Solar Turbines – FT8 & Solar – Turbine Technical Information https://www.dmba5411.com Training and Troubleshooting documentation and videos. Tue, 12 Nov 2024 19:47:16 +0000 en-US hourly 1 https://wordpress.org/?v=6.6.2 ART242 – Solar Mars Borescope Inspection Part 2 https://www.dmba5411.com/art242-solar-mars-borescope-inspection-part-2/?utm_source=rss&utm_medium=rss&utm_campaign=art242-solar-mars-borescope-inspection-part-2 Wed, 19 Jun 2024 19:38:55 +0000 https://www.dmba5411.com/?p=8368 Continue Reading →]]>

ART REF – ART242

This is part of a Solar Mars training course. This video discusses the Diffueser and Combustion Chamber.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

In Part 2 we will look a the exit of the compressor through the diffuser access ports and we will also review the combustion chamber insprection requirements.

Just aft of the compressor is the diffuser. It has four borescope plugs, giving access to the rear of the compressor exit guide vanes.

You may have to remove some of the thermal blancket around the turbine to gain access to the ports.

A 6 mm flexible borescope is used for this inspection. The 6 mm flexible borescope is the most commonly used, because it has a good light source, while being small enough to get into most areas.

The smaller diameter scopes are great getting into tight spots, but if used in a large area such as the combustion chamber, you will not be able to see very much until the tip is very close to the item.

Any metal parts liberated from an earlier stage of compression, or if some solid item enters the front of the turbine, it will pass this last stage vane. If the part has enough mass it will dislodge other parts of blades and vanes along the way. This can end in what is known as corn cobbing. It is common when carrying out a quick inspection of the turbine, to check this location, as if anything serious happened to any of the compressor blades, you will see some damage at this location.

Combustion chamber

Moving back along the turbine we have the combustion chamber.

There are obvious differences between the standard combustion and SoLoNOx combustion chambers, however the general defects you will come across are similar. In fact the type of issues discussed here are typical of any gas turbine.

There are fourteen SoLoNOx injectors, or twenty one standard injectors fitted to the Mars. It is important that the injectors have their location marked when removing, to ensure they are replaced in the same location.

The condition of each injecter needs to be documented for the report. The condition of the injectors will depend much on the cleanliness of the fuel used on site.

If you are using liquid fuel there should be ongoing control, of the supply and storage of the fuel. Liquid fuel degrades the turbine hot section quicker that gas fuel.

Contaminants in the fuel will be seen on the injector and the combustion chamber and hot section. These normally discolour the metal where it is deposited.

If the fuel supply is coming from a wellhead or production area you can get all sorts of contamination in your fuel. Pipline gas for example is ideal, but not always available.

Because of the small diameter orifices in injectors they often block up, especailly the SoLoNOx injectors. This will have an effect on the flow to that injector, often supplying less fuel than the other injectors. If this condition is identified you will need to clean or change the effected injectors. The T5 spread profile will change and is the primary indicator of injector problems. We will discuss this shortly.

A common proble with gas fuel is sulfur in the fuel. This can be very difficult to remove as it often passes through filters before “coming out of the gas” as the gas expands entering the combustion champer. This deposit of sulfur can effect the movement of fuel valves. Sulfur when mixed with other chemicals will casue hot corrosion which will be describel later.

Removal of the injectors is the primary access, to carry out a borescope inspection of the combustion chamber, and first stage nozzle guide vanes.
There are also four borescope plugs that can be removed from the combustion casing. These can be used for a preliminary inspection if need be, but the main inspection is through the injector mounting location.

Access to the combustion chamber:

We will normally access the combustion chamber via the fuel injectors.

Here we see a diagram of one of the guide tube field tools that are used to guide the borescope right to the first stage nozzle guide vanes.

A shorter guide tube is also available which allows the external inspection of the combustor, liners and baffles that protect the combustion casing from high temperatures by directing layers of relatively cooler air to protect it.

The principle guide tube, bolts to the injector mounting holes. A copper pipe-bending-spring is then passed through the guide tube, and finally the borescope flexible lead is passed through the spring. The spring bends going around the ninety degree turn entering the combustor, and internally guides the borescope lead to exit in the correct direction of the nozzle guide vanes.

This makes it much easier for inexperienced technicians to carry out the inspection.

We are now going to discuss the damage commonly found in the Combustion Chamber:

Fretting due to combustor vibration.

Fretting is when two or more pieces of material slide over one another repetitively causing wear.

Because of the need for the combustion chamber to expand and contract, it is hard to avoid surfaces where one slides over the other.

There will always be a certain amount of vibration present in the combustion chamber due to combustion pressure pulses.

