What Role Does Vapour Pressure May Have On NPSHA?

, ,

Previously we have discussed net positive suction pressure but avoided the role vapour pressure may have on NPSHA. Today, we will shed some light on this subject.

The study of vapour pressure starts with developing an understanding of evaporation. Evaporation of a liquid is a concept most of us generally understand, but do we really know the intricacies of the process?

Let’s start by examining a beaker of material in a liquid state. If that material is water, the beaker at a molecular level would be full of H2O water molecules. If the water was cold, say 1 degree C, then these molecules would be moving slowly and would not possess much kinetic energy. The molecular attraction between the molecules would keep almost all the surface molecules contained.

Periodically, due to random collisions of molecules, one surface molecule may accumulate enough kinetic energy to break the molecular attraction and leave the water’s surface. This molecule is now in a gaseous state and has “evaporated”.

As the temperature of a liquid increases so does the kinetic energy of the individual molecules.  As the drawing below illustrates, the increase in energy within the molecules at the elevated temperature translates into more molecules obtaining enough energy to transition into a gaseous state.

If the beaker is open to the atmosphere, air currents carry these water vapour molecules away, and the level in the beaker slowly drops. However, if the beaker is sealed as in the illustration above, the water vapour is contained and pressure starts to build.

While evaporation is occurring within the sealed beaker, a few gaseous molecules lose kinetic energy due to collisions with other molecules and return to the liquid state. A process we know as condensation.

If the liquid within the sealed container is held at a stable temperature for a long enough period of time, an equilibrium is reached. That is to say, water vapour is being formed by the process of evaporation and at the same rate vapour is condensing back to a liquid state. The pressure at which a state of equilibrium is reached at any given temperature is known as the liquid’s vapour pressure at that temperature.

Some people crudely describe the vapour pressure as the push the liquid has to jump from a liquid to a gaseous state.  When the liquid is in a container that is open to the atmosphere, the force exerted by the vapour pressure is pushing against atmospheric pressure.

When water is heated to 100 degrees C (212 F) it has a vapour pressure of approx 14.7 psi.  The pressure exerted by the water, in an effort to jump to a gaseous state, is now equal to the downward pressure on the water exerted by atmospheric pressure at sea level.  Atmospheric pressure at this point is incapable of stopping the rapid formation of water vapour throughout the entire volume of the liquid, and the water boils.

As discussed in earlier blogs, pumps often rely on atmospheric pressure to push liquid into the eye of the impeller. If the vapour pressure is pushing up against the downward force of atmospheric pressure, then it is detracting from the pressure available to push liquid into the pump.  In terms of earlier blogs, it is reducing the net positive suction head available. (NPSHA)

Looking at the formula used to calculate the net positive suction head available,

NPSHA = Ha +/- Hs – Hf + Hv – Hvp

we see vapour pressure is shown as negative or subtractive in nature. This coincides with today’s discussions of how vapour pressure acts against atmospheric pressure and detracts from the pressure available to feed liquid to a pump.

Fortunately, the majority of liquids pumped are water-based and not particularly hot but if you have a more volatile liquid or is at an elevated temperature best consider its vapour pressure if you have low NPSHA or you may be in for pumping problems.

If you need help or advice, the Hevvy/Toyo’s team of application engineers are well versed on this subject and are always willing to help.

Cheers

RJ

Understanding how pump curves are generated

, ,

Industrial pump manufacturers always have available pump curves for fixed-speed pumps with full-size impellers, but what if you want to run the pump at a different RPM or use a trimmed impeller?  Well, the manufacturer should be able to provide you with a modified curve but if that is not easily obtainable just use affinity laws to modify the manufacturer’s standard curve.

Affinity laws are not courtroom jargon, just some simple formulas that can be applied to operating points on a pump curve to predict a new point when pump RPM or impeller diameter is modified.

Changing the Pump RPM: When the impeller diameter of a centrifugal pump is held constant the effect of changing the speed (RPM) is in accordance with the following formulas, where  N= RPM   Q = Capacity   H = Head, and  BHP = brake horsepower.

Q1/Q2 = N1/N2

H1/H2 = (N1/N2)2

 BHP1/BHP2 = (N1/N2)3

Changing the Impeller Diameter: The effect of trimming the impeller without changing the pump speed is virtually identical to what happens when you alter only the pump speed. Therefore the formulas are also very similar, as illustrated below, where  D= impeller diameter   Q = Capacity   H = Head and  BHP = brake horsepower.

