Excessive shaft endplay and runout can cause premature failure of bearings or mechanical seals. Some of these concepts were discussed in mechanical seal root cause failure analysis as potential reasons for seal failure. It is important to check endplay and runout periodically to ensure they are within the allowed tolerances.

Endplay is the axial movement of the shaft, or a movement towards the casing and back towards the motor. Some endplay is required to give the bearings clearance to create a thin oil layer or for thermal expansion. Too much endplay can lead to pitting and fretting around the mechanical seal area.

Runout is the radial movement of the shaft, i.e. the shaft wobbles from center as it rotates. This can occur from a bent shaft, unbalanced parts like the impeller, misalignment of the pump, or even deflection from cavitations. Too much runout will result in bearing damage, which will lead to vibration and seal failure.

To measure the endplay, put a dial indicator on any step in the shaft. Push the shaft in one direction as far as you can, and set to dial indicator to 0. Then push in the other direction and get the reading off the indicator. This value is your endplay. Most ANSI pumps require endplay to be equal to or less than .005” T.I.R., which stands for total indicator reading.

To measure the runout, take measurements of the shaft near the stuffing box, and another measurement behind the thrust bearing (the motor side). Rotate the shaft and note the highest and lowest measurements. The difference is the runout. If your pump has a sleeve, measure with the sleeve on the shaft. The tolerance for runout is 0.001."

If you plan to oil lubricate your pump, it is important to select the proper oil. Using the right oil for the job can extend the life of your bearings and prevent unexpected down time.

So what oil should you use for your pump? In most cases, you'll want a high quality turbine oil with ISO viscosity grade of 68. This viscosity level covers the majority of normal operational conditions in which bearing temperatures run between 120°F (50°C) and 180°F (82°C). For higher temperatures, consider ISO grade 100.

Turbine oil contains rust and oxidation inhibiting additives which help extend the life of the pump. Do not use other types of oils like the ones you can find for the auto industry. Their formulation are not designed for pumps and can negatively impact your bearing life.

Most pump manufacturers will list the recommended oil brands in their IOM (installation, Operation, Maintenance manual). Here are a few recommended by the industry:

Do a local search for "oil distributors" in your area, who may stock some of these brands.

Root cause failure analysis for pump packing is similar to RCFA for mechanical seals. Just using your eyes and this guide, you can identify with confidence why the packing failed. Then, you can determine if a change is necessary and/or cost effective.

Potential changes include:

• Using a different style or material of packing
• Cutting the rings differently
• Buying them pre-cut and pre-formed
• How the packing is installed
• How the packing is “broken in”
• Use of environmental controls like circulation or flushing

Here are the steps to root cause failure analysis for packing.

1. Keep all the rings of packing in the order they were removed.

2. Try to determine how long the pump ran before the packing failed.

3. If the packing looks or smells like it was burnt, the packing experienced temperatures higher than the what the packing could handle. This could be a result of:

   • The pump running dry
   • The packing being installed too tightly
   • The process fluid temperature itself
   • Wrong choice in packing material
   • Combination of these

   For example, some packing made from poly-acrylonitrile (PAN) can only handle a few hundred degrees F. Between the process fluid temperature and the additional heat from the friction of the packing, this temperature limit can easily be exceeded. It is important to select the proper packing material that can not only handle the temperature of the process fluid, but also the additional heat from friction.

4. Measure the packing from the inside to the outside and from the front to the back in several places.

   The reason you should measure the packing in several places is to confirm if the shaft is centered in the stuffing box. It is not uncommon for the packing to compress a little from the front to the back in service.

