 |
|
The Army must upgrade legacy equipment so it can be maintained quickly and efficiently by Force XXI multicapable mechanics. Embedded diagnostics will reduce troubleshooting and allow mechanics to repair more equipment in less time.
During the past year or so, the Project Manager for the Abrams Tank System (PM Abrams) has explored ways to extend the life of the current M1A1 tank fleet to the year 2025. At that time, M1A1's will comprise 80 percent of the Army's tank fleet. However, the costs of converting the total M1A1 fleet to M1A2's and M1A2 System Enhancement Program (SEP) tanks are prohibitive. The Army's challenge is to find a cost-effective approach to improve the supportability of M1A1 tanks. At the same time, the Army plans to combine the current mission requirements of organizational and direct support mechanics by assigning them additional maintenance tasks. To make the mechanics more effective (multicapable), the Army must provide them with the means to improve diagnostic efficiency. In line with this need, plans for maintaining the Abrams tank fleet in the future identify a need to incorporate built-in-test (BIT) and fault-isolation test (FIT) capabilities in the Army's aging tank fleet.
To reduce the operating and support costs of the M1A1 fleet, PM Abrams proposed a recapitalization plan that includes—
This article addresses the last point of the plan.
Electronics Obsolescence
The Abrams tank fleet is experiencing the same problems with electronics obsolescence as are other weapon systems. In 1986, a weapon system with military specification (MIL-SPEC) components could expect to undergo one modernization during its 30-year life. Today, for MIL-SPEC components, two modernizations can be expected during the 30-year life of a weapon system. This assumes the Government is willing to pay a premium price for outdated technology. Because of the expense of MIL-SPEC components, the Government is sure to purchase more and more commercial devices. Seven modernizations can be expected during the projected 30-year life of a weapon system that has commercial components.
The ever-accelerating pace of technology development is causing these changes to come faster and faster. Before 1990, complementary metal oxide semiconductor (CMOS) devices operated on 5 volts of electricity. (A CMOS is used when storing permanent instructions in a computer. It usually controls a repetitive action, such as startup.) By 2005, it is projected that the 5-volt devices will have disappeared from the commercial marketplace. Civilian applications are transitioning to 3-volt CMOS devices, and development of 1.5-volt devices is sure to follow. This means that all weapon system technology that uses 5-volt devices will become military unique and obsolete. Even in state-of-the-art programs, such as the M1A2 SEP, individual components have become obsolete. Either the manufacturer no longer builds a particular device, or the device is available only until current stock is exhausted. For legacy systems, such as the M1A1, the problem is much more difficult, because the electronics technology in the tank is over 20 years old.
PM Abrams approaches this problem in two ways: piece-part level (short term [reactive]) and circuit card assembly level (longer term [proactive]). The piece-part approach provides needed components, including "last-chance buys," qualified replacement parts, and third-party manufactured and recycled parts generated from upgrade programs. The short-term objective is to ensure the availability of spare parts.
The circuit card assembly approach begins with a feasibility study of a particular circuit card assembly to determine if redesign is more cost effective than solving component problems one by one. The intent is to extend the life of the circuit card assembly 2 to 5 years beyond what replacement of individual parts could achieve. The long-term goal is to maintain a producible technical data package for the next production contract. Based on this approach, PM Abrams is redesigning the M1A1 turret and hull network boxes (TNB's and HNB's). The design effort for the TNB is nearing completion, with the HNB to follow. This redesign effort will insert a BIT capability in the M1A1 turret and hull that will identify faulty line replaceable units (LRU's).
Abrams On-Board Diagnostics
M1A1 and earlier Abrams tanks have very little on-board diagnostics. The only subsystem with a true BIT that can identify a faulty LRU is the Thermal Imaging System. However, the turret does have an auto self-test that is initiated each time the turret power is turned on. The self-test shows if a component is responding incorrectly or if a cable is disconnected. It alerts the crew by turning on the fire control malfunction light on the tank commander's control panel and displaying an "F" symbol (for fault) to the right of the range symbology in the gunner's primary sight. It does not identify the component or cable that caused the malfunction, which is similar to the way an "idiot" light in an automobile works.