Capacity: Q1/Q2 = D1/D2

Head: H1/H2 = (D1/D2)2

BHP: BHP1/BHP2 = (D1/D2)3

Note; Before  I show you how to use these formulas to generate a special curve there are a couple of cautionary notes.  Firstly, the accuracy of affinity laws or formulas diminishes as the percent change increases.  Generally changes of approximately 15 % or less still yield acceptable accuracy.  Secondly, when trimming an impeller you are modifying the effective length of the impeller vanes and that is what results in a modified performance.  Since the impeller vane on most pumps does not start at the center of the impeller the percent change in impeller diameter does not accurately reflect the percent change in vane length. Since impeller diameter is easier to work with than vane length, affinity laws substitute it. On high flow- low head impellers this substitution can introduce significant error and actual test tank data should be used to create modified curves instead of affinity laws.

Creating a curve for a special pump speed or impeller trim is basically the same procedure. I will therefore only demonstrate one today and that is the creation of a  pump “Head–Flow” curve.

Below is a pump curve for a 1750 rpm pump.  If we needed to create a curve for 1650 rpm we would start by calculating the shut-off point. The current shut-off point is zero flow at 125 meters as indicated by the green star. The new shut-off point at the reduced rpm would be calculated using the following two calculations

The new flow point at shut-off is calculated using the flow affinity formula  N1/N2=Q1/Q2

OR

1750/1650 = 0/Q2     therefore  Q2 is zero M3/hr

The new head point is calculated using the head affinity formula  (N1/N2)2=H1/H2

OR

 (1750/1650)2 = 125/H2    therefore H2 is 111 meters

Plotting zero M3/hr at 111 meters on the curve below, we establish the calculated shut-off point for the pump at an rpm of 1650.  (indicated by the blue star below)

pump curve

We would next calculate the new run-out point. The current run-out point is 17.2 M3/hr at a head of 85 meters, as indicated by the red star. The run-out point at the reduced rpm would be calculated using the same two formulas as used with the shut-off point

N1/N2=Q1/Q2

OR

1750/1650 = 17.2/Q2     therefore  Q2 is 16.2 M3/Hr

The new head point is calculated using the flow affinity formula  (N1/N2)2=H1/H2

OR

 (1750/1650)2 = 85/H2    therefore H2 is 75.6 meters

Plotting 16.2 M3/hr at 75.6 meters on the curve above, we establish the calculated run-out point for the pump at an rpm of 1650.  (indicated by the orange star above)

Additional points can be calculated to fill in the points between shut off and run out using the same procedure as used above, thereby filling in some points on the 1650 rpm curve as shown in the illustration below. (black stars)

pump curve

Finally, as shown below, connect the points, and you have a 1650 pump curve.

pump curve

In closing, this is how you can create a curve by yourself, however, if it is a Toyo pump curve, you may forget this blog altogether and just call 604-298-1213 and have our application team e-mail you what you need!

Cheers

RJ

What’s the Difference Between Water Pressure & Water Head

, ,

In a previous blog, I quoted the number 2.31 when discussing the relationship between water pressure  & water head.  Since then I have had a request to elaborate on the number 2.31 and where it comes from.  Frankly, with many gauges still calibrated in PSI  I still relate to the US system of measure. The quoted factor therefore only applies to pressure in units of PSI and water head expressed in feet. The basis of the factor 2.31 is as follows:

  • One cubic ft of water weighs 62.4 lbs and contains 1728 cubic inches. Therefore one cubic inch weighs 62.4/1728 or 0.0361 lbs.
  • A column of water with a cross-section of one square inch, that is 12 inches tall (one foot), would weigh 0.0361 x 12 or 0.433 lbs. Exerting a pressure of .433 lbs per square inch.
  • Based on this, a pressure of 1 lb per square inch would require a one square inch column of water equal to 1/.433 or 2.31 ft tall

PS —  In the metric system the conversion factor is 1 Kilopascal [kPa] = 0.101 974  Meters of water column [mH2O]  My apologies to my fellow Canadians for not providing the metric Stay tuned, and bye for now.