   If the packing is thin from the inside to the outside, it can mean several things:

   • It can be the wrong size packing
   • Shaft whip or runout

   Double check to make sure you are using the right size packing for your pump. For an STX pump, you will need 5/16" square packing. For an MTX pump, you will need 3/8" packing. In general, you can arrive at the correct packing size by first measuring the diameter of the stuffing box, then subtract the shaft/sleeve diameter, then divide that number by 2. For an STX standard bore:

   • standard stuffing box bore diameter: 2.00"
   • shaft diameter: 1.375"
   • 2.00 - 1.375 = 0.625
   • 0.625 / 2 = 0.3125 (5/16")

5. Examine the first ring of packing, the one closest to the throat of the stuffing box. The throat is the restricted part of the stuffing box at the opposite end from the gland. If the first ring has a feathered area where the packing has thinned out to an almost fabric-like condition, it indicates the throat may have opened up past the .030” on a side typical clearance. An easy fix here would be to provide a split washer to put at the throat of the stuffing box. The outside diameter of this bushing just needs to be a slip fit to the inside diameter of the stuffing box. The inside diameter as mentioned before should be .030” on a side larger than the shaft or sleeve.

6. Examine the inside of each packing ring. If it very hard, like a glaze, it is evidence that there was too much friction between the packing and the shaft or sleeve. This can be caused by installing the packing too tightly or a combination of tight packing and high operating temperatures. Most, not all, packing needs to weep or drip a very small amount of fluid. This insures that the process fluid or the flush is sufficient to lubricate the packing. While some modern packings such as pure flexible graphite are self-lubricated, most packings have a small amount of break-in surface lubrication plus some internal lubrication. When that is gone, you are left with the fibers of the packing itself. These may not be enough to seal but may be enough to damage the shaft or sleeve they run on. Damage from packing is called scoring.

7. Examine the packing for torn fibers. Most packing is braided. The most work is done at the corners of the packing. This is why some packings are made of more than one fiber with a particular fiber like PTFE or Kevlar at the corners. If the fibers are torn the cause can be an extremely abrasive process fluid. However, be careful if you are considering switching to a partial or all-Kevlar packing as Kevlar can be extremely harsh on shafts and sleeves.

Measure the seal face and measure the wear track. The seal faces are the lapped parts of the seal that rub against each other. Measuring the wear track isn’t always easy as the wear track on a seal face is not always distinct. Therefore, take several measurements of the wear track. Keep in mind that double seals will have two sets of rotary and stationary faces, so keep each set together.

• If the wear track is the same width as the seal face that made the marks, it tells you that the pump was running true. In RCFA, finding out what is right is just as useful as finding out what is wrong. It eliminates guessing.

  

• If the wear track is wider than the seal face that made the mark, the pump experienced runout or radial movement.

  

  Subtract the width of the seal face from the width of the wear track and you have absolute proof of the exact movement the seal experienced. Since ANSI pumps call out a maximum runout of just .005”, you can easily determine if your pump is within that limit.

  Excessive runout may result from:

  • Poor pump alignment (a critical requirement, especially for ANSI-type overhung impeller pumps)
  • A bent shaft
  • Deflection caused by cavitation

  Corrections include proper pump alignment using the laser method rather than eyeballing or using a straightedge. Selecting the right coupling helps in proper alignment. The use of expansion joints reduces or eliminates pipe strain. It is said that if you remove the fasteners at a flange and there is movement greater than 1/32nd of an inch, there is too much pipe strain.

• If the wear track is narrower than the seal face that made the mark, the mechanical seal was over-compressed. This causes the seal to bow.

  

  By measuring the outer diameter of the mark, it will probably be the same reading as the outside diameter of the seal face. The inner diameter of the mark will normally be larger than the inside diameter of the seal face. This is because when the seal bows, it tends to make contact towards the outside of the seal face. Making sure the seal is at the correct position on the shaft is one requirement. However, some equipment thrusts the shaft on startup and this must be taken into consideration.

• No wear track means there was either no contact at all or the rotary seal was slipping and not turning with the shaft.

  

  No contact at all might mean that the mechanical seal rotary is not at the correct location on the shaft and therefore not making proper contact with the stationary. A slipping seal is the results of loose set-screws or, in the case or a rubber bellows seal, possibly an incorrect lubricant was used. Do not use a lubricant that can contaminate your process fluid. Do not use silicone or PTFE based lubricants. They can remain under a secondary seal like an O-rings and allow process fluid under pressure to channel underneath the O-ring.

• If you see a wear track in an arc and not all around the stationary ring, it means the faces did not run flat against each other.

