Red Ribbon Panel Study

 

V-22 NATOPS, Reports, Issues and how they relate to the Marana Accident

                                          (by the Red Ribbon Panel, Jan 2002)

The following are the results of our studies to review and determine if there was any
direct or indirect connection between:

(a) Deleted/deferred flight testing (as reported by last years GAO and DOD/IG Reports), 
(b) Limits for Pilots in the NATOPS Flight Manual,
(c) Performance Test Data and JORD requirements,
(d) Potential impact/errors in JAGMAN investigations and conclusions.


1. Background Information

a. NATOPS Limitations directly related to the Marana Accident:
Normal Load Factor for VTOL/Conversion Mode are -0.5 G to 2.0 G  --- with Max Angle of Bank of 60 Degrees.
("     "     "     "     "    "     "    Airplane Mode"      "    -1.0 G to 3.5 G  ---   "     "    "     "     "     "  "   60-75  Degrees) 

Note: These limits as related to VTOL/Conversion mode are not supported by actual flight testing.


b. Warning concerning Settling with Power in NATOPS states:
"Avoid descent rates of 800 ft/min or greater at airspeeds less than 40 KCIAS." 
There have been several questions asked as to where/how this was decided to be included in the NATOPS Flight Manual. We have found no information based on actual V-22 flight testing for determination of these limits, which are related to VRS. There has been no explanation as to how these Limits were determined as they relate specifically to the V-22 Tiltrotor.

The only indication found, lacking any evidence of  V-22 flight testing prior to the Marana accident, was contained in a Navy Helicopter Training manual which applied to the much earlier and smaller TH-57 Trainer Helicopter:

"Vortex Ring State - The uncontrolled rate of descent caused by the helicopter rotor encountering disturbed air as it settles into its own downwash, also known as power settling. This condition may occur in powered descending flight at low airspeeds while out of ground effect, when rate of descent approaches or equals the induced flow rate. At 300 - 600 ft/min descent, vortex ring state may begin and will not clear until exceeding 1500-3000 ft/min. Glide slope of 70 degrees (nearly vertical) seems to increase the possibility of settling. When these conditions are met, the rotor pumps air into a large bubble underneath it, which then bursts, disturbing air flow and blade thrust. Because approach angles less than 50 degrees and airspeeds of 15-30 kts allow enough new air to enter the rotor system, the TH-57 is limited to descent rates less than 800 ft/min, with airspeeds greater than 40 KIAS, and approach angles less than 45 degrees. "

If in fact, this or similar helicopter flight manual were the source for the V-22 NATOPS Manual Warning - as opposed to actual flight test verification by the Contractor/Navair/NASA  - it would be a serious oversight by all involved.

The application of past experience in smaller Helicopters to derive Limits for an unknown/untested laterally displaced - side by side - rotor aerodynamic system, would be a gross error.

Note: It is a matter of interest that one month AFTER the mishap,  Navair issued Urgent Change No. 1 to NATOPS, adding a description of high rates of descent in VTOL Mode and the procedures to recover from this "regime"

It is also of interest throughout the Jagman report, that flight in the V-22 is only discussed in a two-dimensional consideration. There is no discussion related to the fact that the V-22 must and does operate in a three-dimensional
airspace. Discusion of lateral and yaw controls and their impact on the above is not presented, nor are any of the other flight situations involving Multi-Axis controls presented or discussed.

As will be shown later, the only guidance found was the NATOP Limits of 2.0G and 60 Degree Angle of Bank in VTOL/Conversion mode. The reason for a Prohibition of rapid multi-axis controls is not explained so as to alert the pilots of how this impacts on the foregoing.  

It appears through what little information/research is available in the public domain, that neither Navair nor NASA  had any idea or direct knowledge, through V-22 Flight testing prior to Marana, about the real flight "boundaries", other flight limits or any critical effects of maneuvering.   They do in fact appear to be arbitrary and based on earlier helicopter limitations.

Despite the foregoing, the above limits/warnings in the V-22 NATOPS appear to have been incorrectly applied in the  JAGMAN and Safety Investigations in determining if pilots had exceeded them in violation of the NATOPS.



c. Navair/Bell/Boeing Flight Test Performance Data Chart.
This chart (attached) provides results of actual flight testing, presented to the Blue Ribbon Panel, which should have been used as the factual/experimental basis to derive the proper limits for NATOPS Flight Manual for Pilots.