And these pressure pulses can become a major concern during SoLoNOx operation, when the flame is running very lean.

These pressure pulses on newer Mars untis are monitored using a system called “Burner Accoustic Monitoring (BAM for short)”.

Cracking:

Cracking is a common problem on turbines with high operating hours.

The cycle of expansion and contraction of metals, vibration, stresses from distortion, erosion of surfaces make cracking quite common.

There are no published limits from Solar regarding allowable tolerances with respect to cracks.

Any cracking of structural material is a cause for immediate action, whereas cracking of the annulus and heat shields are less critical.

The length and location of the cracking should be recorded and then seek the opinion of Solar. The recommendation may be for a reduced time interval between inspections.

The danger with cracking is the possibility of a flame pattern change as the cooling flow of air may be effected. There is the added obvious danger of metal being liberated into the turbine.

Bulges and deformations.
Bulges and deformations are often caused by the combustor metal not being able to expand and contract normally. Once this happens the airflow is disrupted, which can lead to other issues such as local overheating, cracking and material liberation.

Erosion.
Erosion often eats away the leading edges of blades and vanes. The particles which cause the erosion flow along with the gases, and impinge on the metal surfaces, eroding it. This can be material such as salts, sand, pollution etc, entering the air intake. The erosion can be from soot and coke, the result of fuel quality. Or from particles liberated from thermal coatings.

Burning.
Burning is often caused by the combustor flame not being kept in the center of the combustor. The fuel exiting the injector is swirled and mixed with air, to improve the mixing of the two, and to help keep the flame in the center. If any of the injectors becomes partially blocked or is damage, it will not be able to provide a proper mixing of the air and fuel. This has the effect of heating some areas more than others. It will cause buckling and burning of the combustor metals, and hot spots on the nozzle guide vanes.

Sulfidation – is not so much an issue in the combustion area so it will be covered in the turbine section.

Injectors

Just as with the spark plug of an internal combustion engine, the injector condition can give you much information, as to what is happening inside your combustion chamber.

All injectors are not equal, and you will find that SoLoNOx injectors are much more complex components, than the basic gas or liquid injector. They are more likely to crack, erode, overheat, form blockages etc.

Technology has made amazing advances, but it often comes at a price. Solonox injectors are costly and complex items to manufacture, and will not tolerate poor quality fuel or air.

It is difficult to clean injectors that are very dirty. The high temperature in this location burns anything inside the injector flow orifices. Sometimes the only way to clean the injector is to heat the injectors to a very high temperature in a special oven, that converts the dirt into dust. At the same time the oven injects gases through the injector, to remove the dust. The injector has to be heat treated after this procedure. The injector is normally flow tested after this process, to ensure it has not changed significantly and is within specification.

If the fuel being supplied is not very good quality, you might consider an additional inspection, at less than 4000 hours. This could be done by removing a few injectors, to allow access with a borescope, to inspect the rest in place.

While not a part of the borescope inspection, the T5 profile gives an early warning to pending issues in the hot section.

The T5 measurement for each thermocouple is monitored for deviation, from its normal profile. Changes to the profile means a possible problem upstream.

For example if one injector is partially blocked, due to dirt in the fuel. Some of that fuel, will then be diverted to the other injectors, which will in turn, run a little hotter and the blocked injector will run cooler.

As well as seeing changes to the temperature profile, the flame at that partially blocked injector, may move position. This may cause localized heating of a casings etc, with its knock-on effects. This can be seen as a shadow or change of color on the combustor.

The way the T5 profile works, is that when new, the temperature profile is recorded on a graph, similar to the one in the presentation. This is the, “as new” condition, which is typically when the turbine is new, or just overhauled. What you are subsequently looking for is a change from the, “as new” profile. Sometimes the change in profile is understandable, such as when the injectors are replaced.

]]>
ART241 – Solar Mars Lube Oil Logic https://www.dmba5411.com/art241-solar-mars-lube-oil-logic/?utm_source=rss&utm_medium=rss&utm_campaign=art241-solar-mars-lube-oil-logic Wed, 19 Jun 2024 19:38:00 +0000 https://www.dmba5411.com/?p=8366

ARTICLE REF – ART241

This video is going to look at the lube oil logic for the Mars turbine.

]]>
ART061 – Bently Nevada 3500 – Connect to the rack. https://www.dmba5411.com/art061-bently-nevada-3500-connect-to-the-rack/?utm_source=rss&utm_medium=rss&utm_campaign=art061-bently-nevada-3500-connect-to-the-rack Wed, 19 Jun 2024 12:25:02 +0000 https://www.dmba5411.com/?p=8160 Continue Reading →]]>

ARTICLE REF – ART061

]]>
ART059 – Solar – T5 Control https://www.dmba5411.com/art059-solar-t5-control/?utm_source=rss&utm_medium=rss&utm_campaign=art059-solar-t5-control Wed, 19 Jun 2024 12:22:45 +0000 https://www.dmba5411.com/?p=8156 Continue Reading →]]>

ARTICLE REF – ART059

 

Here is an interesting question – what is the difference between air at station 1 and 2?