RJ

Let’s Explain the Confusing Term: Bulk Density Ratio

, ,

A couple of months ago I used the term “Bulk Density Ratio” and said that it was a subject for another day. So today is the day.

Pea gravel

The term bulk density ratio describes the relationship of a material or substance mass as found in a particular sample vs its theoretical mass, assuming “complete compactness” and no voids of any type.

For example, Pea gravel is made up of small particles of stone, and a solid block of stone usually weighs approximately 168 lbs per cubic foot  ( 2.7 times the weight of water, an SG of 2.7).

When we measured a particular sample of pea gravel we found that it weighed  108 lbs per cubic foot (1.76 times the weight of water). But stone is stone!! Why the difference?

Clearly, the difference is the pea gravel sample is full of voids between the stones. The ratio of solid stone to the sample of pea gravel is 168 lbs / 108 lbs or 1.55

Said differently,  pea gravel occupies 1.55 times more space than solid rock for the same mass of product. This ratio is referred to as the “Bulk Density Ratio”.

If we take the weight of the pea gravel sample and put it over the weight of a sample of the same volume of solid stone and then convert it to a percentage (108/168  X 100 =  64%) we find that stone occupies 64 % of the space. Assuming it is a dry sample, air must occupy the balance of 36% of the sample’s volume.

Great information but how is it useful in the pump world? Well, the answer to that question is that Bulk Density Ratio is the key to calculating production rates when moving products like sand or gravel that are often measured by volume.  The easiest way to explain this is by using an example.

You are a contractor and your job is to remove a sand bar made up of coarse sand that is 100 meters wide by 100 meters long by 5 meters thick, or 50,000 m3 of material.  If you can pump at a rate of 800 m3/hr of slurry with a density of 1.23,  how many hours must you operate?

You can assume the coarse sand has a dry Sg of 2.7. But the pumpage is not all sand, it is a slurry with water occupying the space between the solids.  The formula below can help you determine the percent solids, by volume.

Cv =( Sm-Sl)/(Ss- Sl) = (1.23 – 1.0)/(2.7 – 1.0) = .135 or 13.5 %

  • Cw = Percent solids by weight
  • Cv = Percent solids by Volume
  • Ss = Sg of the dry solid
  • Sm = Sg of the slurry
  • Sl = Sg of the liquid

At a slurry flow rate of 800 m3/hr this equates to .135 x 800 m3 or 108 m3 of “solid” rock per hour, but the end product is not solid!!  This is where you need to know a Bulk Density Ratio to properly estimate production by volume.

Applying the bulk density ratio of 1.55 for coarse sand would mean the contactor is removing the sand bar at a rate of 108 x 1.55 or, 168 m3/hr. Based on this rate of production, the project of moving 50,000 cubic meters being pumped at a rate of  168 cubic meters per hour will take 297hrs to complete.

Hopefully, this short blog helps to clear up some of the confusion around this subject, but if you still have some questions or need help on a specific project please feel free to contact our very competent applications team.

Bye for now

RJ

Understanding the Effects of Specific Gravity on Centrifugal Pumps’ Head and Pressure

, ,

It is very important when working with centrifugal pumps that there is a clear understanding of how specific gravity, commonly abbreviated to just  SG, affects or doesn’t affect pump output in terms of head and or pressure.

Let’s first define SG. The dictionary definition is simply “ the ratio of the density of a substance to that of water”.   As shown below one cubic meter of water weighs 1000 kg, while an equal volume of granite weighs 2700 Kg. ,  2.7 times as much. Hence granite is said to have an SG of 2.7.

A word of caution! Whatever the sample size, there must be no voids. Materials such as sand or gravel must be adjusted by a factor to allow for the air space between the particles. This factor is referred to as the “bulk ratio”.  A subject for another day!

Turning attention to the head. It is sometimes defined as “the vertical distance (in feet or meters) from the elevation of the energy grade line on the suction side of the pump to the elevation of the energy grade line on the discharge side of the pump.”

When applied to a centrifugal pump’s output, it is more simply described as “the height a pump can lift a liquid to”.  For example, if a centrifugal pump tries to pump straight up a 300 ft pipe, but flow stops as the water in the pipe reaches 231 ft, the pump is said to have a shut-off head of 231 ft.

You may ask, why not express the output of a centrifugal pump in PSI? Well, the answer to that question is based on the operating physics of a centrifugal pump.

