Hello Everyone,

I am a new controls and instrumentation engineer and I've been tasked with Writing a Spec for a controls upgrade on our LM6000 Units. The current control system is a GE FANUC 9070 system.

I wanted to reach out to everyone here and see what are somethings that you would never leave out of a LM6000 Controls Upgrade. Ie, Alarms, Fuel Control , equipment upgrades, Critical Function Redundancy, hot standby plc system design?

Any information is greatly appreciated.

Thank you.

 

Many of these older LM6000 systems were a split architecture controls, the GE 9070 sequencer with separate Woodward Micronet or GE Mark VI purpose designed fuel controller "PLC". A redundant Ethernet link glued them back together.
It is better to combine all the functionality into one control system as it is easier to troubleshoot and common hardware.

The speed sensing and servo control (IGV, VSV, VBV, TBV) IO is very important. Most PLCs are not really designed to handle these signals. A turbine controller is designed for this type of IO, MPUs, Prox, bipolar current drive, and LVDT feedback. It is best to stay away from external converters for these to improve response, accuracy, calibration, service, alarming, and troubleshooting. You may run into this with many PLCs as they do not inherently have this type of IO. The Woodward control will also include automatic servo calibration with sanity checks on voltage ranges and polarity. This is a nice feature.

Keep an eye on the PLC hardware quality. Good isolation and differential inputs/outputs are best for accuracy, noise immunity, and better protection from ground faults.

The Mark VI and MicroNet are very similar in architecture so upgrading to a full Woodward Micronet+ is pretty straight forward and has been done many times on these systems. If it is Woodward today all the logic ports over directly. Woodward has executed many of these systems to date on retrofits and new units. A Mark VIe is a different architecture and I do not have any experience on how the logic ports or how difficult this is to upgrade.

The Woodard or Mark VIe options, both are qualified control systems, meaning GE aviation has reviewed, approved, and vector tested the hardware and software. An Aero-turbine like this has some special control performance requirements such that you stay away from compressor stall and can handle heavy load transients for a few examples. General PLCs will not be qualified by GE aviation to meet all the filtering, timing, accuracy, and response requirements.

Going with the Woodward or Mark will also allow you to get any fuel core/engine upgrades you may want or need, Such as: flameout algorithm that remove the UV detectors for improved reliability, thrust balance valve removal, variable IGVs, Sprint power enhancements, fast start, MW bump, Overfire, reserve power control, enhanced variable geometry servo and feedback fault detection, model based backup sensor fault detection, Optimized variable geometry schedules, Purge credits, Variable/Sprint Nox credits, etc.

The Woodward controls now integrate the HMI and logic together so you can branch from the screens or an alarm directly into the logic to troubleshoot or look at functionality.

Woodward can integrate the rest of the plant into one common hardware if for example that is a need, say boiler control, water treatment, etc. can all be done in the Woodward platform.

The Woodward system offers local and distributed IO. The distributed IO is rugged high temp/high vibe so it can be placed in a junction box with no worry.

Most PLC remote IO suffer from reliability if not air conditioned. The new Woodward remote IO does not require this as it is rated to 100 Deg C continuous operation.

Alarm history and historical data is important. Integration to plant systems with OPC or Modbus is a must. 10 msec alarm time stamping for sequence of events with clock synchronization is good to have. Being able to Search the alarm history is also very beneficial. High speed 5 to 10 msec continuous data logging is also very important to provide data to experts to help analyze any issues that may come up. Trending and plotting data is also key, so good tools are a must. Redundancy in power, CPU, and IO is important for maximum available and minimizes spares to keep in stock.

Addressing the fuel system and electric fuel/water metering valve drivers is an important upgrade to avoid obsolescence. Woodward has also just released our latest GS50 gas fuel metering valve technology with smart metering software technology to improve accuracy over changing fuel conditions. This valve is designed for extreme duty situations, single shaft eliminating a gear train, with high torque, low leakage, self-cleaning, on-board driver with ethernet remote access.

Also consider any other equipment in the panel that maybe obsolete, power meter, vibration, etc. If you are considering also to upgrade the generator exciter, staying with the same model is the smoothest upgrade in my experience. If you have a Brush generator switching to a Brush exciter I have seen also executed well by Brush when migrating from a GE ex2100 exciter.

Some have considered a separate or new cabinet, but the time spent on re-wiring and re-testing is a major risk, so an in-place replacement is most of the time better.

You can also faze in upgrades and tailor it to any budget. Say for example control system now, fuel valves next year, voltage regulator the following etc.

Get some feedback from other users who have upgraded. You can try the WTUI forum or other user groups or plants that have done these upgrades already.

In the end I would say the MOST IMPORTANT thing is experience, knowledge, and support so that the outage time is seamless and short.

A good Factory Acceptance Test and plan is important so that you are best prepared for the field install.

I have seen upgrades where the logic is just duplicated. In some cases, the engineers know how to program but lack the knowledge of the system and the turbine. When there is a problem most often it is not the hardware, but analyzing and understanding the turbine data is critical to helping identify the real issue. Tools and people with experience to analyze this data or provide this data to the OEM/Control supplier is important.

I have also seen upgrades where the logic is all new and you are debugging quite a bit because the simulation and testing is not a strong suit of most "PLCs". Woodward has a full PC based seamless, closed loop, simulation environment using physics-based models to prove all the logic rigorously tested prior to deployment. It is also used for training and debugging field problems. I like to avoid hardware simulators anymore as they are typically more open loop testing or modeling "short cuts" taken. A simulation model can be good enough to tune PIDs, take it to the field, and not touch it. Woodward has this capability.

If outage time is critical Woodward has done these types of upgrades in 3 to 5 days depending on scope and other activities even with mechanical fuel valve updates at times.

 

Email me at [email protected] I can get you a general spec and budget pricing.