  

  Look for an anti-rotation pin in the stationary face. The anti-rotation pin must be shorter than the depth of the groove it fits into. Sometimes, the pin is not even in the groove. This cocks the stationary and makes for a partial wear track. If the anti-rotation pin is too long, you can just trim it down. In most cases you can remove the pin and allow the secondary seal to keep the piece from rotating.

• If you see chipping on the inner edge, outer edge or both, it is a sign of flashing.

  

  Flashing is caused when water is trapped between the seal faces, gets hot, converts to steam and the face blow open. Then, the pressure drops rapidly and the faces slam back together, over and over. The remedy here is to reduce the temperature. Sometimes, just a few degrees is required. Sometimes, an additional environmental cooling control is needed.

• If a seal face is peeled or flaked, it was most likely a coated surface.

  

  For example, Stellite is a hard-coating over stainless steel. A seal face can heat up to the point where the differences in the thermal coefficients of expansion for stainless steel and Stellite cause the Stellite face to crack. Usually, this leaves an unrepairable seal face. Heat can also cause the bond between the substrate and the seal face to separate. Simple fixes include using solid material seal faces such as silicon-carbide or tungsten-carbide.

• If a seal face is blistered or pitted, it is usually due to the material being of lower quality.

  

  Seal faces like ceramic, carbon, silicon-carbide and tungsten-carbide start life as a blend of dry materials. The ingredients are mixed in a ball mill to grind the particles to a very small and uniform size. The longer it takes the more money it costs. When less time is taken, the result may be a grain size ranging from 5-50 micro-inches in diameter rather than a smaller, more uniform 3-5 micro-inch diameter. Larger, less uniform chunks of material can just pop out of a seal face in service. In addition, carbon, the most common seal face material, should be impregnated rather than just machined from raw carbon stock. Finally, some process fluids will actual attack carbon. In these situations, alternative seal face materials should be used. These may be a PTFE based material. While not as reliable long-term as carbon in many applications, PTFE’s resistance to more aggressive chemicals might extend seal life compared to carbon.

• A broken ceramic face could be the result of a cold shock or heat shock if the ceramic grade is commercial rather than high-purity.

  

  The difference between commercial grade and high purity is the aforementioned process of making the ceramic out of smaller and more uniform grains of material. Also, in the production of seal faces, sharp edges should be slightly chamfered or “broken”. This prevents stress risers and premature failure. When a ceramic is driven by pins, this can cause mechanical shock which can also break a ceramic seal face.

• If you see a worn area or channel in a seal face, look for a circulation or flush port that could be directing high-pressure fluid directly onto a seal face rather than tangentially.

Conducting root cause failure analysis on a failed mechanical seal can in the long run minimize mean time between failure (MTBF) and operating costs by preventing repeat premature failure. If you can find the culprit and resolve the issues, you can prolong the life of the next mechanical seal and save time and money.

A tremendous amount of useful information is available from a failed mechanical seal. With nothing more than your eyes and a measuring device like a dial or digital caliper, you can identify failure modes with great dependability and detail. With this information you can decide if any corrective measures are worth the resources required. In many cases, the payback is many times the outlay in a very short period of time.

And because RCFA (Root Cause Failure Analysis) provides evidence rather than conjecture, it can support the changes needed when other departments are involved. Following are the steps to examine a failed mechanical seal.

1. Keep and examine the entire seal assembly.

   Whether seal parts or entire seals will be repaired or not, whether they are repairable or not, they have much information to reveal.

2. Don't clean the seal.

   Unless required by Process Safety Management, don’t clean the seal before examining it. Cleaning may wash away critical evidence.

3. Examine the pump for clues.

   Whenever possible, examine shafts, sleeves, bearings, gaskets and other pump parts. Doing this, you can support your interpretation of what you see on the seal parts or even change it as necessary.

   • Shafts and Sleeves

     Look for fretting - damage caused by O-rings. Measure the diameter to ensure dimensional tolerance of +/- 0.002."

   • Bearings

     If it is discolored black, it is evidence of heat. If it is discolored brown, it is evidence of contamination.

   • Gaskets

     Look to see if gaskets are hard, dissolved, missing, or extruded.