Centrifugal pumps use centrifugal force generated as a liquid is spun down the vanes of the impeller to create pressure within the pump’s casing/volute. The heavier the liquid, the more force or pressure is created.  Some pump manufacturers rate performance at a stated output pressure based on water, usually in terms of PSI, but the majority of pump manufacturers use the term head, as it does not vary with liquid SG.

The basis of the statement in bold above is illustrated to the right.  The pumps are identical, but the blue column represents water being lifted while the orange column represents a liquid with an SG of 1.5 being lifted.  Notice the output pressures vary in proportion to the SG while the rating for lift remains constant.

The centrifugal pump handling a liquid that has an SG of 1.5 is accelerating a liquid weighing 1.5 times that of water down its impeller vanes, creating a force (pressure) that is 1.5 times that of an identical pump pumping water.  The column of liquid, however, also weighs 1.5 times more than water. Hence it needs a pressure equal to the pressure on water times 1.5 to lift the heavier liquid to the same height.

The obvious question then becomes, if pressure gauges measure in PSI, how do we relate that to a pump’s output in ft or meters of the head?  The full explanation of that is again a subject for another day, but the equation below and the diagram to the right will get you by for now.

 

 

 

I hope today’s blog helps explain the relationship between SG, Head, and Pressure. As mentioned in today’s blog, I will in future blogs address the “Bulk Ratio” and fully explain the 2.31 ratio of head to PSI.

Stay tuned, and bye for now.

RJ

Differences Between Cutter Fan Agitators and Shaft-Connected Inducers

, ,

I  was recently in a discussion with an engineer regarding the use of a cutter fan to supply agitation. I was very surprised to find that the engineer, although having some pump experience did not understand the difference between the title objects. Not only could this lead to backward rotation and subsequent damage to a pump but the lack of understanding would also lead to the implementation of ineffective dredging techniques. Therefore today’s blog is on Cutter Fan agitators and shaft-connected Inducers.

The Cutter Fan also known as an Agitator is connected to the main shaft of the pump. The first manufacturer to employ this type of agitation was Hevvy/Toyo Pumps on their heavy-duty submersible dredge pumps. Their patented design utilized a curved three-blade stirring attachment that was threaded onto the pump shaft just below the suction inlet.

The cutter fan or “agitator” as it is sometimes called is typically protected by a stand attached to the bottom of the pump. For added protection on the larger pumps, Toyo places a stub shaft in between the pump main shaft and the cutter fan. Operationally, all cutter fans redirect a portion of the fluid heading toward the pump suction and push or “fan” fluid away from the pump. The agitation provided by the cutter fan dislodges solids and re-suspends them into a slurry. As these solids are drawn toward the pump inlet some of the slurries is redirected by the cutter fan back down into the solids deposit providing a more effective form of agitation than a jet of purified water. This redirecting of slurry/solids continues in a cyclic fashion forming a “pocket” of high solids content slurry directly in front of the pump suction inlet. This of course maximizes the number of solids being pumped, an important feature for any dredge pump.

The Inducer, like the cutter fan, is attached to the main shaft. It can be located anywhere in front of the impeller. The optimum location is entirely application-dependent. Technically said, its normal function is to raise the inlet head by an amount sufficient to provide the NPSHR, thereby preventing significant cavitation in the pump. In short, it can help in applications where initial priming is difficult or the fluid just refuses to flow well into the pump.  Below is a picture of a style of inducer.

In Summary, cutter fans push product away from the pump to aid in agitation while inducers help draw product into the pump. If your new pump arrives with an item that looks like an inducer or maybe a cutter fan, do not guess as to which it is and use it to confirm correct rotation. Since they are both normally attached to the shaft by some form of thread, reverse rotation may result in components unscrewing during operation and some very expensive repairs. Pumps always have a rotational arrow. The cutter fan,  and sometimes the inducer, can be excellent items to watch when bumping for rotation as they are easily visible, but when wiring the pump observe the marked arrow to obtain correct rotation!!  Once you have set rotation you can now look at the blade/vane angles and it will be easy to determine whether the mystery item is a cutter fan or an inducer.

Bye for now!

RJ

There Is No Such A Thing As A Bad Pump Only A Misapplied Pump!

