We are looking to convert our Frame 6 GT from Base load operation to OCGT in the balancing Market. Current start time to base load is ~20 minutes. Has anyone any experience of reducing the start time and any mitigation that can be put in place to retain reliable starting.

 

JW, I have had experience with reducing start time on a frame 5, but not a frame 6. I can tell you that we worked to shorten the time from start to FSNL and the loading rate of the unit from breaker closure to baseload. I am not familiar with what the differences are between a frame 5 and 6.

We had an existing selection on the HMI for a "Fastload" start, not sure if you do. We also use a MKVI control system, not sure what yours is. I can make some suggestions, but not knowing what control system etc. makes it difficult to know what to say.

 

GE has an Opflex offering that provides faster starts I believe it is less than 10min start.

 

We do have a fast start button but it has never been commissioned. At the moment the diesel warms up for 2 minutes, there is a purge for 5 minutes and then it follows a normal start which takes us to 20 mins @ base load. In GE manuals it implies that the normal start of a Frame 6B is 16 mins and a fast start is just short of 10. We would like to get near the 16 minute mark and be at premix steady state.

 

JW - have you contacted GE about fast start options? Fast Start and Fast Load Start are options available on most of GE's gas turbine product line.

You need to understand that fast starts do take more life out of the unit than normal starts.

 

The "fast start" button was never "commissioned". What does that mean? Does it mean it's never been used? Or it was not tried during start-up/commissioning (which is an oversight on someone's part)?

Usually, most units have a FAST <b>LOAD</b> START button, which results in a faster loading from breaker closure to Base Load, sometimes very fast. But, it's really just for emergency situations as it puts a lot of thermal stresses on the hot gas path parts. FAST LOAD START usually didn't change the time to FSNL, only the time from synchronization to Base Load (the loading) was faster.

Some machines were shipped with a FAST START button, which could be programmed to do many different things. In the case of a machine with a diesel starting motor, for example, it could be programmed to reduce the warm-up time (but it wasn't usually eliminated). In some cases warm-up time was decreased slightly and the acceleration rate was also increased. But, this isn't a typical option.

We don't know much about your machine except that it's a Frame 6 with a diesel starter, something we only just recently learned. That information should have been included along with the type of turbine control system in the original post. Along with the fact that your machine has DLN combustors.

There are several companies who would gladly quote a faster start, including GE.

But, because the unit has DLN combustors, there are limits to just how fast the unit can be accelerated, and loaded once synchronized. If you were to buy the option from GE, they would add timers and counters to keep track of the number of these starts, because they must be factored into the calculations for determining maintenance outages. They are not easy on the machine.

You could certainly decrease the diesel warm-up time, but then that decreases the life of the diesel. You haven't said if you will be starting this machine every start in this mode, or just occasionally, or how it might be operated.

And, be careful; I know of several sites that, when the purge time was reviewed, it was determined that the timer was too short for the axial compressor speed during purge, to result in the minimum exchanges (usually three) of the exhaust duct and HRSG and stack volumes. So, you really need to be careful on this particular point alone. Insurance companies and some local regulatory agencies are requiring notification of any changes to such operating parameters. This reminds me of what lawyers always say: Never ask a question you don't already know the answer to.

And, loading DLN machines quickly is touchy. The Premix transfer can be problematic if the loading rate is increased too much, meaning that you'll end up in Extended Lean-Lean and have to reduce load in order to load up to attempt another Premix transfer. Which can mean you would miss your target.

And, we don't know what document you are reading this information about start times from. Is it specifically for units with DLN combustors and diesel starting motors? Or is it for units with conventional combustors and electric starting motors?

To get a concise response, please provide as much information as possible about your site and the unit and its control system. You might work with it every day, but we don't and we can't know very much about your site just from knowing that it's a GE-design heavy duty Frame 6 gas turbine-generator. The fact that the unit has DLN combustors and a diesel starting motor is very important to this discussion. And we're only presuming (because of the length of the purge time) that the unit is being operated in combined cycle mode.

If the control system has a FAST START "button", then it's likely that there is associated sequencing/application code in the turbine control system. It should be possible to review the sequencing/application code to determine what "pushing" that button will do.

 

JW, as CSA mentioned the fact that your units are diesel start and DLN presents some challenges. Again as CSA mentions there are multiple things that can be done to shorten time from start to baseload.

The most common change I have seen is to raise the loading rate. You did not mention what type of control system you have. In my case of a MKVI control system, the TNRV1 block for droop speed control has the ability to select different TNR increment rates for different situations, on of those being a fast load start.

Another place to shave some time from start to FSNL is to raise the accel rate. This typically can't be done until the machine is self sustaining, usually above 60% speed. There is a block called FSRACCV1 that has the ability to select a higher acceleration rate for a fast load start. As CSA mentioned all these things are very hard on the machine, VERY HARD, but if it is truly an emergency then someone has to make the call to "push the button", but they really need to know what is going to happen, why it happens, and how much it costs to do it.

Lastly echoing CSA again the best way to know what might happen if someone selected fast load start is to review the logic. Depending on the type of control system this may take a while to fully see what all is going to happen. Good chance options might be there but are falsed out, or lastly control constants that control TNR increment rate, accel rate, purge, warm-up times might all be set the same as a normal start if the unit was not commissioned for a fast load start.

More information from you might help, but most work will need to be done on your end to understand what the logic will do.