The average Maximum Load Factor (Nz = "G") displayed on this chart show a Maximum average of Nz = 1.15 to 1.20 'G"s for Hover/Conversion modes.

Note: This indicates level flight demonstrated Load Factors. No information was given as to the Load Factor Capability
in descending flight, which would have been more critical in the conditions involved in the Marana accident case.

This minimal Load Factor in turn projects a Maximum  Angle of Bank of about 25-30 Degrees.

Note:
This estimate is also based on the level flight test only. It might actually be less in descent if the Rotors are Stall Limited. In descent the pilot would have more collective control available to use, which would make it easier to readily exceeded these minimal levels, before power (torque) limits were reached.

It should also be noted that the performance envelope "goals" - as shown by the dashed lines for each nacelle angle test segment on the chart - indicate that the V-22 Goal for Load Factor in the VTOL/CONV mode  was approximately  Nz= 1.75 G's, which was significantly higher than actual testing demonstrated.

d. JORD Requirement Charts,
dated 15 Feb 2001   (included in Appendix to Final Blue Ribbon Panel Report).

The established JORD Requirement Number 028 requires:
Instantaneous G Loading of  +3.0 to -0.5 instantaneous G's in Helicopter Mode, and
"     "     "     "     "    "     "        +3.5 to -1.0  "     "     "    "    "   in the Airplane Mode.  

It is obvious from the performance testing  that the V-22 cannot generate +3.0 G's in the Helicopter Mode

It is therefore interesting to note that the Current Status of this requirement is Graded in Green indicating "Full Capacity Exists or Threshold Met"

The inference of this claim also suggests the conclusion that a bank of 65-70 Degrees,  which has also not been (and could not be) demonstrated.

No data to support this gross conclusion has been found in any test reports.

The various items for  Limitations required of pilots as noted in the NATOPS are not supported by either the Flight Test Performance Charts, nor the grading shown in the JORD Requirements which has been graded as being accomplished.

We now have the following variations in "G Loadings"  in Helo/Conversion mode ranging from:

+1.2 G  in actual Flight Testing
+2.0 G  claimed in NATOPS
+1.75 G goal of Flight Test
+3.0 G  claimed as completed in JORD Requirement Status Report

It is obvious that the unexplained variations of information in this area of V-22 Performance capabilities and limits,
are completely impossible to reconcile. (Person or persons responsible should be interviewed/investigated as necessary)

e. Pilots Decisions
Assuming that most Line Pilots are not aware of anything except the NATOPS limits, which are considered by pilots as their flight "Bible" (i.e., not open for argument), we can conclude that a pilots primary basis for decisions are  dependent on published limitations in the NATOPS Manual.

Pilots would normally be correct in assuming (but incorrect in reality) that they could safely make 60 Degree banks at 2.0G during flight in Helo/Conversion Mode; which in reality could lead them immediately into an asymmetrical STALL.

It is inconceivable, given the foregoing  that a pilot following NATOPS could be found the Cause of a mishap such as the Marana Accident.
                                               

2. V-22 Test Program problems directly related to Marana Accident:

As related to several above reports, the testing in the 0-60 Knot airspeed ranges, which was required to provide accurate data for determining limits for NATOPS, was deferred/deleted/ignored, before the aircraft was released to line Pilots ( who were NOT Certified Flight Test Pilots ).

Load Factors for VTOL/CONV Modes, as well as warnings referenced to VRS,  were apparently and wrongly adapted from other standard Helicopter NATOPS Manuals.


No evidence has been seen which would allow any other conclusion to be made.

                                                  
3. Jagman Investigation of Marana Accident:

A review of the Jagman investigation indicate that several items of interest were either ignored, missed or left out which  directly involved the findings and conclusions. Many of these would have been changed if they had been studied, evaluated, and included in the investigation. Absent the information above, it becomes apparent that insufficient data and knowledge was available to the investigators to permit more accurate/complete conclusions.

Extracts from Jagman Report

a. "The Mishap Aircraft's flight profile in the terminal area (high descent rate/low airspeed) most likely resulted in the aircraft experiencing a Vortex Ring State (Power Settling) and /or blade stall condition, which resulted in departure from controlled flight and the subsequent mishap" 

b. The report also stated that this condition is "more extreme than the results experienced in most rotorcraft to date."

