On 14 May 1966 the first PHYLLIS ANN aircraft, serial number 43-15112, landed at Saigon’s Tan Son Nhut airbase—minus ARDF equipment, which had been removed for the ferry flight across the Pacific. The outfits designated to operate the aircraft were not much better off. The month-old 6994th Security Squadron could man just 5 of 11 authorized officer and 52 of 178 enlisted slots. The “front end” 360th Reconnaissance Squadron was in slightly better shape, with 72 of its 102 officer billets filled and 143 of 181 allotted enlisted troops in place. Nonetheless, operations continued with the two DRILL PRESS aircraft and the original HAWKEYE bird, which had flown a particularly successful sortie on 11 April when 13 enemy transmitters were fixed in the Tay Ninh area north of Saigon. That record was eclipsed when aircraft 112 finally became operational; on 7 June the USAFSS operators aboard her bagged 19 targets.
The ARDF plan sold by the Air Force called for ten EC-47s* to be in Vietnam at the end of June, 1966. It was not to be. The second PHYLLIS ANN aircraft did not arrive until the 26th, and it too was without Doppler components and other mission-essential equipment. Personal equipment was likewise in short supply. Lack of survival vests and radios prevented assignment of individual kits. Survival gear was pooled, with crew members signing out a kit prior to each mission.
The ARD-18 system and its associated Doppler and C-12 compass caused frequent aborts, both ground and air. Even standard operating procedures proved troublesome. Excess vibration caused by running the engines at the tech order-recommended brake horsepower settings in order to stretch endurance forced the already overworked ground crews to put in unscheduled maintenance hours. Changing cruise setting to a constant 2,000 rpm eliminated the problem. Back Stateside, other issues plagued the production program, causing even further schedule slips.
*The designation change from RC to EC was not made until 1967 but for consistency's sake the better known EC-47 is used here. Likewise, the later Tactical Electronic Warfare Squadron (TEWS) designation is shown instead of Recon Squadron in the diagram below.
Command and Control of ARDF
By June, the organizational structure was in place which would oversee the in-country SIGINT effort, including ARDF, as long as American forces remained in Vietnam. At the top of the heap was ASA’s 509th Radio Research Group (RRG), which replaced the 3rd RRU and took over Davis Station on Tan Son Nhut. Subordinate to the 509th were two Radio Research battalions already in country as well as the 8th Radio Research Field Station (RRFS) at Phu Bai. Assigned to one of the battalions or to the 8th RRFS were various Radio Research companies (RRC) and their detachments, serving as Direct Support Units (DSU), typically at the brigade or division level. On 1 June, ASA activated the headquarters and headquarters company of the 224th Aviation Battalion, Radio Research. The 224th commanded ASA’s four ARDF companies, activated on the same date, one of which was assigned to each of the four Corps Tactical Areas in the Republic of Vietnam.
ASA had fostered ARDF in Vietnam and from the beginning had tasked the missions, flown the aircraft, and delivered the processed intelligence to the consumer. In the Army’s view, the field commanders should directly control “all cryptologic assets supporting them.” NSA was opposed to any fragmentation of SIGINT assets—consolidation and control of these assets was, after all, the reason for NSA’s very existence. A compromise of sorts was reached whereby DSUs in support of ongoing operations would be controlled by the field commander. But when the crypto troops returned to main base, control would revert to one of the permanent ASA field sites under NSA tasking.
With the arrival of PHYLLIS ANN the issue of control and tasking of ARDF assets erupted, in the words of an NSA history, “into a three-cornered donnybrook.” Seventh Air Force regarded the EC-47 as an electronic warfare asset and demanded complete tasking control. Gen Westmoreland insisted that all ARDF assets be tasked centrally through MACV. NSA was willing to concede in-theater tasking but maintained that ultimately ARDF was a cryptologic asset, the title to which was held at Fort Meade.