Answer a little further on.

 

T5 is the abbreviation for the gas temperature at station 5 or temperature of the gases at the exit of the gas generator turbine. All industrial and aero turbine manufacturers use this station designations and they have found their way into the names of the instruments used in these locations.

How a thermocouple works

The following information which explains the subject perfectly was taken from TEGAM.COM ……. Designers & manufacturers of test & measurement instruments.

What Exactly is Cold Junction Compensation? How Does It Relate to the Use of Thermocouples as Temperature Sensors?

I was recently asked, “What does Cold Junction Compensation do?” This question requires that I summarize how thermocouples work and the physics behind them.
A thermocouple develops a thermoelectric voltage based upon the principle known as the Seebeck Effect. Now, what is the Seebeck Effect?


In the 1800s, Thomas Seebeck was trying to make electricity from heat and he experimented with a circuit of Bismuth-Copper and Bismuth-Antimony and showed that when the two junctions of the two materials were at different temperatures they produced a sustained current; converting thermal energy to electric energy.

I am going to try to turn this into an analogy that will make it more understandable:

Think of a hollow tube with low pressure on one end and high pressure on the other. What will happen? Air will flow through the tube from the high pressure end to the low pressure end (like blowing through a straw). To relate this to a thermocouple, at the hot end the heat has excited the electrons and they are moving faster (high pressure). At the cold end the electrons are closer together and moving slower, low pressure. The result is that electricity flows from hot to cold. The greater the difference in temperature, the greater the voltage – The Seebeck effect.


That is the basis of a thermocouple. What is not clear yet is how the voltage measured is converted to a temperature measurement. The amount of thermoelectric voltage that the thermocouple produces is based upon the difference in temperature between the two ends (and the material). Difference in Temperature? Yes, a thermocouple actually measures a differential temperature – not the actual temperature at the hot end. In order to know what the temperature is at the hot end, you need to know the voltage produced AND the temperature at the other (cold) end.

To make the thermocouple a useful tool in the 1800s, the cold end was put in an ice bath; something that any laboratory could readily reproduce. Next, thermocouple voltage tables were developed based upon the cold junction end being in an ice bath. At that point, the thermocouple wires were connected to copper wires with no thermoelectric effect and carried by the copper wires to the measuring device. The voltage was recorded and looked up in the tables referenced to the ice bath (0°C /32°F) and the hot end temperature was derived.

Today
COLD JUNCTION COMPENSATION replaces the ice bath. An electronic circuit replaces the ice bath by adjusting the voltage, compensating as if the cold end was in an ice bath – hence cold junction compensation.

A thermocouple thermometer has a connection point where the thermocouple attaches to it. This is the “cold junction.” It is not at the ice point (0°C /32°F) so the thermoelectric voltage measured does not correlate to the thermocouple tables. The point where the thermocouple is connected to the copper connections of an instrument is the cold junction. Either a precision RTD or thermistor is used to measure the temperature at that point.

The electronics of the system determine the temperature at this point and then computes the voltage that a thermocouple would make from the ice point to that temperature. The hot end is only producing voltage equivalent to the difference in temperature from the hot end to the instrument connection. This signal is not properly referenced to the ice bath tables. The cold junction circuit compensates for this missing voltage by adding (or subtracting if the cold junction temperature is below the ice point) to the measured voltage coming from the hot end. The combined voltage is then properly referenced to a cold junction at the ice point and can be accurately converted to the true hot end temperature using the standard thermocouple tables.

In Summary
Cold junction compensation compensates for the missing thermoelectric voltage due to the fact that the thermocouple cold end at the instrument is not at (0°C /32°F). This then allows electronics to use the established thermoelectric voltage tables (or polynomials) to determine the temperature at the hot end. Cold Junction Compensation is the reason that the thermocouple moved from the laboratory to the most widely used temperature sensor in industry today.

 

Answer to question at the start
These stations were developed for aero turbines. As an aircraft is flying along at say 600 Km/Hour air entering the turbine inlet duct is slowed down and compressed due to its divergent shape. It is like getting free compression. You don’t have the same effect on an industrial turbine, and you may not see the need for having “Station 2” before the first compressor blade.

Solar T5 temperature control.