The Deputy Commandant for Aviation did NOT concur with this statement.

Note: In the history of helicopter flight operations, we have never found a case where a Helicopter did a snap-roll onto its back and crash into the ground - because it got into a Vortex Ring State (VRS). An encounter of helicopters of a VRS condition is a very rare condition.

When a former FAA Flight Test Pilot with some 40+ years of flying helicopters, was asked what comments he had to offer about the recent flurry of attention by Navair and NASA of  the "VRS" problem, he seemed surprised - and asked "What is VRS?"

Given that this is a well known fact to all helicopter pilots and aero-engineers; someone should determine why the General non-concurred, and also why others - including Contractors/NAvair?USMC/NASA - have repeated similar statements!!


c. "This unfortunate mishap appears not to be the result of any design, material, or maintenance factor specific to tiltrotors"

This conclusion appears to be inconclusive and misleading and has been repeated for the past few years.
It asserts that the V-22 Tiltrotor aircraft  "appears" not to have any design problems.

Efforts by the Red Ribbon Panel over the past year has shown that this is not true.
It has a very basic and major design flaw in its rotor design and configuration. 
This has been demonstrated by: default of testing and performance, numerous limitations and warnings related to control uses by the pilots, and more written emergency procedures and warnings/cautions in the V-22 NATOPS flight manual than any other military/civil aircraft known in current service.

Note: A thorough human factors engineering analysis of the amount of material to be absorbed by the pilots would be of great value to explain the level of difficulty required to safely operate the V-22 in combat maneuvering conditions -- which are prohibited (even after some 10+ years of flight operations). As will be seen later, the Marana accident aircraft were, by any evaluation, one of the first actual examples of routine combat maneuvering in formation at low airspeeds altitudes. They operated in an envelope area which had NOT been tested by either the Contractor or Navair.


The Major Design Flaw is directly related to the laterally displaced Rotor/blade design, which produces little or no combat maneuverability.  This was reported by the Red Ribbon Panel on 14 May, 2001.

The only test data known are the Six flight test Data Points shown on the Bell-Boeing/Navair Performance Flight Test Chart, which indicates a maximum limit load capability of  Nz= 1.2 "G"s. This provides  a very small 2/10 of a G margin for any maneuverability or agility, which when coupled with the Lateral  (side by side) Rotor design creates a significant and deadly design fault.

This has been known or should have been known for many years. This finding is supported by (1) lack of flight test, and (2) the use of numerous warnings or cautions in the V-22 NATOPS Manual directly related to the lack of capabilities above.
1. Combat Maneuvering is prohibited.
2. Autorotations are Prohibited
3. Abrupt multi-axis control inputs are Prohibited


As reported before, the V-22 is not - and cannot - be used in any terminal mission Combat area, susceptible to enemy fire.

DOD studies of  rotary-wing losses in Vietnam and subsequent areas of hostility , have shown that 91% of losses were in the Terminal Mission Area. (As one of our pilots have said - it is a "sitting duck" in any Combat Situation! Another combat helicopter pilot offers that  "a Combat V-22 is an oxymoron.")


The "cover phrase" added to the Gen  McCorkle comment in item c. above is:   "--factor specific to tiltrotors" .
This is an intriguing phrase.

The design flaw in question is ONLY a factor specifically applicable to the V-22 tiltrotor!

The Major V-22  design flaw is not associated nor applicable to ANY helicopter.

The use of laterally (side by side) displaced rotors creates a complex problem - which has been subject to little or no testing in low airspeed/low altitude ranges.

The impact of winds, gusts, turbulence and control inputs have not been subject to any specific testing, nor any Technical Reports from either NASA or NavAir. 
Note: A search of both AHS (American Helicopter Society) and NASA Technical Report Databases reveals not a
Single Report Addressing VRS or Maneuvering Issues with the V-22 prior to 2001!


The coupling or addition of laterally displaced propulsion,  with twin rotors which have the lowest limit load factors for hover than any known Military Combat Capable Rotorcraft; results in what IS a Major Design Flaw, which SPECIFICALLY applies to only this tiltrotor - the V-22. It is a Flaw in that it has not been tested nor demonstrated that it can perform safely in its intended role as a combat capable aircraft.