   • Other Pump Parts

     Look at the impeller, casing, and rear cover for excessive wear. Compare dimensions with tolerances show in the IOM.

4. Try to get an idea of how long the seal ran before it leaked.

   Knowing when the seal leaked narrows down the cause of failure. The critical division is whether the seal leaked when liquid was brought to it or before start up, or after the pump started running

   Why is this important? If it leaked when initially flooded or before startup, it was due to one or more of the following causes:

   • One or more seal faces were not flat enough to seal.

     Normal seal flatness is just 2 helium light-bands or 23 millionths of an inch! Corrective action here would be to test the seal assembly in a pressure testing rig before installation and/or to inspect the seal faces with a monochromatic light and an optical flat.

   • A secondary sealing mechanism failed

     A gasket or O-ring was damaged during installation, is missing, or is the wrong size. Shaft O-rings can get cut during installation or it can even twist and roll during installation, opening up a path of leakage.

   • The seal is not in the proper location on the shaft.

     Most seals require a working length with only plus or minus .030” tolerance. The working length must come from the factory. There is almost no reliable method to determine the working length of a seal when in the field. This is one reason why cartridge (pre-measured) seals have become so popular.

   • Some other seal part is missing, the wrong size or installed incorrectly.

     If you have a PTFE wedge seal design, the wedge can be in backwards or be an incorrect size. An O-ring can be the wrong cord diameter. An O-ring groove could have been machined incorrectly.

   Conversely, if the seal leaked after startup, it was not due to one or more of those four issues. You can eliminate these four issues and seek other causes.

5. Examine the wear track on the seal faces.

   This is the mark made by the mechanical seal rotating face rubbing against the stationary face. The seal faces are the lapped parts of the seal that rub against each other. The wear tracks may contain clues that may identify the reason the seal failed. Double seals will have two sets of rotary and stationary faces, so check them both (but keep each set together)

   Refer to the diagrams inRoot Cause Failure Analysis - Interpret Wear Tracks, where we show you what to look for when examining the wear tracks.

6. Inspect the mechanism that rotates the seal.

   Look for signs of wear on the rotating mechanism. This could be pins, lugs, set-screws, a spring or rubber or metal bellows. These parts can be worn when the faces are pressed together too tightly. The faces jam together then break free. This repeats over and over until the drive lugs or pins get worn. This happens when there is insufficient lubrication between the faces.

   • Check to make sure the mechanical seal is not over-compressed.

     The working length of the seal must be provided by the supplier. It is almost impossible to determine the correct working length in the field. And, this is almost always just a plus or minus .030” tolerance.

   • You may have too much system pressure acting on the seal face.

     This can happen if you are using an unbalanced seal rather than a balanced seal. Balance refers to hydraulic balance not mechanical balance.

   • Process fluid has no lubricating properties

     Hot water, lignin and other liquids have little or no beneficial lubricating properties, which can cause slip-sticking.

   • The face materials themselves could be the problem.

     Tungsten-carbide running against tungsten-carbide is not always the best choice for abrasive service.

   • Cavitation

     If the pump is cavitating, the faces are literally blown apart then they come back together fast when the pressure drops. This jams the seal faces together.

   • Vertical pump configuration.

     If a pump is vertical, any entrained air will eventually makes its way to the highest point inside the pump. This is often where the seal faces are. If this happens, the faces run dry. Venting this area or having circulation will prevent this.

7. Inspect the springs.

   Broken springs are almost always a result of stress corrosion. This happens when stainless steel is flexed in the presence of halogens such as fluorine, bromine, iodine and others. Stress corrosion is actually a chemical attack where small cracks or fissures are created in the stainless steel. Particles can enter this crack and when the springs tries to compress a strong mechanical force is created that can break a spring. Most high-quality industrial mechanical seals have been converted to Hastelloy C springs. Hastelloy C can withstand halogens without cracking. It is interesting to know that springs and metal bellows can actually flex almost indefinitely without breaking if they are flexing within the limits of travel that they were designed for.

   Clogged springs indicate that solids in the process flow have accumulated. Seals with multiple small springs are more prone to clog than seals with one large spring or metal bellows.

