This web page talks about US Navy Submarine Ship Control Stations in general, and the USS Virginia SSN-774 Class Diesel Generator Control Panel, which I designed, in specific. The Navy has generously posted detailed photos of almost everything I want to talk about.
- Compare Ship Control on Los Angeles, Ohio, Seawolf, Virginia Submarines
- Compare DGCP on SSN Virginia, SSN Seawolf, USS Sturgeon
- DGCP on Virginia
Segway – Serial Troubleshooting Tips
Segway – Hard Disk Drives (HDD) and Solid State Drives (SSD)
- Inverted Synchronous AC Generator with Power Rectifier on Virginia
- Virginia Class
1. Compare Ship Control on Los Angeles, Ohio, Seawolf, Virginia Submarines
Before we get into the DGCP, it is interesting to compare the Ship Control Stations on different classes of submarines. I included these photos to demonstrate how the technology on subs has evolved from bulky equipment to streamlined flat panels and joysticks.
Top-To-Bottom = Older design -To- Newer Design
The first (Los Angeles) and last (Virginia) Ship Control Station design have one key thing in common: both have a chair, there’s still a human in the loop. Everything else is designed completely differently.
Pre-Los Angeles Class SSN – I’m not sure what Sub Classes these are.
Los Angeles Class SSN – USS Montpelier SSN-756
Ohio Class SSBN – USS Florida SSGN-728
Ohio Class SSBN – USS Pennsylvania SSBN-735
Seawolf Class SSN 21
Hybrid of traditional levers and modern computer displays
Virginia Class SSN-774
Remember the LCD flat panel you see here, you will see it again on the DGCP.
Virginia Class – New Mexico SSN-779
Virginia Class SSN-774 Trainer
2. Compare DGCP on SSN Virginia, SSN Seawolf, USS Sturgeon
These are shown older-to-newer, The immediate predecessors of the Virginia DGCP are the wide, massive Seawolf DGCP, and the tall, narrow Ohio DGCP, which do not have flat panel displays.
Salmon Class: USS Sturgeon DGCP (from Bubblehead’s blog, see link at bottom).
Los Angeles Class: USS Jefferson City SSN-759 Diesel Control Booth
Left – Virginia: USS Virginia SSN-774 DGCP
Right – Seawolf Class: USS Seawolf SSN-21 DGCP.
Note the 40 to 50 compact, narrow Weschler Bowmar electronic bargraph meters and other gauges.
3. DGCP on Virginia
These photos are dated August 2004 taken aboard Virginia during the pre-commisioning unit (PCU) trials. These U.S. Navy photos were taken by Journalist 1st Class James Pinsky. On his blog, a retired submarine officer nicknamed Bubblehead referred to my design as as “Not your Grandpa’s Diesel Gageboard”, let’s see what he meant:
A picture of the DGCP Training Simulator at the Submarine Learning Center (SLC) and the Naval Submarine School (NSS) in Groton CT is shown.
These are excellent photos of the DGCP. Notice the diesel generator set is running, the selected HMI is the graphical piping “System” mimic, the View Readouts All option is selected, most of the instrumentation appears to be operating properly, and most features on the front of the cabinet are visible. Notice how the engineer uses the trackball to select options on the display panel.
What’s a diesel doing on a modern Navy sub, you may ask. Virginia is of course a nuclear powered vessel, and, like all nuclear subs, has an emergency diesel generator onboard for backup power generation. The design details associated with the diesel systems and electrical power generators are demonstrated by these Navy photos, and by the Technical Manual whose link appears at the bottom of this page (Naval Ship Tech Manual – Electric Power Generators and Conversion Equipment).
I’m thrilled to discover that these detailed photos of my panel are posted, unclassified, public domain (www.navy.mil/privacy.html) on the Internet. I assumed I would never see the final product after I left EB. I wrote this web page for my own pleasure, out of pride for how it worked out, and as a tribute to the challenging and productive years I spent working at Electric Boat. If you are a crew member of a Virginia Class sub, a student of engine mechanics / human-machine interface design / hardware integration, or are just curious about the world around you, I hope you enjoy learning about this little project of mine.
