During summer vacations, my wife and I stay in a cottage that draws water from a clean lake about 30 feet away. A reciprocating Southern-Deming Model S-250 pump has worked well for more than 40 summers, and we have kept the pump and motor well maintained even though we use it only for a few weeks a year. A trip to the nearest hardware store or pump-repair business takes some time, so we have spare parts on hand -- drive belt, seals, plungers, fittings, and so on. I've replaced the leather plungers several times, the drive belt once, and have had to create a new line to the pressure switch, so it made sense to prepare for a pump failure.
This year when I primed the pump, I got a nasty surprise. The connection between the steel water-inlet pipe and the cast-iron pump body had completely rusted out. I never expected that connection to rust away and fail, because it always seemed solid. Unfortunately, it would have cost more to repair the faithful pump than to replace it, so we bought a new and simpler pump, as shown in the photo.
The point here is that the most "solid" part of a device can fail even though we test it thoroughly and don't expect it to fail over the life of a product. In mechatronic systems, it can pay to subject components and complete systems to highly accelerated life test (HALT) and highly accelerate stress screen (HASS) regimens.
The electronics industry has applied these techniques for many years, and you can find a series of Web pages that describe these types of tests. Although this information and other resources on the Web apply mainly to electronic components, you can apply the underlying principles with mechatronic designs, too. Even if you contract with a testing lab to perform accelerated testing, you should know what tests you need to run, and how you want to run them.
Years ago, the US military created a handbook, "Reliability Prediction of Electronic Equipment," MIL-HDBK-217, now at revision F, and available as a pdf. This information pertains to components such as op-amps, memories, and logic devices, but it provides useful reliability-prediction guidance that can apply to other types of devices, too.
Accelerated and stress testing often involve what engineers call a "shake-and-bake" environment that literally vibrates equipment and exposes it to temperature extremes. Formally, engineers know this area as environmental stress screening (ESS). Here is a detailed "Environmental Stress Screening Tutorial." Find other ESS information here.
Sorry to hear about your pump, Jon. Aren't the so-called "shake and bake" tests a standard best practice for mechanical parts and assemblies or is that just another example of a no-brainer routine falling through the cracks due to engineering short cuts?
Wouldn't it be difficult to determine the life of a component against corrosion in a "shake & bake" test? How would you accelerate corrosion and still get a meaningful result for component life-time?
People would run special corrosion tests, which many commercial test labs know how to perform. The "shake-and-bake" tests look for structural weaknesses at various frequencies and g levels, while the environmental temperature tests look for premature electrical and electronic failures as well as failures induced by the stresses and strains created by heating and cooling. --Jon Titus
Jon, I understand the difference between vibration, temperature, and corrosion tests, and know there are standards for each.
Applying the results of a vibration and temperature tests are MUCH easier to deciding the life of a component than it is to using the corrosion results.
The specifications for a component will say "up to 50g, 155deg F". It may even say something about the 20% salt bath immersion for 20 days.
I saw a new sink fixture component corrode away in 5 years on city water when I was a kid. I think it would be difficult to relate the corrosion test results to city water to come up with an expected lifetime.
I have seen similar failures due to conditions we can't easily test for. Your sink drain provides a good example, although I'd think the manufacturer--knowing the use conditions--would have tested drains under different water conditions. Maybe a cheap knock-off manufacturer went for low cost instead of quality. --Jon
Those types of tests should be part of a comprehensive testing program, but many small companies lack a strong background in testing or they lack the time and money to have a contract company run them. Also, when a company custom builds mechatronic equipment, they might not think such tests deserve attention for one-off designs. Also, I doubt many colleges or universities teach engineers about real-world product or system tests. --Jon
On a related subject, the US National Institute of Science and Technology (NIST) recently published information about the failures of carbon nanotubes used as electrical conductors in microcircuits. Tests showed failures of the nanotubes after only 40 hours. Although many researchers have touted the capabilities of these carbon structures to carry high currents, they seem to have problems that deserve more research before companies use them in integrated circuits. You can read the complete article at: http://www.nist.gov/public_affairs/tech-beat/tb20110816.cfm#cnt. --Jon Titus
Was the pipe carbon steel or stainless steel? I wouldn't expect a galvanic couple between carbon steel and cast iron, but depending on the type of stainless, there could be a significant galvanic couple between stainless and cast iron. A plastic or rubber isolator between the two could help prevent this.
I didn't install the original pump so I assume the pipe was carbon steel. I know it wasn't stainless steel. The pump body was cast iron and we run polyethylene pipe down to the lake. We thoroughly drained the pipes and the pump at the end of the summer. Perhaps the pipe-joint compound used by the installer dried up and let water into the pipe-to-pump threads, which caused the rust over the years. --Jon
Might help to keep the housing filled with water during the offseason - if that is possible. Oil would work better - however this is drinking water.
I had a similar problem with a pool pump that only runs occasionally to power a fountian. The iron outlet pipe will oxidize to the point that it restricts flow.
Seems like continuis running pumps suffer less corrosion due to a lack of oxygen and wet/dry cycles.
In this case it is a matter of more expensive stainless steel, zinc, or aluminum housings, or perhaps plastics such as PVC or Nylon
Or the iron housing, with a constant battle against corrosion...
I have done repair work on a very large range of things, and I have looked at various tests, such as the military salt spray test.
What I see is that material choice controls the rate of corrosion, and the test can be done at a desk prior to the first physical model. Some steel alloys will corrode very rapidly, and some assemblies will fail due to rust very quickly, simply because of how they are made. Cheap alloys or grades of steel that are cold-worked seem to be the most likely to fail, and if the design does not include a means to keep moisture away from them, they certainly will fail. There is no need to do actual testing to prove that the failure will occur, the determination can be made simply by inspection.