The temperature of the gases exiting the combustor and entering the turbine is critical to the life of the turbine. Manufacturers push their equipment close to their limits so there isn’t much room for error. It would be ideal to be able to measure the temperature at this point called Turbine Rotor Inlet. You may have heard of TRIT – this is the temperature at this location. Every Solar Turbine of similar make (eg, Mars 15,000 Gas Fuel) will have the same TRIT limit. However, thermocouples operating at these high temperatures would not be practical. The temperature of the gases is measured after the third stage of turbine.

As stated above all turbines with similar make will have the same TRIT limit. The exhaust gases are measured at Station 5, but as each turbine manufactured and assembled will be unique and there will be slight differences between them. Take two turbines at its TRIT limit, now read the temperature T5 at each turbine – there will be slight differences. At this stage Solar had to make a choice, compensate for this difference so that when it comes to the shutdown limit, they are always the same number or have a different shutdown for every turbine. Solar chose to keep the shutdown limits the same number on all turbines.

To achieve this a compensating resistor was added in the circuit in series with the signal for cold junction RTD. When a turbine was being put in service for the first time or when it is replaced for whatever reason you had to replace the compensating resistor. The resistor travelled with the turbine. These resistors were used prior to the use of PLC. With the introduction of the PLC the resistor continued in use for some time until logic was introduced to replace the resistor.

This is an example showing the compensating resistor in series with the Cold Junction RTD on the Isothermal Plate. This is from a time when the thermocouples went to Isothermal Plate and then were connected to the PLC using copper cables. Now thermocouples are connected to local processors directly, where some of the signal processing is completed before being sent as data to the PLC.

BASE T5
Base T5 is according to Solar the temperature read at station T5 and has a direct relationship to TRIT. Shutdown values are based on this reading, modified by compensating resistor (or logic). On the engine data plate or test cell report you will see values BASE T5 GAS and BASE T5 LIQUID, these are the temperature readings while the turbine was at full load in the test cell.

T5 Setpoint

The T5 Setpoint value and the value of BASE T5 will be the same if there isn’t any T5 compensation. T5 Setpoint is the nominal value expected. In the example below the topping limit will be  will be 1400F for both gas and liquid. When the turbine is running on gas fuel and has a T5 BASE reading of 1317F the control will add in the compensation value to bring the T5 Average up to 1400F.  This way will ensure the same control, shutdown temperature on all of the same units.

Shutdown limit.
Normally the shutdown limit for continuous duty turbines is T5 Setpoint plus 65F degrees. As long as the turbine stays under this limit it will not shut down. Once this value is reached a 10 second timer starts to count and when it times out you get an alarm. It the temperature does not decrease below this value another timer will time out after 20 seconds and will shut the turbine down. There is also an instantaneous limit that will shut you down without delay and this value will depend on the turbine type and duty.

Your PLC control will always be trying to return the turbine to the maximum value allowed (T5 Setpoint). Peaking or Standby duty turbines normally have a higher limit as they need at times to over-fire. These units normally don’t run permanently and therefore they sacrifice the lesser hours to overhaul for having the ability to over-fire the turbine for short periods of time. These units have modified logic to allow them to do this.

Operating temperatures.
The turbine is considered lit-off when the T5 is greater than 400F. There will be a limit while the turbine is accelerating to speed and then there will come under the T5 Setpoint control limit. If any thermocouple has a temperature 200F less than the average of all the good thermocouples it will be removed from the T5 Average calculation and an alarm will be issued to say that that thermocouple has failed. If two thermocouples fail the turbine will initiate a cool down. If one or more thermocouples have a reading 200F greater than the T5 Average there will be an alarm to say there is a T5 spread.

]]>
ART058 – Solar – Mars IGV https://www.dmba5411.com/art058-solar-mars-igv/?utm_source=rss&utm_medium=rss&utm_campaign=art058-solar-mars-igv Wed, 19 Jun 2024 12:21:41 +0000 https://www.dmba5411.com/?p=8154 Continue Reading →]]>

ARTICLE REF – ART058

 

The Mars 90 & 100 use a total of six stages of variable vanes to avoid compressor surge during the acceleration and deceleration of the turbine. As you go from stage to stage on the compressor the blades become smaller. The air is being compressed so therefore the area it occupies is smaller. For optimal performance the blades are designed for normal operating conditions. While accelerating and decelerating the smaller blades are not able to process all the air being delivered to it by the early stages. It needs to be dumped by bleed valves or blocked by closing the air path with variable vanes. In Solar terminology the IGV (Inlet Guide Vane) is the first stage vane and the next five stages of vanes are called Variable Stator Vanes. There is no physical difference in any of the six stages. The angel of the individual vanes is turned by a lever connected to a ring surrounding the turbine. As the ring turns the angle of all the stator vanes on that stage turn by the same amount. All of the six rings are turned together in unison.