Screen descriptions are given below. It will be interesting reading for those who like to “look under the hood”. It will be helpful if you make a printout of the 3 photos to follow along as you read.
SSN-774 DGCP Screenshot
SSN-774 DGCP Trackball
SSN-774 DGCP on Wikimedia
DGCP video At timestamp 1:32, the DGCP is featured
SSN-774 DGCP Screenshot
We are looking aft starboard. The DGCP is on the port side of the DG set. The DG set is mounted in the submarine “backwards”, with the generator forward of the engine, as shown on the HMI. Facing the DGCP, the generator is to the left of the engine. The DGCP is a refrigerator-size consolidated control and monitoring panel for the emergency diesel generator. The two smaller enclosures to the right are the Voltage Regulator and something else I can’t remember.
The HMI software application is Wonderware FactorySuite 2000 Intouch WindowViewer, running on Windows NT. Wonderware is a very flexible, easy to use, somewhat expensive HMI/SCADA development program. I also considered using ICAS, a, condition based monitoring program (NAVSEA Philadelphia is the ICAS Lifecycle Manager), but Wonderware has a multitude of HMI design tools that ICAS lacks. Caterpillar also makes engine-monitoring software, which can be run on a laptop for troubleshooting and maintenance, but when you use that you only see the sensors that come with the DG set, not the HM&E ships sensors that hook up to the DG set.
The crewman (sitting) is using a ships notebook computer, perhaps running the EngineVision software serially linked to the Electronic Control Module (ECM) on the Caterpillar Series 3512B turbocharged V-12 diesel engine. Virginia is the first Navy submarine to use a commercial diesel engine, instead of a militarized one like a Fairbanks-Morris diesel engine, which would not fit in Virginia’s AMR. Reports are that the crew is very pleased with the rapid-start Caterpillar engine. Look very closely at the bottom of the HMI and you will see “EVIM OK”, indicating that the serial interface to the EngineVision Interface Module is OK. The Intergraph computer communicates using a simple custom message structure over RS-232 serial to the EVIM. The TAL Technologies Winwedge software application moves the incoming serial datastream to Windows DDE memory space, where Wonderware can directly read it.
Segway – Serial Troubleshooting Tips
The EVIM-To-DGCP live serial data streaming into a Windows NT/2000 box demands that special precautions be taken to avoid COM port failure upon bootup.
SYMPTOM: The data from the CAT EVIM isn’t displaying. Stange COM port errors are appearing.
ROOT CAUSE: When the EVIM is turned on and hooked up to the DGCP, it behaves like a firehose of streaming serial data, with no flow control, into the computer’s COM port. This is normal. When NT/2000 boots, any running Serial Mouse Drivers try, and fail, to identify that firehose of serial data, and consequently treats it as a conflict on the serial port, and Disables the port. Ouch. The default Microsoft Serial Mouse Driver (SERMOUSE), or a third-party one like the Logitech Serial Mouse Driver (LSERMOUS), is causing the COM port to be disabled upon boot-up. The DGCP does not have a Logitech mouse, but someone might have inadvertently installed a Logitech Mouse driver on the DGCP computer, without realizing the impact LSERMOUS has on the COM Port. WinWedge FileWedge program reads the COM Port’s data, but it nothing to do with this COM port issue. Took me about 2 months to figure out this pain-in-the-ass problem when it happened, circa August 1998.
(1) The DGCP Install CDROM setup.bat batch file is supposed to prevent the problem, by putting the /NoSerialMice switch in the boot.ini file, but someone might have inadvertently changed it, or it might not work on your PC for some reason.
Check this by doing: Control Panel > Devices > Disable “Sermouse” device.
The /NoSerialMice switch disables Microsoft serial mouse detection (disables SERMOUSE) on all COM ports. It does not disable third-party serial mice drivers like LSERMOUS.
The DGCP Install CDROM has file setup.bat, which replaces the boot.ini file using the command “copy c:dgcpmiscboot.ini c:”, which contains the following line:
multi(0)disk(0)rdisk(0)partition(1)WINNT=”Windows NT Workstation Version 4.00″ /noserialmice
Also try /NoSerialMice:COM1 although you should not need to put :COM1 there according to the Microsoft Support Website.