The Marana accident would appear to be first proof of this. Given the test results, it is believed that this aircraft would not be certifiable by the FAA for civil use.
It is interesting that the Memorandum of Understanding between Navair and the FAA, which provided that the FAA flight test pilots would be directly involved in the testing of the V-22 was later waived by the Navy. Given the normal thoroughness of FAA Flight Testing Standards, it may be of interest to investigate why/how this occurred.    

Roll Damping and Roll Inertia
Unlike a helicopter where the tail rotor act as a large Roll Damper, and the central location of the rotor precludes any real problems with Roll Inertia;
The V-22 has little or no Roll Damping, and a very large Roll Inertia.

Locating the engines and transmissions and rotors at the end of the wings (even though the wings are fairly short) makes the roll inertia huge compared to a helicopter and large even by commercial airplane standards.

This large inertia will mean that when a significant roll rate is established, it will be very hard to arrest by normal control inputs (differential collective).

Arresting a significant roll rate caused by control inputs (maneuvering bank or turn) or a powerful external disturbance (such as flying into the wake of another aircraft) may have:

1. been beyond the pilots normal control limits,
2. beyond the power (rotor torque) limits,
3. within the control/power limits but one of the rotors stalled because of the low stall margins on these rotors and large differential collective pitch applied through the control inputs, or
4. the control response rate was just too low to be effective (aerodynamic lag).

The application of differential collective in a banking situation probably helped to stall one of the rotors, promoting an even more powerful roll in the same direction.

With the high roll inertia combined with the low roll damping, it could well be that this is one reason that maneuvering limits were placed on this aircraft when in helicopter mode. We suspect that this was determined early on from flight simulation work, and written into the NATOPS.
If this is correct, then it may become another example of where the JORD called for a certain level of maneuvering capability and the contractors went ahead anyway hoping that they could get around this deficiency somehow, most likely by changing the JORD to fit the capabilities of the aircraft. 

There is NO fix or design change which will cure this Design Fault -- without a complete aircraft redesign.

d. Annex A of the Jagman Report Synopsis.
It is of passing interest to note that this annex discusses VRS in BOLD face print,  while 'stall' is in normal print and not capitalized.

The discussion of Blade Stall correctly relates Angle of Attack (not mentioned in VRS discussion), to blade stall -- and to stall region on "one prop-rotor while decreasing the stall region on the other."



4. Asymmetric Stall Considerations
The probability of Asymmetrical Stall was apparently not considered as a serious factor in the investigation, or it might have been noticed that the maneuvering of Mishap aircraft from formation high left to formation high right followed by realignment with lead aircraft,  would be a perfect setup for asymmetric stall on the right rotor. The rotor at this time enters an unusually high Blade Loading and low Stall Margin.

It should be noted that none of the published reports or tests include any multi-axis control testing. All of the past TR-64 tests for examining VRS conditions have been done only with fore and aft cyclic inputs to help establish boundaries.

The absence of differential lateral cyclic or rudder inputs may result in the realities that the new boundaries/limits MAY NOT be adequate or SAFE if and when any lateral or multi-axis inputs are required!

Given the higher density and temperatures involved at Marana, the Nz (G's) available for maneuver would have been reduced enough (G = 1.15-1.20to cause any significant banking in formation maneuvers -- added with yaw pedal activation -- to exceed the Stall Margins during the interval between the approximate times of 19:57:30 (aircraft in conversion mode) and 19:57:57.

There is no indication in the Jagman Report that any consideration of the relationships noted in the foregoing information and data was given.

There was also no information found in the Jagman Report on pages 53-54, where some Eight or so NATOPS provisions were discussed; which might have drawn attention to the specific limits to the pilot in regard to the "incorrect" NATOPS 'G' capability,  nor the incorrect banking capability/limits for the pilots.

It can reasonably be accepted that the average line pilot does keep in mind the normal/standard limits - as opposed the aircraft special limits.

The fact that NATOPS could lead a pilot to believe that under the conditions for hover/conversion mode (2.0 G and 60 Degree Bank) when such capability did not exist, could easily have been the primary factor in this accident.

The apparent rush or jump to discuss VRS as opposed to Asymmetric Stall apparently caused the JAG to assign fault to the pilot. There was no indication that consideration was given to the assumption that the pilot might have been following the correct or specific procedures following NATOPS stated rules/limits.  The selection of criteria for avoiding a little known phenomenon of "VRS" as opposed to clearly stated control limits in NATOPS would have been an incorrect judgment by the investigators. This would be a tragedy and should be reviewed and corrected as necessary.