8. Inspect the O-rings.

   O-ring materials can include various compounds of Buna-N (Nitrile), fluoroelastomers such as Viton, EPR/EP/EPDM, and perfluoroelastomers such as Kalrez or Chemrez.

   • If an O-Ring has dissolved or is swollen or sticky, it is evidence of chemical incompatibility.

     This chemical incompatibility could be from the process fluid itself. Keep in mind that chemicals become approximately twice as aggressive for every 20 degree rise in temperature. The stuffing box area of a pump can be several hundred degrees F higher than the temperature of the discharge. Additionally, the problem can be a flush fluid or even what a pump is cleaned with. For applications where there is no available elastomer that can handle the chemicals the sealing device will see, you can select a mechanical seal with either PTFE or flexible graphite secondary seals. Neither PTFE nor flexible graphite is a true elastomer. They are considered “plastic” in the sense that when deformed they do not recover their original shape.

   • If an O-Ring is hard, it is evidence of heat.

     Keep in mind that temperature limits published for various O-Ring materials are based on hot air only. These limits are for ultimate failure. Lower temperatures can start the degradation of an elastomer. Again, the temperature in a stuffing box can be hundreds of degrees F higher than at the discharge. Only a temperature gauge connected to the stuffing box, preferably with a tell-tale to indicate the highest temperature reached, can verify how hot the stuffing box gets. The solution for heat is either to reduce the temperature the mechanical seal and its O-Rings actually experience or to select an elastomer with suitable temperature resistance.

   • If an O-Ring is deformed it is also a signal that it was exposed to temperatures higher than it could handle.

     Measure the O-Rings from the inside to the outside and from the top to the bottom. This determines the actual deformation. Some deformation is normal. However, if the O-Ring is deformed to where your measurements are at or lower than the nominal dimensions, the O-Ring can no longer seal. Following is some specific dimensional information:

     • -000 series O-rings, from -004 through -050, have an actual cord diameter of .070”. They are designed to compress to about .0625”, so they are referred to as a nominal 1/16” O-Ring.
     • -100 series O-Rings, from -102 through -178, have an actual cord diameter of .103” and are referred to as 3/32” nominal.
     • -200 series O-Rings, from -201 through -284, have an actual cord diameter of .139” and are referred to as 1/8” nominal.
     • -300 series O-Rings, from -309 through -395, have an actual cord diameter of .210” and are referred to as 3/16” nominal.
     • -400 series O-Rings, from -400 through -475, have an actual cord diameter of .275” and are referred to as ¼” nominal.

     Note: Metric O-rings are sized by their actual cord diameter and their actual inside diameter.

]]> Thu, 07 Apr 2016 19:48:01 +0000 http://onhandsupply.net/blog/know-your-pump/


One of the benefits of owning an ANSI B73.1 process pump is the ability to buy replacement pumps or parts from your choice of vendor. But just as you would need to know the make, model, and year of your car to buy replacement parts, you also need to know the brand, model, and size of your pump to ensure that the parts you are buying will fit your existing pump.

see also:
• Understanding the information on your serial plate

Brand: Parts May Not Interchange Between Brands

Although the pumping unit of an ANSI pump are interchangeable between brands according to the dimensional standards of ANSI B73.1, the individual parts that make up the pump may not necessarily interchange between brands. Do your homework before buying a replacement part and make sure the replacement parts will interchange 100% with the pump you have. Parts from OnHandSupply interchanges with the Goulds 3196 style of ANSI pumps, but does not interchange with Flowerserve Durco Mark III style.

Model: Replacement Parts are Model Specific

The model number identifies whether or not you have an ANSI pump. For example, Goulds makes many different types of pumps besides the ANSI pump, but only their model 3196 is the ANSI B73.1 line. Pay close attention to any letters or suffix in the model number, as it may indicate a specialized model whose parts may not interchange with other brands.