(2) If you installed a Logitech mouse, like a Trackman Marble, you must disable LSERMOUS too.
The lower case letter L looks like an upper case letter I, so don’t confuse lsermous (that’s an L) with Isermouse (that’s an I). See how fun troubleshooting this stuff is?
(3) The above steps should work for all cases. If they don’t work, use the Brute-Force solution: temporarily unplug or power-down the EVIM, reboot the DGCP computer, and once it’s fully running, reconnect the EVIM.
From the top left, there is a Barco ruggedized 20″ RFD-251S Liquid Crystal Display (LCD), and the Dial Telephone. The grab handles on the Barco LCD make it look a little bit like a microwave oven. Compare the full-sized photos of the DGCP with the SCS and you will see that they use the same Barco LCD. The LCD on the SCS is rotated 90° because its designers chose to use the “portrait” orientation. The dial phone has a jack for a boom-mike headset instead of having a built-in handset.
In May 2010 I toured the USS Jefferson City SSN-759, a Los Angeles-class submarine, home ported in my home town of San Diego CA. Its ship control area has been modernized with multiple Barco RFD 251’s on many enclosures and overheads.
Below, sound powered telephone jacks JA and 2JV, speaker cutout switch, loudspeaker.
Below, collision alarm (red), voltage regulator control transfer switch. If you’re wondering why I put the voltage regulator remote controls on the DGCP when the voltage regulator itself is right next to it, it is because the local controls inside the voltage regulator are very small and not readily accessible; that was never intended to be the place where the operator controlled voltage. The DGCP is designed to be the remote-manual control for it.
Right, behind a protective Plexiglas shield with finger holes to help prevent inadvertent operation, the DG Set control and indication panel consisting of Staco pushbuttons and indicator lamps. This part of the DGCP is similar to the features on the electric plant control panel (EPCP) in the aft end. From here the engineer can do things like start and stop the DG, adjust its output voltage, and control the space and jacket water heaters.
Right, the recess is for the Q70 style trackball. The Barco LCD is not a touchscreen. I avoided a touchscreen because when you are working in the AMR your hands may get dirty and that would make the screen look messy.
Below, a pull-out keyboard tray (shown stowed). Only needed for maintenance and configuration.
Below, a rackmount Intergraph TDR-3000 ruggedized computer. Look very closely and you will see the oval Intergraph logo. I leveraged off the use of Intergraph ruggedized PC’s on Smartship.
Segway – Hard Disk Drives (HDD) and Solid State Drives (SSD)
Due to concerns about meeting shock requirements, later DGCP’s are furnished with SSD’s (Flash Memory, no moving parts) instead of conventional HDD’s (spinning platters).
SSD’s are a clever idea, have clear advantages over hard drives, and are probably the wave of the future. But I don’t like them.
Wikipedia SSD article, ” Flash-memory drives have limited lifetimes and will often wear out after 1,000,000 to 2,000,000 write cycles … As a result of wear leveling and write combining, the performance of SSDs degrades with use.”
I’ve used a dozen different solid-state Flash Drive / Jump Drive / Thumb Drive / keychain USB devices, and they all died after 6 to 9 months after frequent daily use.
SSD’s should not be used in computers that serve in very long-term, high-reliability, industrial applications. Sound familiar?
Mitigation of drive failure, should be sparing a Removable Pre-Imaged HDD, instead of using an SSD.
Below (not show) a small power distribution panel, and a hinged front door for two shelves of Opto22 SNAP IO B3000-ENET data acquisition modules. Look very closely at the bottom of the HMI and you will see “Top B3000 OK” and “Bottom B3000 OK”. The Intergraph computer communicates using Modbus over Ethernet to the B3000’s. Wonderware comes with a device driver for Modbus over Ethernet. My design decisions were happening at the same time that the HMI/SCADA industry was rapidly transitioning from proprietary interface protocols to open standards using TCP/IP. If I were doing this design today, I would probably use Profibus-DP or Profinet.
Inside, power supplies, relay boards, a hub, and miscellaneous hardware.
