API RP 687, Chapter 1, Section 10: PHASE I Rotor Inspections - Dimensional inspections (PART 1)

Sample of the forms filled during a Phase I.

Today, let me tell you about the dimensional inspections that are part of the API 687 Phase I rotor inspection.

Below, I list the typical sequence of events in a Phase I inspection, and in my last post, I covered about half of the steps.

List of activities in a typical Phase I Rotor Inspection.

First a quick recap:

Let’s remember that the purpose of the Phase I rotor inspection is to assess the condition of the rotor that is being considered for overhaul.

For this we can divide the tasks into three main groups:

1. Visual Assessment and Non-destructive Evaluations

2. Dimensional Inspection

3. Check Balance

   
   

Let’s dive into some interesting details of these inspections, starting with the “Visual” ones.

Visual

Visual inspections reveal interesting macro features that are easy to spot.

We can easily detect the effects of the rotor being in operation and “wear and tear”  or even collateral damage suffered during the installation, removal, or transportation of the rotor.

The most common findings are:

• Scratches on shaft areas or probe areas

• Erosion

  • on seal areas due to interstage leakage

  • on airfoils or impeller vanes

• Corrosion on shaft areas

• Pitting on steam turbine disks

• Foreign object damage on airfoils or impeller vanes

• Rubbed areas from rotating and stationery touching in operation

• Galling over coupling areas

• Broken threads

• Large or unusual balance corrections

• Displaced shroud bands or missing tenons on steam turbines

• Cracked or chipped coatings

Non-Destructive Evaluations (NDE)

NDE will provide more details on some of the visual indications or detect features that would have not been noticed with the naked eye.Most of these indications would also be representative of damage or wear during operation, but often more severe.

Some of the indications revealed during non-destructive evaluations may point to chronic or deeper issues, such as material fatigue or stress corrosion, that develop over time.

The most common:

• Heat checks or small indications/cracks related to rubs between rotating and stationary components

• Cracks or indications around keyways

• Indications around balance holes

• Indications on steam turbine blades (tenons, root profiles, high stress areas)

• Indications on impeller vanes (leading edges, training edges; at the connection to the backplate or coverplates)

• Indications on threads

• Indications on coatings

Remember that the term indication is to call out “any feature of interest” that should be interpreted and evaluated.

When we detect an indication or a crack, we must avoid passing immediate judgement. Instead, we should determine the cause through proper evaluation.

If an indication is relevant, meaning it impacts on the safe or reliable operation of the equipment it should be addressed.

Now, the next set of inspections.

Dimensional Inspections

Steam turbine and compressor rotors must fit inside cases, and they must fit very tightly without any of their parts touching or rubbing.

A tight fit is required to maintain proper clearances on seals so the working fluid (steam for turbines, or some process gas for compressors) stays within what we call the “flow path”.

The steam and gases we expand or push through steam turbines and compressors are like mighty rivers. We need to keep them within their channels. If flow escapes the defined path, you will start to see erosion.  The greater the erosion, the greater the efficiency loss and damage the machine will endure during operation.

Here comes a great challenge: how do we seal or keep gases from escaping if we cannot create a tight contact seal since rotors are spinning?

Let me tell you a secret. Inside a steam turbine or compressor, we really don’t completely seal the rotating shaft against the case. We just try to “discourage” as much of the flow from not travelling outside the intended flow path.

It is mostly on the outermost seals of the case where we have what we call atmospheric seals. This is where we must keep an airtight seal to prevent losing steam or releasing some nasty gas into the atmosphere.

To illustrate this, let’s look at this cross section I drew, of the very first centrifugal compressor I had to rerate back in the day.

Gif illustrating how close components fit inside a compressor case

Outer Diameters

To assess and control the clearances between the shaft, rotor, and impellers, we measure all the different outer diameters or ODs on a rotor assembly.At every step change, we take a measurement.

We can then compare all these dimensions to the original prints, to previous inspection reports, or compare them to the Internal Diameters (IDs) of the case to calculate clearances.

Engineers following API RP 687 expect to see the data collected in a form like this:

Sample OD Report from API RP 687

In each one of those white boxes, you would find the measured diameter.

Now you can appreciate on a multistage rotor like this one, inspectors will easily have to record dozens of measurements.

If there is no room on one page, you may need to add additional pages to fit more measurements.

Also consider the fact that ODs should be measured with a resolution of 0.0001 inches (0.0025 mm).So don’t rush your inspections or you may end up with inadequate numbers.

Journal Diameters

Journals require very tight (precise) dimensional control. These measurements are taken with a resolution of 0.0001 inches (0.0025 mm) at multiple locations and positions.

Remember that on fluid film bearings, the metal shaft runs inside a metal bearing, floating on an extremely thin oil cushion, just a few thousands an inch thick.

Journals need to be round, straight and smooth!

To determine if they are round, we measure them at three positions around their circumference.

To make sure journal areas are not tapered (conical) we measure diameters at three axial locations.

Typical locations where diameters are measured on a journal.

These measurements will be recorded on a form that looks like this:

Sample of an API RP 687 Journal OD Report

If you look closely at my form, you will notice that under each journal I have a place to record the measured “journal weight”.

We may already know the total weight of the rotor, but we also need to know what the weight of the rotor is under each journal, so we can calculate balance tolerances.

Sometimes we can assume rotors are “symmetrical” around their midspan and estimate that each journal weighs half the total rotor weight. But this is not always true, and if you have a scale, it is always best to measure.

To ensure accurate determination of balance tolerances, we measure the journal weight of a rotor as follows. We support one of the journals on a sturdy pedestal and lift the other journal with a scale or load cell. Make sure the rotor is completely horizontal, and the scale will display the journal weight. The procedure should look like what I have illustrated below.

Sketch on how journal weights should be measured.

Electric Runout (ERO) Probe Location

The probe areas are basically how you measure the displacement or vibration of a shaft while it operates.

Shaft vibrations are the main indicator of rotor condition or well-being that can be measured during operation, much like taking a machine’s pulse during operation.

Now imagine going to the doctor and getting hooked on an EKG machine. If you have a very hairy chest, you may have trouble getting the electrodes to stick to your skin and get good electrical readings.

The same principle applies to probe areas on a shaft. We need to make sure the probe areas are smooth and have extremely low Electrical Runout or ERO. These probe areas need to be ground and burnished to be very smooth mechanically and in electrical signature to get clear readings without unnecessary “noise”.

There is no official form but I created my own, and I made it look like this:

My version of an ERO Location Report

Once the machine is running and operating, we will be reading displacement quantities as low as a few thousands of an inch!

A good machine will typically exhibit vibrations lower than 2 mils or thousands of an inch (0.002”) in amplitude. This is half the thickness of a sheet of printer paper!

On a later post we will do a deep dive on charting probes since there are tons of details to discuss there.

Coupling

Last but not least today, we measure the coupling.

The coupling surface on the shaft is critical because it through this surface and the interference fit between the shaft and the coupling hub that the torque gets transmitted between machines.

There are different types of couplings used in turbomachinery, but in API 612 and API 617 equipment, tapered couplings are the most common.

A taper is a cone, and measuring its diameter with a precision of 0.0001 inch (0.0025 mm) requires some skill, care, and a special tool.

The form that captures this information looks like this:

Sample Coupling Diameter Report

I hope these explanations shed some light on the reason why the things we do and measure during a rotor inspection are important.

Once I start to write I realize how many details there are, and it feels like I will never end explaining or exploring tangents to each topic.This is a journey for sure, that for now it seems it does not have an end.

Nonetheless, we shall take one step at a time.