The Reason for Condition Monitoring

Okay, folks, um, thank you very much for coming along. Uh, we're going to talk in this short half an hour session on condition monitoring and very importantly, discuss condition improvement.

Um, as a subset of that.

Um, my name's Dean Whittle, I work for a company called Reliability Maintenance Solutions.

Um, I'm a training manager for RMS.

So, hopefully this information we're going to convey today will be clear, concise.

And hopefully very visual to to make it understandable.

Um, there's a lot to go through, so.

Here we go.

Okay, so we have this.

Um, opportunity, um, in front of us.

To monitor and understand the health of our assets.

Obviously, a key aspect of that is to improve availability.

And indeed, um, reduce downtime.

So all the blue things are good things.

And all the red things are not so good.

Things we want to actually improve.

So, you know, less downtime.

More availability.

So increase capacity.

So we're not going to dwell on these too long, but just key points here.

Things that we're looking to actually improve.

Um, less planned downtime.

You know, if we don't need to take planned downtime, let's not do it.

Let's measure on condition.

The benefit is, um, less costs.

Reduced waste.

So there's a number of benefits that if we focus on these key areas in red.

That will glean and gain aspects of the blue areas in, as you see above.

I guess shareholders are kept happy.

Regulators, statutory requirements on assets, you know, we're complying with conformity by measuring condition.

And equally, maybe our employees are happy as well.

But the key point is, how are we going to actually achieve this?

How are we going to improve condition and and understand the condition of our assets?

Well, the key aspects here are management of assets.

Management of people via the use of condition monitoring.

So we're going to use condition monitoring methods, but equally, not just focus only on measuring condition.

But equally as important, measure, um, focus on condition improvement.

So we're looking to improve the condition.

Not just measure, um, the condition and wait for something to degrade.

Let's improve the condition.

But as we work through, we'll see that.

So we kind of got this really, really focused responsibility.

With many, many complex machine types.

That we want to understand the condition.

Um, you know, if if we don't understand the condition, those machines will eventually run to some point and then fail.

So measuring condition will allow us to understand the health.

And you know, the responsibility we've got is these machines are very complex.

You know, they're delivering a very important function to the business.

So it's not just the asset here, a compressor.

It's the individual parts of that compressor.

You know, the fact that the bearing we're going to use is correctly lubricated.

It's fit for purpose.

It's got the right clearances.

And it's installed properly.

And it's a quality bearing.

We buy that.

We buy quality.

We buy reliability.

Equally, the gears, the same applies.

You know, obviously the gears need to come together, need to mesh properly.

The right lubrication needs to be used.

Um, and obviously our gears need to be aligned, etcetera, etcetera, etcetera.

So we won't dwell too much more on that.

So, condition monitoring, we're going to, um, minimize the risk of failure.

By understanding the condition.

Via early fault detection.

Okay, so we're going to measure machines.

Um, obviously failure will happen.

But it can be avoided.

I think that's the message.

We can avoid failure by understanding condition.

So equipment gives us warnings.

It talks to us.

It doesn't speak English, it doesn't speak Scots, but what it does speak is signals.

As we'll see in a moment.

You know, it speaks vibration, maybe it speaks sounds, ultrasonics and so on.

There's a number of things we can use.

So the principle, the basis of what condition monitoring is.

We've got an asset here that runs for a duration of time.

We, we trend, we track, we monitor, we choose the right measurements.

And it runs defect-free.

Happy days.

Good asset.

No problem at all.

But then at some point, and quite soon after a defect is initiated, we have the potential for failure.

Maybe here it's rather exaggerated, maybe that time is maybe a little bit shorter than that.

Generally speaking.

And from that point forward, we're going to see a degradation in the asset health.

Okay, so we're we're moving more and more closer and closer to actual functional failure here.

And there's a journey we take where vibration and oil analysis and ultrasonics.

Um, you know, and and infrared thermography can be used to measure the condition.

And understand at different stages of failure where we are on this kind of P to F interval.

Point of detection through to point of of of failure.

Of the asset itself.