The HMI shows the EDG running. All of the following values are displayed. Engine speed is about 1800 RPM. Engine cylinder temperatures are around 1000 F, and exhaust air is 157 F after being cooled by the seawater mist injection. The ambient compartment pressure is 0.2 inches of mercury vacuum because the engine is drawing ambient air. All three exhaust valves (isolation, backup and hull), all three induction valves (backup, hull, and head), and all three seawater valves are open, of course.
There are 16937 gallons of fuel remaining in the normal fuel oil tank. You may learn more about other important benefits of some tanks by reading here.
You can trace the path of heat flow through the coolers and heat exchanges on the lube oil (yellow), fuel oil (yellow), ventilation/exhaust (tan) , freshwater (blue), and seawater (green) systems to see how much heat is added or removed by each equipment. The HMI color code matches with valve handwheel paint color scheme. The HMI does not show every valve and equipment in these piping systems; only the parts of the system that are instrumented are shown.
There are tiny nicks and scrapes on the Barco handles, which give you an idea how rough this environment is.
Note the engineer reaches into the recess to move the trackball for the LCD. In this photo the display is in “datasheet” view. This view contains the same data as the “system” piping view above, in the format of an instrument list instead of the format of a piping mimic. The operator can use the display format he is most comfortable with.
The DGCPs (or equivalent equipment) on previous submarine classes, such as Los Angeles (SSN-688, about 60 boats), Ohio (SSBN-726, about 18 boats), and Seawolf (SSN, 3 boats), have individual vertical bargraph meters and dials to provide indication and monitoring of instrumentation. On Virginia, the DGCP design goes high-tech with the use of a flat panel display, providing virtually all indication and monitoring (not control). The HMI may look “busy” at first, but the operator can simplify it by selecting different options for which readouts are displayed and which readouts are hidden. I’ll be honest with you, this flat panel display is a glorified bargraph meter. Count the quantity of numeric readouts and valve-position icons, it comes out to around 80 plus admin-buttons. It would have taken up a lot more space than an LCD to foundate 80 bargraph meters and what-not on a steel cabinet. That’s why Seawolf’s DGCP is so huge.
Something operators like about bargraph meters is that they can make quick status assessments by merely glancing at the heights of the bargraphs, without considering the numeric values. The HMI accomplishes the same thing by using dynamic font color coding (green Good, yellow Warning, red Bad). The color coding matters when the engine is running. For most readouts, the color coding does not mean anything when the enging is not running.
This HMI describes all major elements of the auxiliary systems that interface with the DG Set.
The NFO tank is down to 16690 gallons, so this photo was taken after the engine had been running for awhile (see previous picture), and the cylinders have cooled down.
There is an option button labeled Instrument Numbers that displays the instrument number of each readout in a small font. This option is shown turned off (otherwise the HMI would look even busier). The operator knows what a readout is for by noting its location within the context of the piping mimic.
The diesel seawater (DSW) pump and diesel freshwater (DFW) pump icons represent fluid centrifugal pumps. The fuel oil (FO) pump and fuel oil transfer (FO Xfer) pump icons represent axial pumps.
The bottom right hand corner shown that the engine shut down due to high exhaust backpressure, which means that the one of the exhaust valves was shut, which tripped a pressure switch sensing exhaust gas pressure (IC Circuit 2SN).
The 8H switchboard breaker is shown open (the bar is not aligned with the bus).
The DSW pump valve DSW-1 valve is shown closed (the bar is not aligned with the pipe).
It is interesting to point out an “open” electrical breaker looks same as a “closed” piping valve. The graphical representation of an electrical single-pole switch, when drawn “open”, represents the absence (opening) of a continuity-path for the flow of electronics.
The graphical representation of a valve, when drawn “open”, represents the absence (opening) of a restriction-path to the flow of a liquid or gas. I like to quiz mechanical engineers by asking them which word they use to describe the 8H breaker symbol, shown – they usually say closed 🙂
Along the top, the long flat rectangle, filled with diagonal hatch marks, represents the pressure hull. Seawater (green), fresh induction air (tan), and exhaust air (tan) penetrates the pressure hull boundary. The snorkel head valve VH-1, shown outside the pressure hull, is the flapper that uses water-sensing electrodes to automatically slam shut when waves lap over the hull while running the engine. In a high sea state this valve will open and close quickly and frequently. If excessive wave action causes it to close for a long time, the ambient compartment pressure switches will sense a low vacuum pressure and automatically shut down the engine. Crewman will probably feel their ears pop. If the head valve electrodes sense water, and the head valve does not successfully close within a certain time period (maybe a whale is stuck in the poppet valve), the induction hull and backup valves automatically close to prevent flooding (IC Circuit 1SN).