The first round of the contest went to MACV. Responsibility for tasking all ARDF assets was vested in the ARDF Coordinating Center (ACC) in Saigon, jointly manned by personnel from the 6994th and the 509th RRG. Potential consumers requested ARDF support via MACV J-2 (joint staff intelligence) which turned over the request to a coordinating committee representing the various stakeholders within the intelligence community. The committee met each Wednesday to produce a weekly package of ARDF requirements. The ACC translated the weekly package into “frag” [fragmentary] orders for the flying units, issued the following day.
Proof of the Pudding
As the kinks were being worked out of USAF airborne direction finding operations, the two DRILL PRESS aircraft continued to collect extremely valuable SIGINT, particularly against COSVN (Central Office, South Vietnam), the main communist headquarters in the south, located somewhere in the jungle near the Cambodian border north of Saigon. Although ASA ground sites were capable of intercepting most COSVN communications, reception quality was poor and much traffic was garbled or missed as a result. The airborne DRILL PRESS platforms suffered no such limitations; in 28 days USAFSS intercepts provided the basis for 17 “significant intelligence reports”. An ASA field commander conceded that “Drill Press is providing an estimated 75 per cent of our usable intelligence.”
By mid-July, 1966, four PHYLLIS ANN EC-47s were in Vietnam, but the gain was somewhat offset by the departure of the HAWKEYE aircraft, its prototype ARDF equipment having finally given up the ghost. Crews on the available aircraft fixed 114 targets on 25 missions, logging just over 200 flying hours. Thirty-five fixes (31% of the monthly total) were in direct support of the 1st Division’s EL PASO II operation, aimed at elements of the NVA 9th Division along National Highway 13, justifiably known as "Thunder Road," north of Saigon. An additional seven fixes went to support Operation YORKTOWN, a minor search and destroy mission by the 173d Airborne Brigade against a suspected VC battalion. On the downside, 12 missions aborted, mostly the fault of the ARDF equipment.
August would prove to be even busier as PHYLLIS ANN, bolstered by the arrival of a fifth aircraft, continued to support allied operations in the II and III Corps areas. On 1 August, the 25th Infantry Division launched operation OAHU, aimed at securing the airfield and nearby base camp at Tay Ninh West and destroying enemy guerilla units which controlled the surrounding areas. Although “all data processed” prior to the operation pointed to possible contact with a large VC force, nothing materialized. But the sweep destroyed many unmanned fortifications and captured or destroyed much enemy materiel. The EC-47 crews made 17 fixes in support of OAHU, although for reasons unstated none of them were passed directly to the army DSU.
Additional PHYLLIS ANN missions were flown in dedicated support of Operation TOLEDO, a search and destroy operation east of the Saigon-Bien Hoa area. TOLEDO was a combined operation, conducted primarily by the 173d Airborne Brigade, but also including units of the freshly arrived 1st Australian Task Force (1 ATF), a USMC Special Landing Force, and two battalions of ARVN rangers. The operation kicked off on 10 August based, according to the 6994th history for the period, exclusively on ARDF results. The fixes located the headquarters of the VC 5th Division within the “secret zone” of the May Tao mountains, now to be the upcoming area of operations (AO).
On the 14th, a decision was made to attack the suspected location of the 5th Division headquarters with armed helicopters, preceded by the dropping of CS (tear gas) on the unsuspecting enemy. Heavy automatic weapons fire was received, but the choppers departed unscathed after raking the area for 10 minutes with machine gun rounds, 40mm grenades, and 2.75 inch rockets. Heavy foliage prevented conclusive evaluation of results. When “subsequent intelligence reports”, indicated that the VC headquarters had moved approximately 10 kilometers north within 24 hours after the strike, the target was taken under fire by heavy artillery. The 173d’s after action report noted that “intelligence sources were unable to detect activity for a period of several days.” When ARDF indicated further movement on 19-20 August the attack plan was repeated, although this time the CS-dispensing helicopter took non-lethal hits. Again the target was not heard for several days. When new activity was noted, artillery was again directed at the coordinates.