Um, you know, arguably if we can see things much, much, much, much sooner here, for example.

The planning time we've got in green is great.

We've got a lot of planning time.

The the risks, the safety incidences, the the cost of any incident here is very small.

But the longer that defect's left unattended to, then maybe here ultrasound starts to wake up and tell us a story.

The bearing in the machine's maybe starting to show signs of distress.

Then we see some vibration behavior starting to wake up.

But you notice now that planning time is starting to reduce.

The amount of time to failure is now reducing.

And the risks and the costs will grow as time propagates through to functional failure.

Obviously, here we may have more heat.

You know, the risks are increasing, the costs are increasing, etcetera, etcetera, etcetera.

Through to, you know, the the bearing is virtually collapsed and it's rattling around.

And there's no contact now between the race before it actually functionally fails itself fully.

And our risks, our costs are very high.

And unfortunately, we're at no time whatsoever now to plan any action.

So, these these, sorry, these warning signs.

What are these warning signs?

We talk about.

The warning signs.

We can use, um, for example, vibration.

A very quick snippet of vibration here.

If we take the sound of vibration, maybe just from the motor as a whole.

You can hear that sound of the motor running away.

Or maybe the pump.

Is that coming through?

The sound?

It's a bit faint, maybe.

Thank you very much.

Cool.

So that, you know, that's that's the entire sound of the pump as a whole.

But but individually, the pump blading.

May produce this sound.

So that's a completely different sound to the pump vibration as a whole.

Maybe the bearing in the pump.

Has a defect.

So that's a high pitch sound.

So collectively with vibration, we're able to understand each different heartbeat.

And understand which part of the machine is causing these sounds that we can ought to be here now.

But what's responsible for this increasing signals and increasing vibration?

I mean, collectively as a whole.

With a good running pump, if we measure the pump head.

Anyone got any idea what the sound might be like?

No.

Let's let's maybe hear it.

Maybe this is a good sound, maybe it's a bad sound.

Anyway.

So.

Maybe our machine doesn't really sound like that.

But still.

Anyway.

But the key point is, with vibration, we're able to home in inside the machine.

We're like, we're like seeing through the machine by measuring the vibration on the bearings.

Measuring its heartbeat here with a vibration system.

To understand which component in the machine is degrading.

Okay, that's the key point there.

We can see the bearing failing, we can see the coupling, the blades, the heartbeats.

Of all of those different components is different.

Um, equally, we could look at measuring.

Uh, relative temperatures here with a thermal imaging camera.

So here I can scan with this camera and look at different phases through this supply.

Um, obviously, we have a a a hot spot, sorry, there on the yellow phase.

Equally there on the yellow phase.

And I can read the uh spot temperature.

We have to be aware of some um influencing factors that could affect the values we see.

But it's very, very clear technology that, you know, we can see hot or indeed cold spots.

If appropriate to the asset we're measuring.

With our vibration systems, we can use motor current signature.

Motor motor circuit analysis to understand the condition of the rotor.

The stator and the actual insulation inside the machine.

Um, equally with oil analysis, we've got oil wear particle analysis.

We can measure the health of the oil.

We can know the oil condition.

We can know whether we've got machine wear developing from inside elements that are breaking down.

And indeed, whether we've got external contaminants getting into the into the oil.

Because it's a very critical path that the lubricant is kept clean.

Obviously.

So, you know, taking a bearing as a very good example, obviously it will have a certain heartbeat as it rotates around the system.

But what's a critical path here to stop it entering a failure mode potentially is to maintain that lubrication film.

You know, that lubrication film is like one micron.

One micron of of size, so it's critically important that we get that lubrication film right.

The right grease goes in, the right volume of grease at the right interval.

Um, obviously we get debris in our oil, um, or our grease.

You know, we might have a five micron particle here.

That's going to propagate and cause damage to the actual component.

The race of the bearing and equally metal particles would, you know, detect themselves.

And arguably with an oil sample, we might see that.

But the bottom line is.

The same applies to gears, you know, a damage to a gear.