, ,

No manufacturer that has been around for any length of time makes a bad pump. The bad press that follows an early pump failure is generally the result of a misapplied pump. Unfortunately, it is often the pump’s reputation that takes the rap and not the circumstances that lead to the installation of the “wrong pump”. Any reputable supplier will want the client to be pleased with his pump purchase long after the initial installation. Having said that how does he get it right?

In 1941 the actor and comedian  George Jessel, when referring to “the people”, made the famous quote “Give ’em what they want.” The truth of the matter is in the pump world you need to “give them what they need”. To do this the supplier needs to know as much as possible about the customer’s application.

The need to know

I once had a PA call me up and request a quote on a 40 HP submersible pump. When I asked for more info such as head, flow, liquid, and solids content he cut me off and said I don’t have time for this just give me a price for a 40 HP pump. As I was in no position to argue with him I sent him a quote for 6 different pumps. Needless to say, I got a call from the mine foreman the next day and we discussed the details of the application.  As it turned out the head/flow he needed required a 60 hp pump so I sent him a quote, copied in the PA, and soon after received an order.  Everyone was happy.

If the mine had taken delivery of a 40 hp pump, at best output/flow would have been below the required amount or possibly even non-existent. Would it have been a bad pump or a misapplied pump?

What to Know?

The simple answer here to is give as much information as possible. The more the supplier knows about the application the better he can respond with appropriate information. It is equally as important for the customer to offer up as much information as he can as it is for the supplier to seek all the application details.

Major pump suppliers like Toyo all have questionnaires that can be sent out to prospective customers to help them convey application details to the manufacturer. These are great tools that help both parties home in on the correct pump for the application. Even if the pump user is well versed in pumps and the application thereof, I would still recommend the completion of one of these questionnaires.

Providing specs like; head, flow, liquid type, etc are covered by all questionnaires but more general questions are sometimes overlooked or just not responded to. These can be critical to the selection process. Below I have listed a few:

  • Why is the customer looking to pump the liquids or materials? It is vitally important to obtain a project overview. This gives the supplier a feel for the customer’s end goal and keeps everyone on track while not getting locked into a fixed project plan. Thinking outside the box can sometimes lead to a better plan.
  • What is the anticipated life of the project? This will help the supplier determine if the customer needs premium products that will last a long time, or if he needs a more budget friendly product that will still last for the life of a shorter project.
  • Is the purpose of the application to relocate a liquid, a liquid that contains solids, or is it to use a liquid to relocate a solid. This along with the project life data, will help a supplier focus on the correct product line to offer.
  • Is this a new application or are the existing pumps giving problems? If it is replacing an existing pump, what type of pump is it and how is it failing?  No one wants to offer similar equipment to what is already there and suffer the same fate.

Who to know

I can not over-emphasize the need to obtain and share information. It is the best defense against failed pumps.  I know that all customers vary with regard to pump knowledge, some will need to seek help from within their organization to provide the details of their application.

However, you can rest assured that the sales and application team at  Hevvy/Toyo are here to support you and are just a phone call away.

Bye for now.

RJ

6 Parameters Maximizing A Mechanical Seal’s Performance And Longevity

, ,

Pumps fail for a host of reasons, but as the chart below depicts, more than two-thirds of the failures relate to sealing device issues. With the movement to conserve water and minimize the quantity of waste liquids requiring treatment, the industry is relying more and more on mechanical seals. As the installed population grows so does the percentage of pump breakdowns resulting directly from mechanical seals.

I, like many pump guys, try and separate pump problems from mechanical seal problems. However it’s a package, and the pump supplier has to support the customer in any way that he can. With that in mind, I thought I should delve into some of the general pump requirements for a successful mechanical seal installation.

There are six distinct parameters that must be correct for a seal to provide maximum performance. Some seal designs are more able to accept dimensional inconsistencies than other designs, so specific tolerances must be obtained from the seal supplier prior to installation.

Just to be clear, the tolerances referenced in the diagrams below are for general reference only. You must confirm the tolerances required for specific seals with the manufacturer of that seal.