The control developed over time from simple surge avoidance to later versions which modifying the airflow to control power. The Moog actuator was used on the early Mars with Turbotronic 1 controls. It had a +/- 10 volt output channel which used +/- 3.6 volts signal to position the actuator. This signal was sent to a volt to milliamp convertor (+/- 50 mA) which was sent to a servo controller to position the actuator. It had a position feedback signal used to create an error for IGV servo. The Moog was not Cenelec or CSA approved therefore it was not allowed to be used in many countries. The Moog was easily identifiable with its bright yellow paint. The alternative to the Moog actuator at that time was the Tactair Electro Hydraulic Actuator. This did not have a position feedback loop but it did have the safety certification required by many countries. It used a 4 to 20 mA signal to position the actuator. The latest actuator in use is the 120 VDC PECC actuator. With the introduction of the electronic actuator the hydraulic system could be removed on some units. These actuators are very precise and quick acting.

The control has also change with time. The initial control was to just to avoid compressor surge. Up until 80% NGP corr (corrected) speed the IGV was in its closed stop position of -45 degrees. The control then opened the IGV linearly, until it was fully open at 92% NGP corr (94% on some units). At this point there wasn’t any further movement of the IGV until the turbine was shutting down. Apart from much better control electronic actuators meant you could get rid of the hydraulic system associated with the IGV. The system was redundant most of the time and constantly using energy.

This is the logic used to give it a linear output of -0.36 volts at 80% NGP corr. and +3.6 volts at 92% NGP corr. In the first instruction NGP corr. speed multiplied by -0.6 and then in the second instruction the result is added to 51.6. The table below shows the result.

 

You can see that when the NGP corr. speed is below 80% the voltage is greater than +3.6 volts. The maximum signal of +3.6 is sent to the output. Once the speed reaches 80% the output increases linearly up to 92% where is reaches the programmed limit of -3.6 volts. There were variations of this logic so don’t be surprised to see slightly different numbers.

 

The Mars had a full open stop at a nominal value of +7.5 degrees although many Mars turbines will not even reach 0 degrees. In the test cell the full open position of the IGV is determined and that number is stencilled on the IGV arm. This number is also found on the turbine data plate and the test cell report. All Mars turbines will be a Match 59 or 122 and what this means is that a Match 59 Mars will reach its T5 and NGP limit at the same time on a 59 degree Fahrenheit day. Operators in hot climates choose Match 122 turbines as they are more likely to operate most of the time at the higher ambient temperatures. The Mars will run more efficiently if its speed is closer to its limit of 103.6 NGP. So selecting the correct temperature Match is important.

Solar introduced Active Guide Vane Control on the Mars T13000 and T15000 which allowed the IGV to be closed-down a little on hot days when the turbine was T5 topping. As stated earlier the Mars will operate more efficiently if the NGP speed is close to its limit. This makes checking the position of the IGV a little more complicated and graphs need to be referenced to determine the correct angle given the current conditions. The transition from surge avoidance to Active Guide Vane Control happens between around 89% to 92% NGP corrected speed. This is an example of the three slope control of a Mars 100 with a max open position of -1.5 degrees.

Calibration of the IGV is done whenever a new turbine is installed and is verified on the 4000 maintenance. There is a fixed relationship between the first stage and the remaining five stages, no adjustment is allowed on the last five stages. On the Turbotronic 1 units with surge avoidance you need to verify the ramp as there are variations of logic. It may be 80 to 92 or 80 to 94 and the best place to check is in the logic. In the example earlier it was 80 to 92 which was on a T12000 unit so you need to check. Even then newer units with Active Guide Vane Control use a surge avoidance and this is the first part of the calibration also for these units

The calibration of the Turbotronic 2 actuators that use a 4 to 20 mA signal is requires you to verify the ZDEGDR_CNT – which means verifying the output counts to the actuator when the IGV is at the 0 degree position. Everything else is measured from this point. The other very important K-value which you have to change or verify, is the GVAG_OFFST which is the full open stop offset to GVAG_FULOP which is the nominal full open value (+2.5) for the IGV. If the full open value found in the test cell was +3 degrees, the value you enter into GVAG_OFFST is +0.5. The actual calibration procedure requires a good knowledge of the logic as it can change form model to model and vintage as well as hardware.

]]>
ART057 – General – Smart monitoring & diagnostics https://www.dmba5411.com/art057-general-smart-monitoring-diagnostics/?utm_source=rss&utm_medium=rss&utm_campaign=art057-general-smart-monitoring-diagnostics Wed, 19 Jun 2024 12:20:34 +0000 https://www.dmba5411.com/?p=8152 Continue Reading →]]>

ARTICLE REF -ART057

If you look at something differently, you will not get the same result!