The gunner can run a manually initiated computer self-test of selected parameters. Based on a pre-programmed set of parameters, the computer self-test checks the laser rangefinder, cant (angle), crosswind, super-elevation, and lead. Should any of the checked items fail, a corresponding code number appears in the display window on the computer control panel. The crew can perform limited troubleshooting procedures, such as checking to ensure that the crosswind sensor is clean. However, to identify failed components, mechanics use either the simplified test equipment M1/fighting vehicle system (STE-M1/FVS) or a breakout box and multimeter. The STE-M1/FVS is hard to use, takes excessive time to set up, and is physically bulky. Using the breakout box with a multimeter requires an intimate knowledge of the electrical wiring harnesses, wiring diagrams, and functional logic.
The M1A2 and later tanks are equipped with self-test, BIT, and FIT. The crew can troubleshoot the system with BIT and report the system fault. The mechanic reruns the BIT to confirm the fault and uses FIT to isolate the actual fault in the LRU. The diagnostics set-up time is reduced to zero.
Upgrading M1A1 Diagnostics
Two approaches to upgrading the M1A1 fleet are being considered. The first is to replace the current analog LRU's with digital LRU's, link them together with a data bus architecture, and develop BIT and FIT software to test the new LRU's. Essentially, this is an upgrade program similar to the M1A2 program, and it is inherently expensive.
The second approach keeps the existing analog LRU's, upgrades the TNB'S and HNB's by adding a general-purpose processor card with diagnostic software to each, and connects the TNB'S and HNB's to LRU test jacks with cables. Initial estimates are that this approach will be about 75 percent less expensive than the redesign approach.
Reducing Operating and Support Costs
The challenge is to upgrade legacy equipment designed in the 1980's so it can be supported by Force XXI multicapable mechanics. Since the Army will have fewer mechanics, they will have to be more efficient. No-evidence-of-failure (NEOF) rates (for example, faults reported in engines that actually are in running condition and should not have been sent for repair), time involved in managing the parts inventory, diagnostic time, and maintenance man-hours drive operating and support costs. By far, the greatest cost drivers are the repair parts. Some individual LRU's cost more than $160,000.
Lowering NEOF Rates
By providing the mechanic with more efficient diagnostics, we potentially can lower LRU NEOF rates by identifying the faults on a vehicle under the original conditions of failure.
There has been a lot of debate recently about what contributes to NEOF's, how bad NEOF rates are, and whether the rates can be improved cost effectively. NEOF's can be caused by changes in the operating environment, errors in diagnostic procedures, cumbersome test equipment, or the basic design of the system.
 |
| Soldiers and airmen guide an M1A1 Abrams main tattle tank onto a C-17A Globemaster III for transport. |
The Abrams tank initially experienced failures in some electronic boxes in the high-temperature environment of Saudi Arabia. Failed components were returned to the contractor for analysis, and there was no evidence of failure. Subsequently, it was determined that the heat was causing specific electronic components to fail. They worked fine when the temperature was reduced. The circuit cards were modified, and the NEOF rate was reduced significantly.
It is possible to identify LRU faults incorrectly by measuring voltages on the wrong cable pins when using the breakout box and a multimeter. The more experienced the mechanic, the less likely a component will be identified erroneously as faulty.
The STE-M1/FVS is a classic example of cumbersome, antiquated test equipment. The average setup time and diagnostic run time to execute a test is 3.6 hours. When a fault is found and a repair is made, the test must be re-run from the beginning, because STE-M1/FVS lacks the logic to return to the initial location of the fault in the fault tree. After each subtest, the mechanic must connect and disconnect cables, adapters, and sensors, which is time consuming and frustrating. The natural tendency is for the mechanic to take an educated guess at the location of the fault and swap out the LRU. If that doesn't fix the problem, the mechanic swaps out another LRU. Each time a fully functional LRU is swapped out, we generate an NEOF statistic. The other risk is that, by swapping out the LRU's, the root cause of the failure may not be solved. Then, when power is applied to the circuit, the faulty LRU could cause a different LRU to fail. We observed up to three of the same LRU's being "burned" before the real fault was determined.
