United Refining Company (URC) / Boiler 4 FD Fan Turbine / [Issues & Incidents]

Ramp Rate on Auto Start
Updated 2/13/09

To achieve a faster ramp rate to speed on auto start, the configuration worksheet was updated to have the Peak 150 read Fan Shaft Speed rather than turbine speed .  Since the gearbox reduction is almost 3 to 1, the ramp rates realized will also be 3 times their previous maximum configurable value.  This option is easily implemented and will avoid confusion because the equipment speed is normally displayed as Fan Speed throughout the system.  With the numbers used, the Peak 150 should be able to raise it's setpoint at almost 300 RPM per second (turbine speed) during a "Remote Not Matched" state.

The overall objective is to avoid a boiler flameout during fan drive transition.
The limited stated objective of the Peak 150 is to raise the speed setpoint to maximum speed in less than 25 seconds.  Included in the 25 second time window is the latency for logic sequence and steam to the turbine.  The maximum speed is now about 1750, so the Peak 150 can accelerate from 350 Min Gov (fan speed) to 1750 in about 14 seconds.  The latency issues account for about 6 seconds, so we have about 5 seconds left over.  It should be noted that the actual acceleration rate may be limited by the turbine horsepower.  The turbine will accelerate quickly if it's speed is below the fan speed, but once the turbine speed matches the fan speed the over-running clutch will load the turbine and the acceleration will not be as fast.

Updated 2/6/09

Notes from phone conference between Mike McCluskey, Tom Truax, & Hubert Hoskins on Friday, 2/6/09

The ramp rate appears to continue to be an issue, especially with the boiler operating at high load and EGR active.  There had been weekly auto start test for a couple of months with incident, but only two test with the EGR active.  The first test with the EGR active was at low load.  The second test was at high load and the boiler flamed out and did not recover.

Several possible configuration solutions were proposed to address the objective of avoiding the remote not matched ramp and getting the turbine up to speed quicker, including:

• Setting the minimum governor speed to a higher value.  The reported lower speed range is about 2700 RPM, but the min governor speed configured is 500 rpm
  • Mike reports that even though they don't normally run down at 500 RPM, the do need to control at that speed via the DCS because it is part of the boiler start permissive purge cycle.
• Utilizing the idle function and setting the rated target speed to a high value
  • This idea would need testing to see evaluate the behavior.
• Having the DCS call for a lower speed to match the remote signal, and once matched, having the DCS then ramp the setpoint up at a fast rate not limited by the Remote Not Matched Rate.
• There was some mention of utilizing a 505, but that idea proposed by Woodward seemed to not be well receive by the customer.
• The customer also put forth the idea of scraping the Peak 150 and using a PLC.  Tom noted that although that option may be possible, it would not be cost effective and support would be limited.

Ran out of time on the conference.  Plan to re-conference again on Monday at 7:30 EST.

Updated 11/6/08

The issue below will not be addressed further.  After auto start testing during commissioning, URC determined that the auto start acceleration rate was adequate for operation of the boiler.

Initial Posting Updated 10/30/08

There is an issue with the ramp rate to speed on auto start.

The objective is to ramp the Peak 150 to match the fan speed control signal from the Δ V DCS.  If the time to speed is too slow, the concern is the burner may flame out.  After the initial posting of this issue on 9/24, the auto-start transfer performed adequately at low boiler loads during testing on 9/25 by switching the VDD off, so we may not need to take further action.

The TG-13 governor started with the governor steam valve wide open.  The upstream shutoff valve controlled is controlled by a 50" Valtek actuator that is not modulated other than full on or full off.  On loss of the electric motor, the upstream Valtek shutoff valve would transition from full closed to full open and the steam would be admitted to the turbine through the wide open governor valve.  There are varying reports on how long it took to accelerate, but the reports varied from 5 seconds to 14 seconds.

There are some variables related to turbine acceleration, one of which is weather it is loaded or not.  If the fan is coasting down, the turbine will accelerate unloaded until it's speed reaches the fan speed and the overrunning clutch engages.  Acceleration with the fan fully stopped will be slower because there is more mass to accelerate.  Acceleration rates will vary depending on the fan speed when the motor quits.  Prior to the FVD, the motor was always running full speed, but with the VFD, the motor speed will be variable.

It takes the fan about 20 seconds to decelerate from 1750 (about 5200 Turbine Speed) to about 670 RPM (about 2000 RPM turbine speed).  It takes about 30 seconds to decelerate from 2000 rpm turbine speed to about 1000 rpm turbine speed.

The Peak 150 governor has a limitation on how fast it can accelerate to the match it's speed setpoint with the control signal.  This is referred to as the "Remote Not Matched Rate" in the configuration worksheet.  The maximum value the control will accept is 100 RPM / second.  We would prefer a faster acceleration rate.  There may be some other options to work around the arbitrary limitation, but it would be preferable to keep the logic simple.  Woodward has been contacted as noted below to request permission to change the limitation to a higher value.

The email request to Woodward is noted in the email chain below.