Some parts may interchange between models, most commonly the power end. Although Summit's 2196 is their ANSI pump line and their 2196R line is not, the 2196R model uses the same power end. The same goes for Goulds, who use the same power end for seven different pump models

Size: Some Parts are Size Speicifc

Pumps come in different sizes, and you need to know the size of your pump to order the right part. The pump size looks like this:



Your pump size tells you whether your pump is in the STX series, MTX / LTX series, or XLTX series (think Small, Medium, Large, and Extra Large). Each size in the ANSI line has a unique impeller and casing. The rear cover (or stuffing box) and the adapter is shared among sizes with the same max impeller diameter within a series. The pump size also determines what size foot to use.

Once you have the brand, model number, and size, you're almost set. Here are some additional details to consider:

• Is your pump MTX or LTX? - MTX and LTX (Medium / Large) series share the same sizes, so things can get confusing. Although they share the same size, they use different power ends and rear covers. You can distinguish which series your pump is by measuring the shaft diameter of your existing pump:

  	STX	MTX	LTX	XLTX
  Shaft Diameter	1.375"	1.75"	2.125"	2.50"

• Is your rear cover (stuffing box) a "standard bore" or "big bore"? - You need to know this if you are replacing your rear cover or if you are replacing your sealing mechanism. Mechanical seals are different for standard bore and big bore configurations. Packing cannot be used in big bore rear covers. Measure the inside of your seal chamber and compare with the table below.

  	STX	MTX	LTX	XLTX
  Standard Bore	2.00"	2.50"	2.875"	3.375"
  Big Bore	2.875"	3.50"	3.875"	4.75"

• Check if your impeller was trimmed down - When a pump size is recommended for an application, an impeller may be trimmed down from its original size to optimize pump efficiency. If the serial plate does not have the trimmed diameter size, measure from the eye of the impeller to the furthest point on one of the impeller vanes to determine the radius, then double it for the diameter.

]]> Wed, 23 Mar 2016 17:07:00 +0000 http://onhandsupply.net/blog/serial-plate/

Here's an example of a pump serial plate from OnHandSupply. Serial plates may differ by each manufacturer, but the fields are generally similar. There may be an additional serial plate on the motor, which is specific to the motor used to power the pump. The serial plate for the pump may be on the power end (bearing housing) or the casing (volute).

It's important to familiarize yourself with the fields on a serial plate, because your serial plate is the easiest way to gain key information about your pump. We'll cover the definition of each field on the serial plate.

COMPANY

The company name on the serial plate may be the pump OEM (Original Equipment Manufacturer) brand name, the name of the pump packager, or the name of a private label company. This is the only entity that can interpret the serial number assigned to the pump.

MODEL #

A model number identifies the type of pump. This might be a generic term like "ANSI," or a model number specific to a brand. A model number that is brand specific, like a Goulds "3196," indicates that the pump is from the Goulds' ANSI line. A brand and model specific number is far more helpful, because it determines whether or not parts interchange with other brands. Besides the Goulds 3196, other ANSI brand models include: Summit 2196, Peerless 8196, Griswold 811, Aurora 3550, and others

STD

ANSI designation. This identifies the dimensions of the pump according to ANSI standards. The standard number is not unique to each ANSI pump size, but all pump sizes with the same ANSI designation will have the same pump dimensions (pump can be replaced without changing the base or piping).

GPM

This is the flowrate of the process fluid being pumped, in gallons per minute.

IMPLR DIA

Impeller Diameter is the diameter of the impeller that the pump was designed to operate to achieve best efficiency for a given application. Note that this may be less than the maximum impeller diameter which corresponds to its size. If the impeller does not need to be trimmed down, the impeller diameter may say "Full" to indicate that no trimming is necessary.

SERIAL #

This is a unique identifier for the pump that only the manufacturer or packager can interpret. From the serial number, every part of the pump should be identifiable for reorder; however, a pump can undergo modifications over its life which the original manufacturer may not have on file. We recommend double checking key measurements to verify the information from pump serial numbers.

SIZE

The size of the pump contains significant information. It contains three numbers, typically written like '2x3-13'. The first two numbers are the suction and discharge opening dimensions. The order is not important, although each brand has a preference. The smaller number is the discharge, and the bigger number is the suction.
The last number is the maximum impeller diameter. Note that the actual impeller diameter may be deliberately trimmed to fit the pump application, or unintentionally less than the max diameter due to wear.