A a tooth chips, that's going to erode away the teeth.

They're going to propagate into more and more damage.

We can use ultrasound.

This is very high frequency analysis, um, or or stress wave analysis.

As we sometimes refer to ourselves.

To listen to the sound of the bearing.

Okay, we can listen to that lubrication film, we can listen to the actual bearing.

Um, sound that it's making audibly through a through a monitoring device.

An ultrasonic device, for example.

Here, we've lost lubrication film.

Metal to metal contacts increasing.

Our decibel level will will go up.

Um, less prolonged, that's only going to cause further damage.

So it's important, we've put some grease in now, we feel it's appropriate to put grease into the bearing.

To reestablish the lubrication film.

So we can understand if the grease we're putting in is having an impact on the bearing.

But equally, we check it some moments later, maybe an hour later.

To make sure it's still running sweet and sound.

And our lubrication film has returned.

Okay, so, um, how are we going to apply these programs properly?

Um, it's all very well having high technology, uh, vibration systems or ultrasonic systems, etcetera, cameras.

But how are we going to go about using them in the right way, in the right application?

With the right kind of settings?

That's that's very important.

I think arguably often overlooked, the words mentioned so often, but, you know, it's very important that we focus our asset on critical assets.

We understand the consequences of failure of assets.

The likelihood of failure.

And then we can start thinking about the appropriate method of maintenance to maintain this asset.

Understanding its failure modes.

The failure rates.

The P to F interval, how long or how short could that be for each failure mode?

And the detectability.

Now and only now really can we start to say to choose the right type of technology.

Okay, it's very important we understand the failure modes.

And we apply the right technology or technologies to monitor the health of that asset.

That's very important.

It's not rocket science.

It's something that can be done in a very fluent, clear way.

And the knowledge is on plant already.

But is there more we can do?

We can monitor condition, we can measure the health of the bearing.

We can understand the lubrication film.

But is there more we can do?

We can eliminate root causes of failures in the beginning.

Not just focus on monitoring a bearing that eventually is going to fail anyway.

With condition monitoring, I'm not going to make it better.

Just because I condition monitor it, it's still going to fail.

Unless I do something about what the condition monitoring guy is telling me to do.

I replace the bearing.

So, you know, it's it's it's critical here.

We're not just focusing on monitoring.

We're focusing on condition improvement.

So if we can kind of wind this story back and this guy here.

You know, if we take him back to to this point in time where, you know, we had this area where.

Machine's running perfectly, quality of product it produces is great.

No health conditions whatsoever.

It's running sweet and sound.

Happy days.

So let's let's focus on improving condition.

And just bear with me a second.

Let's take that just to the extreme end.

For this to work, I think.

There we go.

Here we go.

So we're going to wind back that P to F interval.

Pull it back into this defect-free zone.

And then expect and continue to run not weeks, days, maybe months, but years.

Yeah.

Years of running reliability from an asset because we're taking care of it.

We're focusing on causes of failure.

Not just monitoring failure with condition monitoring.

Is that's basically what the goal of condition monitoring is?

Is to detect, monitor, trend and track.

Degradation.

This is going to improve reliability.

By doing focused root cause analysis.

At some point in years in the future, sure enough, failure could then redevelop.

But we condition monitoring still to see that story.

Okay.

So, you know, root cause.

What what's this all about?

I mean, the number of machines that I see, pumps that cost 50, 100,000 pound installed.

On very flimsy structural configurations, the base plates eroded away.

The concrete, you know, the base, the support for the machine is not good before you even put it in.

That's a bad start to begin with.

But we've got a story here where we've got bearings and couplings that we need to align properly.

We need to actually make sure the bearing's fit for purpose.

Um, if the bearing is right for its application, and indeed the lubrication is correct.

Don't just pick up the nearest grease gun.

Put the right grease in in in the machine, in the bearing.

Align the two bearings.

The bearings need to be aligned properly.

The shaft needs to be aligned.

The machine needs to be balanced properly.

The air should circulate freely through the system.

No restrictions.

If not, that's going to cause forces back on the machine.