The Jagman Report apparently does not provide continuous readings of the CSMU (Crash Survivable Memory Unit) for the last fifty seconds of flight between the initiation of Conversion for the 2nd Aircraft; through the Crossover and bankings required during conversion to change formation sides, and the need to to get realigned with lead when he is in High-right position.
The maneuvering necessary to follow lead into a left hand formation turn, while initiating conversion and maneuvering from left high to right high was not discussed in the Report.

A complete time line discussion, including not only the following, but one which should include other registered CSMU parameters, might have provided Angle of bank information which would allow a second by second reconstruction to perhaps validate a rotor stall, instead of reaching a premature conclusion supporting the 'VRS' theory. Any attempt
by the pilot to maneuver under the Standard NATOPS limits for helicopters of 2.0 "G"s and 60 Degree Angle of Bank would have produced an immediate right rotor stall.

(Note: Airspeed has been converted to give a better view of ROD/Airspeed relationship)


Time            Nacelle(Deg) AGL(ft)   ROD(fpm) Airspeed(fpm)  AS/ROD           Comments
19:57:13             0              1280        500         18,840            37.68               begin conversion
19:57:14            convers       1350        885         11,140           12.59         began crossover, unknown Bank Angle
????
19:57:35            51              820        3945        10,229             2.59                     crossover
??
19:57:38            73              unk        unk            unk               unk                       on right side
??
19:57:44            90              566        150            9,317            62.11          slowed, slight balloon
??
19:57:50            90              500         800           5,267             6.53            left bank,  3 O'clock High
??
19:57:57            95              339         2247         4,051             1.80            multiple axis controls
19:58:03 IMPACT


Note: 
Multiple axis controls were clearly involved in the maneuvering control actions required in formation flying.

WHY WERE MULTI-AXIS CONTROL INPUTS PROHIBITED IN THE NATOPS?


This should have raised a Red Flag in the JAG Investigation, that there was in fact something already known by Navair/Contractors/NASA about the V-22 maneuvering control problems.

What was known that established this concern and prohibition - other than lack or cancellation of flight test -- and why wasn't it addressed properly in the NATOPS to preclude inadvertent accidents?

At (:13 sec), a/c began conversion from Airplane mode to Helicopter Mode, with a rapid decrease in airspeed from 18840 fpm to 11140 fpm at (:14sec) (if CSMU correct).

Between (:14sec) and (:35sec) the a/c began crossover maneuver, with unknown angle of bank, while continuing conversion from 0 Degree to 51 Degree nacelle angle, and very rapid descent required to follow lead in his descent from the IP (Initial Point). 

3 Seconds later CSMU indicated at (:38 sec) that the aircraft was somewhere on "right side".  Jagman reports nacelle at 73 Degrees, but provides no data re A/S, ROD, bank angle, or AGL, for this critical time period.

At (:44 sec), conversion was almost finished, with 90 Degree Nacelle, and the a/c slowed and made a "slight" balloon event.
The airspeed was still slowing, and the ROD had slowed significantly from 3945 to 150 fpm.

Given that between (:35 sec) and (:44sec) the a/c was in crossover maneuvering; followed by a slight balloon and needing formation correction to left a few seconds later ( :50 sec), the pilot made a left bank (of unknown amount) with an A/S to ROD ratio of about 6.53/1. 

Given the several time gaps - some significant-in the Jagman Rpt, a definitive time when the aircraft may have entered a maneuvering stall or encountered VRS is hard to establish. The maneuvering time zone between :35 and :50 Seconds
does show that a definitive Left Bank was made at :50Sec.

It would appear that in making a Left bank, the a/c  exceeded  the aircraft's actual "G" and Banking limits
(probably with about Nz=0.2 G  reserve with less than 5-10 Degree bank capability), which probably stalled out the Right rotor! 


(Note: The only real guidance available to the average pilots was that NATOPS limits were 2.0G and 60 Degree Bank. The actual Bell/Boeing/Navair Flight Test Performance Chart indicates that Max Maneuver Lift reserve was about
0.2 G!! in the 0-60Knot range!!!!
.) This indicates that neither the Lead aircraft nor the Accident Aircraft had sufficient maneuver margin to proceed in flight conditions (which would not have affected any other Military Helicopter).