-----Original Message-----
From: Tim Hinde [mailto:Tim.Hinde@woodward.com]
Sent: Monday, October 27, 2008 11:21 AM
To: Tom Truax; Alan Oetken
Cc: hhoskins@peaker.com; Eric Feiler; hhurley@peaker.com
Subject: RE: Peak 150 Remote Not Matched Rate Limitation

Tom;

The answer is going to be no. It takes time and resources to change an
existing product. The GAP would have to be changed and tested, the
manual would have to be changed, this all takes resources that we don't
have available.

Tim Hinde
Woodward Senior Technical Support Engineer
Phone: 970-498-3525
FAX: 970-962-7541
Email: tim.hinde@woodward.com

-----Original Message-----
From: Tom Truax [mailto:ttruax@peaker.com]
Sent: Friday, October 24, 2008 9:46 PM
To: 'Tim Hinde (thinde@woodward.com)'; 'Alan Oetken (aoetke@woodward.com)'
Cc: 'hhoskins@peaker.com'; 'Eric Feiler'; 'hhurley@peaker.com'
Subject: Peak 150 Remote Not Matched Rate Limitation

Tim & Alan,

What is the reasoning for limiting the Peak 150 Remote Not Matched Rate to 100 RPM per second? Would Woodward be willing to give us permission to modify the rate in the GAP program to a higher value?

It seems common that the Peak 150 would be used in a back up standby role where the customer wants to accelerate quickly to some variable speed if the primary drive source fails. In our application at URC the Peak 150 replaced a TG-13 governor. The TG 13 governor was wide open on start and the turbine could accelerate to max governor speed (5200 RPM) in 10 to 15 seconds. They have removed the fan louvers and now need to target whatever speed the new FVD driven motor is running at to achieve their boiler output demand.

I recognize that there are a couple options to work around the issue, including having the customer lower the his remote speed setpoint and then moving it back to where his PID loop wants it after a confirmation that the Peak 150 has achieved a match, but that is excessively clunky and the customer is resistant to the programming hoops on his end.

We just receive a PO for another stand by application (Sunoco Philadelphia Refinery CO Boiler FD fan) that will be directly on point with the current back up standby application at URC. Neither customer wants to slow roll their small turbines due to their experience with negative impacts resulting from extended slow roll. We are building a strong relationship with key personnel at Sunoco. They are enthused about the Peak 150 and we expect them to purchase significant volume over the next few years.

Sincerely, Tom Truax, PSI

-----Original Message-----
From: Tom Truax [mailto:ttruax@peaker.com]
Sent: Thursday, October 23, 2008 4:06 PM
To: 'Dave Quigley (Dave.Quigley@woodward.com)'; 'Tim Hinde (thinde@woodward.com)'; 'Alan Oetken (aoetke@woodward.com)'
Cc: 'hhoskins@peaker.com'
Subject: Peak 150 Ramp Issue at URC Resolved

Tim & Alan,

Thank you for your effective support. We removed the wire from the idle rated input and the unit operated as expected. We then checked the Idle Rated settings, and they were as you suspected with the target speed set for 1000 RPM and the ramp rate at the default of 5 RPM per second. Apparently, the control will still execute the idle rated function even if it is not configured to do so.

Their system works great. Speed control is very tight across the whole span and the actuator is rock steady. The customer is very pleased.

:) Tom Truax

-----Original Message-----
From: Tom Truax [mailto:ttruax@peaker.com]
Sent: Thursday, October 23, 2008 1:37 PM
To: 'Dave Quigley (Dave.Quigley@woodward.com)'
Cc: 'hhoskins@peaker.com'
Subject: Urgent Peak 150 Support Needed

Dave,

I am at United Refining Company in PA. We are commissioning a Peak 150 on a small steam turbine that is driving the Force Draft Fan for the most critical boiler in the refinery. The other side of the fan has an electric motor controlled by a VFD. The intent is to have the turbine backing up the fan in stand by mode ready to start. We need the turbine to ramp up to speed quickly upon command if the VFD motor drops out. They are sending the motor and the fan the same remote speed setpoint signal value. If the motor drops out, we auto start the turbine and need it to quickly ramp up to the speed setpoint requested by the DCS.

I've attached the configuration worksheet for the Peak 150. We are not using idle. The speed range is 500 RPM to 5200 RPM. Remote Control is selected prior to turbine start. My understanding is that the turbine should start and ramp toward the remote speed setpoint at the Remote Not Matched Rate. We have the remote not matched rate set to the maximum the control will accept, which is 100 RPM per second.

Our trouble is that upon start, the turbine speed setpoint goes to 500 rpm, and then ramps up at 5 RPM per second until it gets to 1000 rpm, it then accelerates to the remote speed setpoint at the remote not matched rate of 100 rpm per second. It is taking 100 seconds to go from 500 RPM to 1000 rpm. Based on my reading of the literature, it should move at the remote not matched rate and get to 1000 rpm from 500 rpm in 5 seconds. 100 seconds is too long and may result in a boiler trip.

We are commissioning the new emission control system for the boiler, so we have a full crew on site. Any assistance you can provide will be appreciated.

Sincerely, Tom Truax PSI, (805) 680-7727 cell