else. People need to realize this is not a yardstick for the fluid – its a yardstick for everything.