Introduction

The freedom to be able to collect and analyse the data you want as opposed to what the OEM gives you must be appealing to anyone troubleshooting technical issues. The intension of this article is to show you how to connect to an Allen Bradley and Woodward PLC and write PLC values directly to an Excel spreadsheet. Then give you some examples of what is possible so that you can come up with your own ideas and hopefully share them with us. This article is being co-written with Sergio Stoessel who I have worked with for many years in the field.

Most of the information available on the HMI are analog and discrete values, but there is so much more information available in the logic waiting to be displayed. Add to that ideas which you have such as items you have discovered that nobody had thought about. This is where this article is going.

The article will have three parts, the first two will be this one and the last two will show how to connect to the PLC / CPU (Solar & PW) .

Data collection

This is going to be dealt with in two separate articles, one for Solar and one for PW.

Ideas on how to display your information:

The following are examples of what you might consider doing with the data once you have it.

In this example you have a display of when the pilot fuel valve is opened and closed to control emissions. The graph is constructed as any graph in Excel, the on off lines are defined in the colored table and have fixed values. The red “+” that you can see between the lines is your current position defined by the “x” and “y” value in the red box (NGP and DP). The two lines are fixed while the operation point “+” moved dependent on the values read into the red box. This gives you a visual appreciation for where your operating point is in relation to some limits or control. The surge control screen on a Solar Compressor Set is a good example of this. There are many alarms which don’t have a fixed value and it is nice to be able to see it graphically or to record the value as the turbine is starting to see if some value comes close to alarm etc. This is done in Excel by selecting a Chart Type “Scatter”. The two fixed lines will have various “x” and “y” values defining the lines. The operation point “+” will only have one “x” and “y” value, but it will be changing depending of the “x” and “y” value coming from the PLC.

 

 

Another example is the fuel system of the Mars turbine. This looks at the logic status of solenoids, switches and valves during a fuel valve check so you can use it to troubleshoot issues or for training technicians or operators. Each of the instruments is made up of a box (various cells merged) that changes its color depending where a “1” or a “0” is being written to the cell. The information can be written directly to the cell or as in this case the values were written to the cells in blue and then copied to each valve and solenoid. This uses Conditional Formatting in Excel to change the color depending on the cell value.

 

 

This is a bit of logic from the FT8 which is part of the logic allowing the starter to motor over the turbine. This was done in Excel using conditional formatting. Clearly if all the inputs are not condition “1” the output will be condition “0” as it is shown below. The four inputs that are being used here are coming from other parts of the logic. The top one is the lube oil level switch LSLL601 which is a direct indication of the condition of the switch – it is closed. The second from top is logic that says that the oil temp is greater than 50 degrees C. This is not a switch, it is the analog value from TE601 – which is an RTD measuring tank temperature and then in a separate bit of logic the current tank temperature is compared to the value of 50C and if it is greater than 50C change the status bit to “1”. This is a good example of the type of information that is often not available on the OEM display. The third from top is like the oil temperature because it has separate logic looking at both oil pressure and NH speed to change the status of a bit which is the input to the cell. The bottom signal is a discrete signal like the oil tank level. All that is needed is to connect the four colored cells of the input to the PLC and this will change color on status “0” or “1”

 

 

This is an example of performance data of an FT8. This produced a comprehensive report and is very useful if you use it regularly. You can also use it to evaluate the performance in real time. This was based on a service bulletin from PWPS 96M08 which gives the information necessary to develop it.

 

 

Here is an example of data taken from once the start motor turns the turbine until it reaches crank speed of 2,500 NH on two FT8 turbines. The orange failed to achieve the pressure alarm level indicated by the red line and in the time-period indicated by the blue line. We now know this was due to no accumulator bladder pressure. Because the bladder was depressurized the oil had to fill up the accumulator which would normally be occupied by the bladder. On the subsequent start it built up pressure quicker and did manage to start. Apart from not starting it could also have given problems if there were drop-loads on units with this logic causing the unit to trip. Monitoring this curve during the start phase would give you a warning to check charge and condition of the accumulator.

 

 

Solar used to have three pages of containing every analog value on the turbine similar to the example here. The grey is the current value and the green, yellow and red are the level of normal, alarm and shutdown. This allows you to view 40 or so analog values in a couple of seconds and you see anything amiss straight away.

 

 

Solar used to have three pages of containing every analog value on the turbine similar to the example here. The grey is the current value and the green, yellow and red are the level of normal, alarm and shutdown. This allows you to view 40 or so analog values in a couple of seconds and you see anything amiss straight away.