The 173d Airborne Brigade credited its DSU (the 404th RR Detachment) with providing 90 enemy locations during the twenty-nine days of Operation TOLEDO. Forty-six of these locations—more than half—came by way of PHYLLIS ANN EC-47s and the 6994th Security Squadron. TOLEDO no doubt represented “a serious supply setback” for the enemy, but the operation was “characterized extremely short engagements with a fleeting enemy and the VC’s preference of abandoning base camps and supplies rather than fighting.” The 173d counted only eight enemy bodies for the loss of seven of their own. Supplies could be replaced and base camps rebuilt or relocated. The VC 5th Division, like many other repeatedly engaged enemy units in this long, exasperating war, would be heard from again.
Aiding the Aussies
Phuoc Tuy province, on the coast southeast of Saigon, was the specially chosen Tactical Area of Responsibility (TAOR) of the 1st Australian Task Force, headquartered at Nui Dat. On 7 August, to the west of the TOLEDO AO, the Australians launched Operation HOLSWORTHY, intended to cordon off the French plantation village of Binh Ba and root out VC guerillas and sympathizers. Among the support elements of 1 ATF was 547 Signal Troop, a SIGINT unit patched together, much like the American DSUs, for a sudden deployment to Vietnam. In late July, the 6994th began passing fix data to the Australian DSU in support of operation HOLSWORTHY, even before it officially began. A transmitter associated with the 275th Regiment of the VC 5th Division was soon identified. Day by day, the analysts of 547 Signal Troop tracked fix locations provided by their American counterparts in the air above, including PHYLLIS ANN.
“Like a string of beads”, the plots pointed steadily westward toward the 1st ATF base at Nui Dat.* But SIGINT could do no more than provide the location of a lone radio transmitter. When infantry patrols failed to detect an enemy presence, a ground attack was deemed unlikely. But on the night of 16/17 August, Nui Dat was heavily bombarded from positions to the east of the base. Patrols sent out on the 18th located the now abandoned firing positions. Although SIGINT indications continued to point to the 275th Regiment, the size of the enemy force—and its intentions—remained unknown. At mid-afternoon of the 18th, a company-size patrol searching the nearby rubber plantation made contact with 6 or 8 VC, who promptly scattered. The Australians pursued but within an hour were pinned down and fighting for their lives against what was obviously a far larger enemy force. Of the 108 men in the patrol 17 were killed; another 25 wounded. It could have been worse. Only timely and accurate artillery fire enabled the embattled troopers to hold off the enemy while a relief force assembled and deployed.
Although not apparent at first, the Viet Cong themselves had taken a terrible beating, mostly from intense allied fire power. Scattered amongst the rubber trees at Long Tan lay 245 enemy dead. What had looked like a certain defeat—even a narrowly avoided massacre—now appeared to be an astounding victory over an enemy force that had outnumbered the Australians on the order of ten to one. Exactly which enemy units had been encountered was not immediately established. Opinions likewise differed as to whether the Australians had walked into a trap or had instead disrupted an enemy main force bent on overrunning Nui Dat base. Whatever the case, 547 Signal Troop had sounded warnings, albeit limited in specificity. The larger problem was the failure, or at least the inability, of the command echelon to act upon these warnings. The reasons, in part, could be laid to inexperience in utilizing SIGINT and the extremely limited number of personnel cleared to receive the information. These issues were in due course overcome. Although small in numbers—its original allotment was only 15 spaces—547 Signal Troop was large in experience and expertise. It would soon establish a well deserved reputation for professionalism within the allied SIGINT community.
* This "string of beads" is illustrated by a map published in a volume of the official Australian history of the Vietnam war. The map, however, is hypothetical, and is not a representation of actual ARDF plots made at the time.
Article expanded and updated, 16 December 2016
by Joe Martin