Cause more stress.

So if we take this example here, this fan's trying to show kind of an imbalance story.

That's going to cause a a a radial force on the bearing.

So this bearing, in a way, it's it's contained within the bearing housing.

It wants to do this.

This is the motion the bearing's trying to actually take.

But it's constrained by the bearing housing.

But at the end of the day, the forces are still in the machine.

And those forces will cause additional stress.

Additional forces on that bearing.

Reducing its life.

So, what's the solution?

Root cause, quality, balance your machines to a tolerance that's precision.

And maintain that.

Reduces the forces.

Quality align the machine, etcetera, etcetera, etcetera.

So obviously, a machine could be misaligned, maybe it wouldn't visually move that much.

But if you bolted that machine down and it's misaligned.

Those forces go to the bearings.

Those forces go to the shaft seals that cost thousands of pounds.

Premature failure of rotating component assets due to a very fundamental, simple thing to do.

Align the machine.

It's not rocket science at all.

So maybe the the two shafts are misaligned angularly.

That's going to put stress and load on the raceways of that bearing.

Maybe the bearing's cocked on the shaft itself.

Not common, maybe, but if the shaft is worn, that could be cocked.

It's cocked in the outer race, so the bearing could be cocked over in the outer race.

That's going to induce extra stress, extra forces in that bearing.

Common today.

Variable speed drives.

Variable speed drives.

Can induce the potential of what we call EDMs, um, their discharges.

Basically, spark erosion through the bearings.

That's a common failure mode on a variable speed drive.

Root cause fix.

Potentially, earth the machine properly.

As a simple, simple solution.

Or majority bearings today in that situation will be insulated.

So they insulate the bearings to stop the currents flowing through the bearing.

Okay.

Okay.

So we've got a standby machine.

It must be perfect when we want to use it.

It's perfect.

Well, maybe not.

The pump beside this machine is vibrating.

That transmits vibration to the standby machine here.

What results?

Static indentations.

We call it false brinelling.

So the machine's not rotating and it puts distinct wear marks on that machine.

We start the machine up three months later because we want to do something on the other pump.

And we've got a problem straight away.

So not good situations.

We need to eliminate that as a root cause.

We're back to this lubrication film again.

I I you know, I keep kind of talking about it, but it's it's a very, very critical path.

It's absolutely essential it's maintained.

We've now lost lubrication film.

What's going to result?

Stress, increased friction.

If we're measuring ultrasonics, we're measuring vibrations, they will deteriorate.

But arguably, unless we do something about it, it's still going to fail.

So we need to establish that lubrication film.

And make sure it's it's correct.

It's a very critical path in a bearing.

Okay.

So we we we can use all this technology.

We can monitor condition.

And, you know, with vibration, we can see these shock waves, you know, we've got a repetition here.

Shock, shock, shock.

And we can clearly measure those and we can say with vibration.

It's the inner race, it's the outer race, whatever.

But one thing we cannot do.

Is make the machine more reliable.

What makes it more reliable is getting to the root cause.

Just because I condition monitor it, it doesn't make it better.

It doesn't make it better.

Getting to the root cause, putting the root causes to bed is what makes it better.

Improve condition.

Underline again, improve condition.

So condition monitoring is great.

But hopefully the story we we we are conveying here is we need to go beyond that.

We're not just focusing on monitoring bearing faults, we're looking at improving reliability of an asset.

We've got all this technology, but we need to focus on improving reliability.

Okay, so we need to go beyond standard maintenance practices.

Standard condition monitoring is the message.

We want to improve reliability.

It starts before the machine is ever put in position.

At the design stage.

The manufacturing stage.

The procurement stage.

The shipping stage.

The installation stage.

Let's look at that journey.

So we've got this machine.

It's precious.

It's a precious machine, it's worth a lot of money.

It's critical to production.

And we must maintain the reliability of this asset.

So before it's ever even installed, the journey starts.

But what we're going to do here is create a force field.

Okay, so we've got this kind of force field around the machine to protect it.

And we're going to stop the unreliable suppliers coming into this force field.