This is an example of a cooler which is driven by a VFD, cooling hot compressed air taken from the last stage of compressor. The air after it has been cooled is sent back to the turbine to cool bearing areas. As the cooler gets contaminated with dirt and dust it speeds up to compensate for the inefficiency of the cooler.

First you need to put together a table showing how a clean cooler operates at different ambient temperatures and turbine speeds. This data can be taken from historical records etc. The function “FORECAST” is then used in conjunction with your table to predict the motor speed.

The Current Ambient Temperature, NH Speed and Current Fan Speed are read into excel from the PLC and the Expected Fan Speed is calculated. I this example you can see that the fan is running about 10% faster than it should be.

 

 

This is just a sample of some ideas – your imagination it the only limit to what is possible.

]]>
ART056 – Solar – Connect to OPC Server https://www.dmba5411.com/art056-solar-connect-to-opc-server/?utm_source=rss&utm_medium=rss&utm_campaign=art056-solar-connect-to-opc-server Wed, 19 Jun 2024 12:19:29 +0000 https://www.dmba5411.com/?p=8150 Continue Reading →]]>

ARTICLE REF – ART056

 

The following article is going to describe how to copy information from the Rockwell PLC and write it as live data into an Excel spreadsheet using RSLinx as the OPC Server. This should be read in conjunction with article “Smart Monitoring & Diagnostics”.

RSLinx is an OPC Server which communicates with the Rockwell PLC and facilitates the movement of data to and from the PLC to displays, laptops and in also to Microsoft Excel. There are various versions of RSLinx available and not all will work with Excel. You will need the RSLinx Classic or some other OPC Server software to do this. In this article just RSLinx is being discussed as this is what is normally available on sites where there is Solar equipment and laptops with the Rockwell Logic. The connection to the PLC will vary with the PLC you have as this has changed through the years from PCMCIC cards, KFC modules, PCC cards, 1784 brick to the latest using direct connection to any node using Ethernet cable (not a complete list). The first task is to install RSLinx on your laptop is it is not already there. This is Rockwell OPC Server software and is required regardless of the PLC or type of cable you use to connect.

The next step is to configure RSLinx to communicate with the PLC that you have. If the laptop you are using has already has this connecton because you use it to look at the logic on-line then you can skip this step. If you have PLC 5 with a PCC communications card then the procedure below will help you with this configuration. Otherwise you will have to do some research. If anyone has a different procedure for different PLC, card etc please pass it on and I will share.

 

To configure RSLinx to communicate with PLC 5:

Open up RSLinx and press the “COMMUNICATIONS” key and then select “CONFIGURE DRIVERS”. 

You will see a list of drivers that you currently have. As you can see here there are two drivers, one for the PCC card to communicate with Logix5000 and the other is a Ethernet driver to communicate with Logix5000.

 

When the configure drivers menu comes up press the drop down box to reveal a list of drivers available. The one highlighted is the one needed for the PCMK card Solar uses.

 

Then press the “ADD NEW” button and you will be asked for a name for the new driver.
Leave it at the default if you like.

 

The next menu automatically comes up and it needs some configuration by the user.

 

Select a “STATION NUMBER” that will not conflict with others on the DH+ system.

Now you can see the AB KT-1 DH+ running, so “CLOSE” the box.

Opening up your project file:

Make sure you have only one copy of your project because Logix5 will look in the Project Directory by default.

Select “COMMS” then “GO ONLINE” and this box will appear as it is if the project file is not in the “PROJECT DIRECTORY”.

If your project file is somewhere else select “BROUSE” and select your directory. If there are differences between the project file in the PLC and the version you have you will be asked if you want to MERGE with your file, select this if necessary.

Your project will now open and your communications is working correctly.

 

Next step is to set up a Topic in RSLinx:

With RSLinx open and communicating select DDE/OPC then Topic Configuration.

 

• Select New.
• Type in a name of a new TOPIC (Note down the topic name – no spaces allowed).
• Select Data Collection and change the Polled Messages to 1.

NOTE: Station Octal – If the PLC you are connecting to is daisy chained on the DH+ the station number will probably be 2,3,4 or some other number.
How to check:
Open RSLinx.
Communications
RSWho
Open the DH+ and note the Station Octal (Number).

 

• Make sure the Communications driver is already set up.
• Press Done.

Instruction to use in Excel:

Open an Excel worksheet and type the following into any cell.

=rslinx|xxxxxxx!’F8:0′

Replace the xxxxxxx for the name you selected for the topic name. 

F8:0 was used as an example, pick any PLC address you like.

Save the file and open it up again. This time it will connect to the PLC with the current value for F8:0

Once you are successful with this address you can copy any PLC value to any cell and do what you like with the information. Now it is time to go back to the other article to get some ideas of what you can do with the data. 