These fancy new flat-panel-display based control-and-indication screens on Virginia (Ship Control, EPCP, DGCP, others) went through extensive HMI Fleet Reviews. I remember presenting this design to a crowded room full of NAVSEA reps and Chief Engineers in February 1998. We captured hundreds of comments and incorporated dozens of suggestions to arrive at this design. The important thing to note is that smart, robust HMI design is being taken very seriously as it’s used more and more. Years ago an HMI requirements spec often said “the HMI must have an easy to use look-and-feel”. Today there is a large body of knowledge of what constitutes good HMI design.
Do the pipes animate when fluid is flowing through them? No they don’t. If it had been truly important to visually confirm process flow, the Navy would have directed the Shipyard to install flow sensors in the pipes, which the DGCP Screen could have used as a trigger to activate pipe animation graphics and display flow rates. To animate pipes any other way (e.g. deducation based on valve lineups) is complicated and unreliable and should not be done.
4. Inverted Synchronous AC Generator with Power Rectifier on Virginia
I’m an electrical guy, so this page wouldn’t be complete without a discussion about what the diesel is actually on the ship for, to drive the generator. Virginia has a very compact custom generator.
- Diesel Engine. The Caterpillar 3512B is the standard commercial-off-the-shelf (COTS) diesel (that’s one mighty big shelf!)
- Synchronous AC Generator
Output voltage is 700 volts DC and output power is 921 kilowatts. Virginia is the first submarine to have a DC voltage bus. Why does the EDG set output 700 DC, you may ask. Because Virginia has a 700 VDC distribution system (no frequency-sync of power sources needed). There are lots of invertors throughout the ship to convert to whatever power is needed. The DDG-1000 Zumwalt Destroyer also has a DC bus. Unlike SSN Virginia, DDG Zumwalt is a true Integrated Fight Through Power IFTP system, because it is electric-drive, which means the propeller shaft is turned by an electric motor.
There is currently no American electric-drive nuclear-powered submarine. There was a prototype, the USS Glenard P. Lipscomb SSN-685, and another before that called the USS Tullibee SSN-597.
The electric-drive SSN of the future will be stealthier than today’s mechanical-shaft-driven subs, because electric-drive physically decouples the noisy nuclear steam plant from the propeller shaft, reducing signature.
The generator that is attached to the engine is custom. The generator is an inverted synchronous AC generator with a built-in power rectifier. The term inverted is used because the Main Stator Coil is the Armature. The rotating magnetic field is in the Main Stator, also called inside-out. The term synchronous AC generator is used because the following equality applies: Calculate the rotational speed, where the Main Rotor has P = 6 (pairs of) poles, and runs at 90 Hz.
Synchronous speed NS = 1800 RPM. Number of phases q = 6 x 2 = 12.
The output frequency can be increased by increasing the RPM’s, or add more poles.
Rotors. There are two types of rotors used in synchronous generators.
- Cylindrical. For high-speed, 2 or 4 pole, quiet, balanced, low winding losses.
- Salient-pole. For low and medium speed. Have outward projecting laminated poles.
This generator has salient-pole rotors. The rotor & stator have the same number of poles. The Rotor winding carries DC, to produce constant flux per pole.
Construction. This generator is like three generators in one:
The Diesel Engine turns the shaft. The shaft is part of the rotor inside the generator housing.
Generator Housing contains rotor and stator. Rotor windings include PMA rotor, exciter rotor, and main rotor. Rotor windings are surrounded by stator housing. Stator windings include PMA stator, exciter stator, and main stator. I have not found a concise description of this type of generator on the Internet, so I’ll describe it below, in sequence.