We're going to stop cheap spares.

We buy cheap bearings, we buy cheap seals, we buy cheap.

Do we buy reliability?

Possibly not.

The way the machine's transported before it's going to be installed is critical.

So we've got to make sure that's that's fit for purpose.

The transportation methods.

It's contractors that are used.

You know, we we've got to get these things right.

Because if not, we're buying problems before it's even installed.

Not good.

Poor design, obviously, design stage.

We need to make sure it's fit for purpose.

So once we've kind of got those things right.

And that starts in the bigger picture, no doubt.

We've got to say that anything that comes into the plant is then fit for purpose for the asset.

So that if we need to replenish grease on this machine once every six months.

That the grease is is the correct grease.

Not just the the nearest grease gun is picked up and used.

It's the correct grease and it's not contaminated.

Because if it's contaminated, we've seen previously what can result.

We've got to make sure that parts we use from store.

Stock, they've been stored properly, they're clean.

Dust, moisture has not contaminated before we put these parts into the machine.

Poor installation practice.

We've got to make sure that we actually install our machines properly.

I cannot overemphasize the point I've made about the base.

You know, a nice stiff, rigid base with a quality machine bolted onto it.

That's a good story.

But how many people focus on putting a quality base in?

Quite few to be fair.

Surprising.

We all think people do, but do they?

Shocking, to be honest.

Poor operating practice.

If you maybe with condition monitoring, you have the potential to install online monitoring.

One one benefit that tends to bring is you see a lot more information coming back more regularly.

And you can almost say, you know, guys, last night on that ship.

You were not operating that pump properly.

You know, that pump needs to be.

Cared for in the way it's used.

Just like your motor car.

You know, you don't thrash it down the motorway every every second of the day and come to a grinding halt and so on and so forth.

Because you know, we're going to take care of the way we use the machine.

Unnecessary PMs.

Why do we need to do a PM if we can condition monitor it and it's saying it's fine?

Is this an appropriate method of monitoring?

We don't necessarily need the PM.

Intrusive inspections.

Can they be alleviated by using condition monitoring?

So we use condition monitoring to take away the need to do an intrusive inspection.

We measure it, trend it, track it while it's online.

Doing its duty.

So whilst this is kind of a maze of different points.

They are very simple points to be fair.

And if we get those things right, we've now got the best possible chance.

Of of a a reliable asset with long life.

Good reliability figures.

Obviously, what's important, and I guess what that's what this is for.

We still continue to monitor the health of the asset.

That's still a need, please.

You know, that's still something we need to do.

Um, to continue monitoring.

If we don't do these things right, then basically we're inviting trouble.

Before the machine's ever used.

And arguably, what's going to happen?

The balls are going to get the machine, they're going to make the machine angry.

And failure will will result.

So that's not a desirable situation.

Okay.

Just checking the time here.

So.

Okay.

Um, I'm quite worried about the time I had to be fair.

But it looks like I'm going to finish possibly a moment or so earlier.

So in conclusion.

Breakdowns can be frustrating.

You know, but with condition monitoring, we have the ability to listen.

To see inside the machine and understand condition.

However.

What condition monitoring on its own doesn't do, it doesn't improve reliability.

Improving reliability involves taking a proactive approach.

Looking at root causes of why bearings fail.

Why couplings fail.

Why shaft seals fail.

Etcetera, etcetera, etcetera.

If we get those things right, that's what will increase reliability.

So we're going to focus on condition improvement.

So there is a solution.

We monitor our equipment, we monitor the condition to prevent catastrophic failure.

Um, obviously we get improved safety.

And planning, we, you know, we're in control.

We can plan when we're going to do something.

Because we know the condition.

And equally as important, as I've kind of tried several times to focus on.

We're going to focus on improving the condition.

You know, improve condition, we will increase reliability.

We will get more life out of that asset.

And that's good for everyone, good for the business.

So, I guess at the end of the day, I guess happy days, everyone's happy, and we have a solution to our problem.

Maybe with the correct technology and the correct methods applied.