]]>
ART049 – Solar – Mars oil pressure regulator issue – PCV901 https://www.dmba5411.com/art049-solar-mars-oil-pressure-regulator-issue-pcv901/?utm_source=rss&utm_medium=rss&utm_campaign=art049-solar-mars-oil-pressure-regulator-issue-pcv901 Wed, 19 Jun 2024 01:29:11 +0000 https://www.dmba5411.com/?p=8125 Continue Reading →]]>

ARTICLE REF – ART049

 

The Mars pressure regulation as used on the generator sets with bias control has a tendency to drift low. This article describes one fix to this problem.

 

 

Figure 1 shows part of the Mars generator lube oil system. The design of the pressure regulation is very bad and it is common for the oil pressure to drop without adjustment while in operation

 

The normal practice is to adjust the oil pressure to 10 PSI while the unit is not running, using the spring adjustment on the regulator valve (in red).

Then when the unit is running the oil pressure is adjusted by means of the hand valve.

Note the solenoid is energized at 30% NGP which dumps some of the pressure opening the valve thereby dropping the oil pressure.

 

This is a drawing of the system before modification.

 

A pressure regulator Solar PN 1066290 was fitted instead of a hand valve and we never had another problem with the oil pressure.


This is the regulator PN 1066290 we fitted instead of the hand valve.

]]>
ART045 – Solar – Centaur 50 compressor surging https://www.dmba5411.com/art045-solar-centaur-50-compressor-surging/?utm_source=rss&utm_medium=rss&utm_campaign=art045-solar-centaur-50-compressor-surging Wed, 19 Jun 2024 01:23:22 +0000 https://www.dmba5411.com/?p=8115 Continue Reading →]]>

ARTICLE REF – ART045

The information in this article is taken from a real event where a Centaur H was surging when accelerating to load. A borescope was taken to site as the reason for surging could be damage to the rotor blades. If it was not caused by internal damage, it is always nice to take a look inside to ensure the surge has not done any damage. If you ever heard a compressor surging you know it is a violent force that can do lots of damage.

With nothing showing up in the borescope the IGV was inspected and the servo actuator feedback turnbuckle was discovered to be loose. The jam nut was backed off, probably by vibration (see diagram below). The IGV angle needed to be checked and adjusted to ensure it was within specification. The turbine was cranked over to develop the hydraulic pressure required to move the IGV. The plan was to command the IGV to various angles and then verify that the IGV actually went to that angel as measured by the IGV tool. The logic was modified to allow control of the IGV signal. The required counts were entered in the control to move the IGV to 12 degrees – it moved the IGV to the full open position. The milliamp value was checked with a multimeter and a reading of 30 milliamps was being displayed.

It was discovered that the signal was coming from a 0 to 50 mA card and the servo actuator was a 4 to 20 mA actuator. The other output used by this card is a 0 to 50 mA output to the fuel actuator which is a 0 to 50 mA actuator. We reconfigured that channel in the PLC to give us a 4 to 20 mA output to the actuator and then calibrated the IGV. We ran the unit and the IGV functioned correctly.

Because of this configuration error the IGV would open up 2.5 times quicker than it should and would not be in the correct position for the speed during the acceleration phase between 80 and 97 % NGP corrected speed. The Centaur turbine is relatively old technology and is not as efficient as the newer turbines such as the T70, Mars etc. Because of this there is a large surge margin and is less likely to surge. A retrofit had been done on several of these units and the wrong configuration was on them all. They had been operating without issue for over a year without incident and if it were not for the linkage loosening they would probably be still the same way.

Item 1 is the servo actuator turnbuckle which was found to be loose.

Item 2 is the servo actuator.

Item 3 is the main actuator.

 

 

]]>
ART042 – Solar – Centaur / Taurus Pressure & temp control block. https://www.dmba5411.com/art042-solar-centaur-taurus-pressure-temp-control-block/?utm_source=rss&utm_medium=rss&utm_campaign=art042-solar-centaur-taurus-pressure-temp-control-block Wed, 19 Jun 2024 01:19:11 +0000 https://www.dmba5411.com/?p=8109 Continue Reading →]]>

ARTICLE REF – ART042

In the graphic below you can see two T60 generator units, one with pressure control issues and one with temperature control issues. This has been an issue with this pressure and temperature control block for many years. Solar released a Service Bulletin 6.0/124 and it is at revision E at least, so that gives some idea of the number of changes there have been. This temperature block was used on Centaur and Taurus generator sets and at one time was mounted at the side of the oil tank, now it is located on top of the tank. The old part number was 190185-xxx and there is at least one newer one part number 1061918-xxx.

If anyone has had issues with this part and would like to share your experience please comment below.

]]>