The other potential NEOF-generating situation occurs when a mechanic successfully uses STE-M1/FVS to troubleshoot the fault and arrives at an "ambiguity group." An ambiguity group is the point at which the test tells the mechanic the fault is in one of several possible components. The vehicle technical manual has alternate troubleshooting procedures that eventually will lead the mechanic to the faulty LRU. This is a time-consuming, manual process, and, if the commander or shop officer is pressing the mechanic to "get the vehicle up to readiness," the mechanic may just swap out all of the LRU's in the ambiguity group. In previous years, when the Army had more extensive prescribed load lists (PLL's), the NEOF's could be controlled at the battalion level. Since PLL's have gone away, the NEOF's move to the direct support repair activity.
The STE-M1/FVS does perform cable testing very well. Experience shows that the cables in a tank seldom are damaged or fail except through damage to connector pins caused by repeated removal and replacement of LRU's or by having heavy objects dropped on them. The M1A2, which has no automated cable-testing capability, has had little problem with faulty cables. On the infrequent occasions when cable testing is required, it is accomplished as an alternate troubleshooting procedure using a standard breakout box.
Reduced Inventories
The quantity of LRU's and shop replaceable unit's (SRU's) stocked is driven by field demands. If these demands are overstated because of NEOF's, the inventories are sized artificially. This causes the Army to procure unnecessary parts. By reducing the number of NEOF's, the Army can reduce the size and scope of the wholesale supply system.
Shorter Diagnostic Times
By providing the mechanic with embedded diagnostics capabilities, the Army will lower the actual troubleshooting time and allow each mechanic to accomplish more repairs in a given period of time. When each mechanic is more efficient, the number of mechanics required will be smaller.
Fewer Maintenance Man-Hours
If mechanics can identify a failed component more accurately, they will replace fewer fully functional components. Additionally, many failures, such as those in wiring harnesses and cables, are caused by removing and replacing LRU's. The Army can use its maintenance resources more efficiently if mechanics can improve their ability to identify failed components correctly the first time.
Turret and Hull Network Boxes
As mentioned earlier, PM Abrams is redesigning the TNB's and HNB's, because the companies that originally provided the parts for the boxes no longer build them. To have additional components made would be cost prohibitive. Both boxes consist of multiple relays that are analog controlled.
By leveraging our investment in the M1A2 SEP, we can take advantage of the redesign of these boxes to implement digitally controlled switching. As a byproduct of the backplane (a circuit board containing sockets into which other circuit boards can be plugged) added during the redesign, we can use the M1A2 SEP's general-purpose processor circuit card as the basis for BIT in the TNB for the turret LRU's. An M1A2 SEP common memory circuit card will host the BIT software. We plan to use existing M1A1 electrical schematics, coupled with the existing diagnostic flowchart logic used by STE-M1 and a turbine engine diagnostics demonstrator, to develop the BIT software. We also plan to add permanently connected cables between the TNB and the test jack connectors on the LRU's. A similar approach will be followed with the HNB, using as many common components and software as possible. Both the TNB and the HNB will have a self-test capability as part of the obsolescence redesign.
In effect, when the modification project is complete, the Army will have converted the M1A1's to the M1A2 diagnostic concept of an initial BIT run by the crew, with results reported to the direct support maintenance support team. The direct support multicapable mechanic will verify the results of the initial BIT and then run the more in-depth FIT portion of the BIT.
Although no one is happy doing more with less, it is a fiscal reality that Force XXI will have fewer maintainers. New diagnostics in the Abrams fleet will allow them to be more efficient. ALOG
Leonard M. Konwinski is a branch chief in the PM Abrams Logistics Division, Army Tank-automotive and Armaments Command, Warren, Michigan. He has a B.A. degree in journalism and an M.A. degree in public administration, both from Central Michigan University. He also is a graduate of the Army Materiel Command Supply Management Intern Training Program.
Lieutenant Colonel Paul M. Wilson, USA (Ret.), is a contractor software engineer with the PM Abrams Logistics Division, Army Tank-automotive and Armaments Command, Warren, Michigan. Before his retirement from the Army, he was the Product Manager for the Army's M4 Command and Control Vehicle (C2V). He holds a B.S. degree in electrical engineering from the University of Texas at Austin and an M.S. degree in electrical and nuclear engineering from the University of Nevada at Reno.