MAT'L CONSTR

Materials of construction. This is typically the material of the “wet end” parts, or the parts that come in contact with the process fluid (impeller, rear cover/stuffing box and casing/volute. These parts may need to be more corrosion resistant than other parts of the pump that are not wetted by the process fluid.
Common materials include cast or carbon steel (CS), cast iron (CI), 316 stainless steel (316 SS), CF8M (the cast equivalent of machined 316 stainless steel) and CD4MCuN. There are more exotic alloys available such as Hastelloy C, Hastelloy B, Alloy 20 or titanium for extreme applications.

FT HD

Feet of Head. This is the height of the column of the process fluid that the pump can maintain.
Feet of Head can be converted to psi (pounds per square inch) = 0.433 * specific gravity of the process fluid. In most cases, feet of head is approximately twice the psi.

MAX DIA

Maximum diameter of the impeller. Often not necessary because this information is implied by the pump size.

RPM

Revolutions per minute. ANSI pumps commonly turn at 1800 or 3600 rpm.

]]> Mon, 01 Feb 2016 20:52:00 +0000 http://onhandsupply.net/blog/inpros/

While there is much discussion about which bearing protector works the best, it is safe to say that the worst bearing protector will protect bearings significantly longer than the best oil seal.

The ASLE (American Society of Lubrication Engineers) have determined that the equivalent of one drop of water inside a bearing housing can reduce bearing life 48%. What happens is that the drop of water will eventually get under the bearing ball and cannot support the load that oil or grease can support. As a result, the metal bearing race gets deformed outside its plastic limit where it can return to its original dimension. From then on, every time that part of the race gets a load on it, it deforms in its plastic range until it fails.

The use of oil seals in ANSI pumps is rapidly diminishing in favor of bearing protectors. Bearings can have a legitimate design life of several decades. But if you take the time to study the information provided by the oil seal industry, you’ll find that the legitimate design life for oil seals in typical applications is measured in days up to a few months.

People have tried both multiple oil seals and reversing the direction of the curl of the oil seal lip. The issue with multiple lips is that the outer lips run dry and have a very short life. Plants have tried reversing the oil seal installation. A single lip facing towards the bearing does a little better keeping the lubrication in but not as good a job in keeping contamination out. Reversing the direction of the lip creates the opposite outcome, better at keeping the contamination out but a poorer job of keeping the lubrication in. And, the lip itself can imbed with solid particles and score the shaft under the lip actually opening up the clearance for contamination to enter and lubrication to escape.

Bearing protectors can be made of bronze, stainless steel, PTFE or some other plastics or even a combination of two different materials. Bearing protectors can be reused as they are a non-contact sealing device. Bearing protectors consist of two pieces that are usually unitized. This means that the two pieces do not come apart. The two pieces can be described as a rotor and a stator. The rotor acts as a slinger. Water and bearings do not mix. Whether the source of unwanted liquid is a leaky mechanical seal or packing, water from a cleanup hose, or wet weather if the pump is outside….

The other component of a bearing protector is the stator. This is usually a labyrinth. Most ANSI pumps use the shaft to rotate the oil slinger inside the bearing housing to sling the oil around to lubricate the bearings. It splashes everywhere by design. The labyrinth catches the slung oil in a series of thin lips, hence the term labyrinth. Gravity allows the oil to slide down to the bottom of the labyrinth where there is either an angled trough or one or more drain holes that allow the captured oil to return to the sump area of the bearing housing.

One less understood benefit of bearing protectors is that should a catastrophic bearing failure take place, the bearing protector can act as a sacrificial disaster bushing and can protect the running gear from being destroyed. A few hundred dollars for a new bearing protector is better than a several thousand dollars for new running gear.

A good practice is the use of root cause failure analysis for bearings as well as for mechanical seals and packing. The use of RCFA coupled with a strong managed lubrication program yields benefits that are a multiple of the cost of the program.