In the case of you pump, Jon, I wonder if the part that failed was threaded into the cast iron, or just pressed in? A threaded in part that failed would potentially allow the mating part to have the remains removed and the connection cleaned up with a pipe tap, and a new fitting installed. Use of a galvanized fitting would probably have extended the sys6tem life many years, even if the rust-failure area were not galvanized. And, of course, a heavy coating of moisture proof sealant would also have improved the lifetime.
Hi, William. The pipe threaded into the cast-iron pump body that had a threaded hole. The plumber used pipe-joint compound when he attached the pipe in the late 1960's. The threads on the pipe and on the pump housing were gone and all I saw was a rusty mass on the end of the pipe and an uneven hole on the pump. No threads to clean out. Sorry I didn't take a photo, but we had a lot going on that day. There was nothing to salvage. If I knew of a nearby machine shop, I coulset up and maintain seemed like the better course. Time will tell. --Jon
Jon, Then the repair method would be to silver-solder in a brass or copper fitting, after a great deal of cleaning up-probably a 3-day soak in an oxalic-acid type of cleaner, such as some radiator flush. That would have left you with a quite durable connection. If you have kept the old pump, you could repair it for when the new one fails.
I cannot understand the concern for a rapid test to predict the life of a low carbon steel - cast iron couple (obviously different metals) in a vibratory (spinning un- balanced rotor) in fresh lake water (high akali metal salt content if hard water) that lasted forty years of two weeks or so wet and fifty or so weeks dry (?) per cycle. Was there an expansion tank in the system to absorb water hammer on shut off (high tensile shock)? I'll bet your new pump won"t do as well. Without knowing more about your actual layout, speculating is vanity.
I used the pump as an example of an unexpected failure and as a way to introduce the topic of testing. I would bet, though, that companies that make automobile, aircraft, and ship pumps do subject them to vibration testing. --Jon
My unstated point was that after forty years in such a complex environment failure should be expected. Unexpected failure is after one week or even one day. As time passes failure becomes more likely. Most failures are due to ignorance of exposure margins, stress or temperature cycles, time, misapplication, etc. A notable example is our recent national economic collapse brought on by the abused housing market, a trusted source of capital so good that bonds based on home mortgages were rated AAA as a matter of course over scores of years experience until an environment of greed and lax Federal supervision combined with legislative excess and criminal behavior interfered. Each of the corrosive activities on its own were recognized as dangerous causes of risk of failure. The barrier to reason was the cognitive dissonance of that degree of collapse. The predatory banks are protected by the too big to fail provenance dispensed by our collaborative Treasury Department and our similar Securities and Exchange Commission while all home owners are victims of the loss of equity. If failure of the economy had been expected by the appropriate individuals, investigations would have been conducted to defend against it, much like your stock of repair parts which might have included an entirely new pump, as well.
I am surprised that a junction of steel and cast iron would corrode, but certainly there was a great deal of elapsed time. If there dissimilar metals were connects, eg brass or copper against steel, galvanic corrosion would occur much faster.
However, the real reason I wrote this note is that citing MIL 217 is bad advice. That spec is 40 years old and is flat out wrong regarding the physics. Conversely, JEDEC JEP122 is up to date and physically correct.
While I realize the pump example was a lead-in to the point of the article I wonder if you were a plumber you would have seen this coming?
"I never expected that connection to rust away and fail, because it always seemed solid." "Seemed solid" is key here.
Every time I have to do some plumbing I ask questions. My experience with DIY hot water heater repair and advice from my plumbing store would have had me wondering about cast iron anything. I was about to add a pressure regulator using cast iron fittings from Home Despot and was at my real plumbing house buying a hot water heater because I couldn't get the anode rod out of my current one. The plumbing store told me the cast iron would rust away. Something not on my radar. I switched to brass.
I bought the new HW heater (it's not really a hot water heater, now is it?) and before tearing the old one out I gave the anode rod another try and voila! It came out. I replaced it. Now I have a brand new spare cold water heater waiting in line. Not a bad investment, as the prices have doubled in the last few years. But still wasn't on my radar.
Generally used for Avation / military products. I would assume (?) any other similar specifications from CE (or equ) would be similar. Namely, a primary focus (say vibration or shock) and a expected environment (helicopter?) with a specified range expected (amplitude vs frequency limits).
Bottom line for a given application: there are likely limits already established (from a appropriate agency)..
Assuming the manufacturer/designer is paying attention!
As for a pump working 40 years.. how good does it have to be?
The Machinist Calc Pro computes speeds and feed rates for milling, turning, and drilling: cutting speed, spindle speed, feed rate (inches/minute), cutting feed, etc.
During a recent meeting with engineering-school faculty and alumni, Contributing Technical Editor Jon Titus talked about whether colleges should educate generalists or specialists. What do you think?
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 3
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
A quick look into the merger of two powerhouse 3D printing OEMs and the new leader in rapid prototyping solutions, Stratasys. The industrial revolution is now led by 3D printing and engineers are given the opportunity to fully maximize their design capabilities, reduce their time-to-market and functionally test prototypes cheaper, faster and easier. Bruce Bradshaw, Director of Marketing in North America, will explore the large product offering and variety of materials that will help CAD designers articulate their product design with actual, physical prototypes. This broadcast will dive deep into technical information including application specific stories from real world customers and their experiences with 3D printing. 3D Printing is
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
2 