1)  Shaft run-out

Shaft run-out should typically not exceed 0,05 mm (.002”) TIR (Total Indicator Reading) at any point along the shaft for ball or roller type bearings. (1000 to 3600 RPM)

 

2)  Radial shaft movement

Radial shaft movement is generally limited to 0,05 – 0,10 mm (.002” -.004”)  for ball or roller type bearings. For sleeve or journal type bearings, values will generally be in the order of 0,10 – 0,15 mm (.004” – .006”)

 

3)  Concentricity, shaft to bore

Concentricity of the shaft to the seal chamber bore should normally be within 0,025 mm per 25 mm shaft diameter (.001″ per 1″ shaft diameter) to a maximum of 0,125 mm (.005″) TIR.

4)  Seal chamber squareness

Seal chamber squareness to the shaft centreline should be approximately 0,015 mm per 25 mm seal chamber bore (.0005″ per 1″ seal chamber bore). Note: make sure that shaft endplay does not affect the reading.

5)  Shaft endplay

Shaft endplay should generally not exceed 0,25 mm (.010”) TIR, regardless of thrust bearing type.

6)  Seal  preload

All seals have some type of spring mechanism to provide an initial compression of the seal faces, commonly referred to as a preload. Cartridge seals have a built-in preload that only requires the installer to remove set tabs prior to start-up.

The specification for the preload varies greatly dependent on the specific seal design. You must see the manufactures installation instructions for specifics.

I hope the information in this blog helps some readers confirm that their pumps are ready for a successful mechanical seal install.  Next blog I will try and touch on some operational tips on avoiding premature seal failure.

RJ

When Hevvy/Toyo’s “Biggest Sand Pump” Rescues A Billion Dollar Drill Rig

, ,

Over the holidays my son asked me if I had ever been on an oil drill rig. In enlightening him as to what his old man did in the 80s I realized the story might me interesting to others, so here goes.

It started with a telephone call from a purchasing engineer with the Gulf Canada Resources. He was interested in the “biggest sand pump” we could immediately supply to Gulf Oil. On describing the DP-150B , which was the largest pump in stock at that time. He replied good and could I put it on a plane today to Inuvik NWT, then get a helicopter lined up to fly it to a drill rig. I explained that the only plane we knew of that could carry a pump of that size was a very expensive C-130 Hercules and a helicopter was another issue. His answer was very short, “budget unlimited, we are in danger of losing a billion dollar drill rig so get on it.”

Andy, our brilliant logistics expert, got the C-130 figured out and started making calls for a chopper. The largest helicopter currently stationed in the Arctic was an older Sikorsky but it could only lift about 5500 lbs. We would have to fly it in pieces from Inuvik to the rig and assemble it on site. By nightfall the pump was in a C-130 and I was on a standard commercial flight with as many tools as I could carry, both headed to Inuvik .

The next morning I stripped the entire wet end off the pump but avoided tearing into the sealed motor section. The helicopter crew were also up early and had gutted the chopper to remove as much weight as possible. No one knew the exact weight of the main motor unit so the pilot said lets find out if we can lift it! The attempt failed, the load just hopped and the pilot had to abandon the lift. During the ensuing discussion I asked the pilot if it was true that helicopters gained lift with forward speed and if so we could lay the 500 lbs power cord out on the runway and drag it until lift was gained. He replied “good plan but as I flare out at the rig I will loose lift and I will be dropping that pump whether I am over the landing pad or not.” Fifteen minutes later I watched as a pump flew down the runway at an elevation of 10 ft with the electrical cord sliding along behind like a big wet noodle.

The chopper returned about 90 minutes later. Pilot had done his job well! The ground crew was waiting with the balance of the pump loaded into a cargo net and I had been dressed in a flight suit and helmet. Apparently the helicopter was not allowed to carry a passenger with a sling load underneath and I was to be smuggled on board as the navigator. No time for another trip or for that matter, any rules.

On arrival at the rig I was met by the rig captain. He told me if I was asked questions about helicopter rides, that I was to confirm that I had come in on a special Jet Ranger flight that AM. I was also going to get a 5 minute briefing and be asked to go to work immediately. The rig is on 24 hr notice to abandon the platform but all safety training and formal orientation would take place after the pump was up and running.