Other Resources:

Labyrinth Seals and Lip Seals: An Economic Comparison

Understanding Bearing Housing Protection and Reliable Lubricant Application

]]> Wed, 20 Jan 2016 15:57:37 +0000 http://onhandsupply.net/blog/history-ansi-pump/ Before the ANSI Pump

The history of the ANSI pump is unique and interesting. Before ANSI pumps, the North American marketplace was dominated by Ingersoll Rand and Worthington, the latter is credited with the development of the first modern pump. In fact, pump serial #2 is in the Smithsonian museum for its historical significance.

Before ANSI pumps, each and every pump had to be uniquely plumbed. Install a particular brand of pump and only that brand of pump would flange up to the existing piping. Since changing piping is time consuming and expensive, there was no incentive or pressure to keep the cost of replacement parts reasonable. Replacement part prices and delivery times were outrageous and even the most basic, courteous service was virtually non-existent. I worked for one of those majors and spent most of my time trying to address these issues with customers and distributors who were frustrated beyond description. After several years, I started my own industrial distributorship that competed with them with alternative products. These are the so-called “pirate” parts also called alternative sourced or non-OEM (original equipment manufacturer) parts. More on this later.

The Rise of the ANSI Standard

When the concerns and dissatisfaction reached an unsustainable level, a small number of customers who represented a large percent of the pumps purchased every year approached the pump companies. They wanted the ability to swap brands without changing the piping. Naturally, the majors didn’t want to end their near monopoly and the pump companies resisted because of the design and manufacturing expenses involved in developing an entire new pump system. However, after one company broke ranks the others then had to follow. This was in the 1950’s. The early effort was called the American Voluntary Standards. Not too voluntary, though. Later, this was folded into ANSI standards in 1974 and updated as recently as 2012.

The objective was to have pumps in a variety of sizes that could “flange up” with other brands of the same size. This means that the suction and discharge flange locations, the location of the feet and the location of the shaft for the driver had to be the same. These pumps were designed for the chemical process industry and, at the time, were considered to be throwaway pumps. In other words, when the pumps started to deteriorate, you just replaced it.

Emergence of the ANSI Pump Market

So, the ANSI market developed. A few companies grew to dominate the market. Then, the cycle of expensive parts and slow delivery repeated. If quality, delivery, prices and service were what customers wanted, the door would have been closed to others. However, there was enough demand to launch the alternative aftermarket. First came the easier pieces such as shafts and sleeves. Some improvements were offered by these non-OEM manufacturers. As an example, sleeves could be hard-coated to last longer with packing. Then along came some of the high usage more sophisticated parts like casings and impellers.

Finally, some of these manufacturers realized that just by making a few more parts such as bearing housings and adapters, they could offer the entire pump. Some of these manufacturers set up their own distribution networks and some offer their products to the entire market directly. Today there are a number of manufacturers and resellers who offer pumps and even parts that interchange one-for-one. A result of this and the competition that it brought was to bring replacement parts down in price. A pump could be in service without one original part left in it. You can literally build a pump from parts without paying a premium compared to buying a complete pump to start with. This is unprecedented in the pump industry and, perhaps, in almost any industry.

]]> Fri, 06 Nov 2015 21:59:34 +0000 http://onhandsupply.net/blog/chemshow/

We are excited to exhibit at the 2015 Chem Show, from November 17-19 at the Javits Center in New York City. It is the 100 year anniversary of the Chem Show, where the chemical processing industry gathers together to meet face-to-face and showcase the innovations and solutions to engineers, plant managers, and other CPI professionals.

The chemical processing industry is also looking for ways to boost capactiy, reduce costs, and optimize process operations. While the ANSI pump may not be an innovative technology, we believe OnHandSupply's internet retailing model brings a modern perspective to how the industry can improve its supply chain and purchasing. With pump parts distributed throughout the US, we make it easy for the process plants who can now buy from one entity and reduce inventory holding costs, while still having access to pump parts within 1-2 business days.

If you are planning to attend the 2015 Chem Show in New York City, please stop by our booth to learn more about how OnHandSupply can improve your operations. We will be in booth 361, and we hope to see you there!

]]> Tue, 13 Oct 2015 19:05:20 +0000