Stationary armature, rotating field magnets. The purpose of the Permanent Magnet Alternator is to create a rotating magnetic field to generate enough power to flash the exciter field. PMA rotor (12-pole Lundell) turns, sets up rotating magnetic field (the permanent magnets rotate). PMA field magnets rotate inside stationary armature, interaction induces EMF in PMA stator. PMA stator winding 3 phase 180Hz 150VAC output is hooked up to Voltage Regulator input. Voltage Regulator regulates and rectifies power. Voltage Regulator AC output is rectified and the DC output is hooked up to exciter stator windings. The magnets inside a PMA are extremely powerful, and dangerous if you have to handle them.
Stationary field winding, rotating armature. The purpose of the Exciter is to provide the Main Salient Poles with DC power to create a rotating magnetic field. Exciter stator winding is powered with DC, sets up a stationary magnetic field. Exciter armature rotates inside stationary field winding, interaction induces EMF in Exciter rotor. Exciter rotor (16 pole Exciter Field Winding armature) 3 phase 240Hz AC output is hooked to a full-wave bridge rectifier. The rectifier is also located on the rotor. Rectifier DC output is hooked up to main rotor.
Stationary armature, rotating field winding. Main rotor (the Field Winding consists of 6 Main Salient Poles) turns at 1800RPM 90Hz, sets up a rotating DC magnetic field. Main field winding rotates inside stationary armature, interaction induces EMF in Main stator. Main stator (4 of 3 Phase Windings in Main Stator Coil Armature) AC output is hooked up to Power Rectifier. Power Rectifier outputs 700 VDC which is delivered to the ship’s electric plant.
PMA Rotor > PMA Stator > VR >
Exciter Stator > Exciter Rotor > Rectifier >
Main Rotor > Main Stator > Rectifier > Ship’s DC buss.
Slip. If the rotor is revolving at exactly the same speed as the magnetic field, no currents will be induced in it, and hence the rotor should not turn at a synchronous speed. In operation the speeds of rotation of the rotor and the field differ by about 2 to 5 percent. This speed difference is known as slip.
- www.navy.mil select Photos and search for V-12 DIESEL ENGINE.
- NavSource Online Virginia (SSN-774) Shakedown Cruise (same photos as above)
- Below Decks Pictures Of USS Virginia, dated October 30 2005, on blog The_stupid_shall_be_punished
- Technical Description of USS Virginia Ship Control Station. The fly-by-wire SCS on Virginia is completely different from all previous submarines.
- Article about the Virginia Ship Control Trainer in the New London Day
- Naval Ship Tech Manual – Electric Power Generators and Conversion Equipment
- Armature, on Wikipedia
- White Paper on Synchronous Generators
- DGCP Training Simulator
6. Virginia Class
|SSN-774||Virginia||Groton, CT, Squadron 4||38283||Block 1|
|SSN-775||Texas||Pearl Harbor, Hawaii, Squadron 1||38969||Block 1|
|SSN-776||Hawaii||Pearl Harbor, Hawaii, Squadron 1||39207||Block 1|
|SSN-777||North Carolina||Pearl Harbor, Hawaii, Squadron 3||39571||Block 1|
|SSN-778||New Hampshire||Groton, CT, Squadron 4||39746||Block 2|
|SSN-779||New Mexico||Groton, CT, Squadron 4||40138||Block 2|
|SSN-780||Missouri||Groton, CT, Squadron 4||40390||Block 2|
|SSN-781||California||Groton, CT, Squadron 4||40845||Block 2|
|SSN-782||Mississippi||Pearl Harbor, Hawaii, Squadron 3||40880||Block 2|
|SSN-783||Minnesota||Groton, CT, Submarine Group Two||41524||Block 2|
|SSN-784||North Dakota||Groton, CT, Submarine Group Two||41937||Block 3|
|SSN-785||John Warner||Naval Station Norfolk, Old Dominion||42217||Block 3 - Under Construction|
|SSN-786||Illinois||42583||Block 3 - Ordered|
|SSN-787||Washington||42736||Block 3 - Ordered|
|SSN-788||Colorado||42736||Block 3 - Ordered|
|SSN-789||Indiana||Block 3 - Ordered|
|SSN-790||South Dakota||Block 3 - Ordered|
|SSN-791||Delaware||Block 3 - Ordered|
|SSN-795||Hyman G. Rickover||Block 4|
|SSN-796||New Jersey||Block 4|