Since the NATOPS manual appears to have included incorrect information that had not been validated by any actual flight testing, any pilot left in the above situation-regardless of training to the contrary --  may have had no reason NOT to use standard banking techniques in formation maneuvering. His ignorance of this situation might well have resulted in the unnecessary deaths of all aboard the mishap aircraft.

IF the CSMU data which could fill in the missing time and Bank Angle information was plotted, or if it  is  available in one of Jagman  Appendices, a complete review should be immediately accomplished!
VRS is an interesting subject, but the unknown and untested results of Asymmetric Thrust and Lift parameters associated with a blade/rotor stall limits, may well be far more important to Pilots, particularly in the lower airspeed/altitude terminal mission areas which require a high degree of maneuverability under pressure.

--------------------------------------------------------------------------------------------------------------------------
Conclusions:
Taking into consideration:

1. the conflict of statements and documents related to the Flight Testing results and the determination or signing off of incorrect G Limit factors both in NATOPS and In JORD Briefing Charts,

2. the lack of availability to put correct data, limits and proper discussions in the NATOPS Manual,

3. the lack of adequate discussion related to asymmetric configuration under various situations involving both Stalls and VRS,

4. the lack of inclusion in the 'Findings of Fact' of any scenario formation maneuvering and use of flight controls/timing which would have led to a more probable conclusion of an Asymmetric Stall,

5. the unwarranted decision to place the onus on the pilot for not complying with NATOPS, when NATOPS did not cover the relationships between Maneuvering considerations necessary to fly in formation flight (and at the same time avoiding the PROHIBITION against Combat maneuvering).

Recommendations for Investigation:

1. The basis for statements by any personnel offering as facts the "sameness of the V-22 and helicopters" when such conclusions were baseless, and can lead to carelessness and accidents.

2. The basis for the administrative or negotiated elimination/deferral/deletion of tests which were required (by contract?) for use as a sound basis, for the determination compilation of correct/validated FACTS for use in the NATOPS  Flight Manual.

3. The basis for consideration by Navair for certification of airworthiness and/or clearance for Line Pilots prior to completion of flight testing and acquisition of the data required to establish proper NATOPS instructions, Limits,
Warnings and Cautions.

4. The basis for decisions related to the assignment by Navair to the Contractors for the primary responsibility for the Flight Test Program requirements established under the ITT (Integrated Test Team) concept

5. Lacking adequate testing, the basis on which the Contractors and Navair used to insert in NATOPS material which was in error, false, or misleading and which would in turn lead otherwise qualified pilots into unfamiliar and dangerous situations.

6. The apparent use of outside advisors/consultants/officers, not familiar with the true performance and behavior of Asymmetric tilt rotor Propellers (as opposed to conventional helicopter rotors), and the numerous situations waiting for the uneducated pilot not familiar with the idiosyncrasies involved.

7. The degree of the management awareness of the decisions related to all of the foregoing, and any obvious misrepresentations under contract related to all of the above obvious discrepancies.

8. Given that the real primary cause was the failure in airworthiness of the unique design of the asymmetric propeller system, which had not been fully and properly tested and certified; consideration should be given to set aside any assertion or conclusion that the pilot of the mishap aircraft be deleted from all correspondence, reports and announcements related to this mishap. 

9. Finally it is recommended that based on the above, that the conclusions reached by the Jagman investigation should be set aside, until and unless more testing of the aircraft related to both Asymmetric Stalls and VRS can conclude that the aircraft is certifiably safe in the same degree as that required by the FAA for transport aircraft.

Note: It appears that not a single person on the Blue Ribbon Panel had any real experience in Rotorcraft aerodynamics, flight testing and operations;  which otherwise might have resulted in the discovery of most of the above findings. All of the material above was available in the briefing materials provided to them, and in the NATOPS Flight Manual. These were the same documents used by the Red Ribbon Panel in our research and studies which produced most of the foregoing materials and discussion.

CONCLUSION:
The above information solidly indicates that the V-22 rotor system is in fact a FLAWED Safety of Flight Design, which CANNOT be 'fixed " without  a complete aircraft re-design.
It should be terminated immediately , before more men are killed.



Harry P Dunn
Coordinator, Red Ribbon Panel