The quick briefing explained that this rig was named the Molikpaq and it was actually a mobile island. The island was basically a large hollow steel donut, or “caisson” , with a drill structure constructed over the center hole. When a site was selected a dredge would dress the sea bed forming an underwater berm and then the Molikpaq would be floated over that site and the hollow donut/caisson flooded. When the Molikpaq was firmly sitting on the sea floor a dredge would fill the center with sand. Small explosive charges would then be embedded in the sand and would compact or densify the sand, thus forming a stable island that could deflect the winter ice pack as it impacted the rig

When it was time to move the rig to another location four sand pumps would be lowered into the sand core to remove the sand, water pumps would then pump out the caisson and the rig could  be floated to a new site. However, if during the sand removal stage the sand pumps failed with half the sand core, the Molikpaq could not be refloated and it would be destroyed if the receding ice pack reversed course and impacted the rig.

With the pack ice visible in the distance and four broken competitors pumps no more need be said.  This was serious!!

Working with the midnight sun, a rig mechanic and I reassembled the DP-150B , installed it, and got it fully operational. With endless daylight we set about making one good pump out of the broken ones on site . After 36 hrs with no sleep we had two operational pumps and  I was rewarded with a private room and told to sleep as long as I wanted.

When I finally awoke I took my safety training and orientation. The captain asked me to stay on board as their guest for the day and when they were finally comfortable with the situation he arranged for the Bell Ranger to come and pick me up. He asked if there was anything he could do in appreciation of coming to their rescue.  I asked if the pilot could hesitate as we departed so I could get a picture for the scrap book, the Captain replied of course.  On departure they put a safety harness on me and strapped me to the seat.  At about 500 ft I found out why the harness! The copilot opened the side door, the pilot asked if I had my camera ready and then rolled the chopper on to its side.  I got a clear shot of the Molikpaq  but it was a little terrifying to put it mildly.

The flight back to Inuvik was uneventful except it seemed a bit strange  that we flew the whole way at about 50ft off the deck. I had always been told that helicopter pilots flew at altitudes of over 1000 ft so they could auto-rotate into a safe landing if they had an engine failure. On arrival the copilot walked me in as Frank the pilot shut down the chopper. Curiosity got the better of me, so I asked the copilot about the flight altitude. He said I was correct about the auto-rotation but a safe landing in the Arctic ocean just means that you will freeze to death before help can arrive, so its better to die in the crash. The main reason however is that Frank is an ex Vietnam gunship pilot and a bit different, he still believes he is vulnerable to surface to air missiles if he is flying at over 50ft.

RJ

Getting A Submersible Pump’s Rotation Right

, ,

I spent the better part of 25 years attending hundreds of submersible pumps’ start-ups, and the single most common error I had to correct prior to commissioning was the pump’s rotation.

The majority of Hevvy/Toyo submersible pumps sold feature an agitator.  With the agitator or cutter fan clearly visible when the pump is viewed from below, observing rotation during a bump start is pretty straightforward.  Unfortunately, as discussed in a previous blog, many people assume the object they are observing acts like an inducer and don’t check the nameplate for correct rotation.  A potentially costly mistake.

Correct rotation on this DP style pump is CCW when viewed from below

Sometimes I would arrive at a start-up and the pump was already installed and positioned in such a way that viewing the underside of the pump was not possible. When this was the case I would make sure the pump was freely suspended and do an in-place bump test. Newton’s Third Law would cause the pump to jump rotationally as power was applied.   (Law states “For every reaction, there is an equal and opposite reaction.” ) For electric submersible pumps the “initial reaction” is the spooling up or acceleration of the pumps rotating element.  The “equal and opposite reaction” is the force trying to spin the enclosure of the pump in the opposite direction.  This means if the main housing jumps in a clockwise direction the pump impeller is rotating counterclockwise, and visa versa if the housing jumps in a counterclockwise direction then the pump impeller is rotating clockwise.

Correct rotation should always be noted on or near the nameplate. Some manufacturers use an arrow to denote correct rotation others use terms like “ CW when viewed from the drive end” or CCW when viewed from the suction end. Many pump casing designs utilize a tangential discharge which can provide a good clue to the rotation if nameplate data is long gone.

Typical CCW pump rotation when viewed from suction end

In closing, I offer a note of caution. Periodically I would arrive and the electricians would say they had already checked rotation by running the pump. They would say that “rotation was all good because there was flow when we test-ran the pump”. Unfortunately, centrifugal pumps will pump quite well running backwards and correct rotation cannot be confirmed by this method.  We would now not only have to check rotation using one of the two procedures described earlier but also physically check that the agitator and possibly the impeller were both still tight.

Bye for now!

RJ