First-Hand:Signal Processing for the Lunar Module

submitted by George Yabroudy

I am an IEEE Life Senior Member. I worked on a Grumman subcontract for the Lunar Module (LEM) as a project leader from 1962 – 1972. My company, (Arma Division of American Bosch Arma Corporation), designed and produced two major Electronic Assemblies which were located in the aft equipment bay of the LM ascent stage. Those signal processing units were part of the LM Instrumentation System and were known as the SCEA (Signal Conditioning Electronic Assembly) and the CWEA (Caution and Warning Electronic Assembly).

I wrote and presented a paper, at the SAE Electronic Packaging Conference in NYC in February 1967, entitled “Packaging of Electronic Signal Conditioning Equipment for the Project Apollo Lunar Module”. My paper describes, in detail, the functional and environmental requirements of our equipment and the resulting electronic packaging design. Our equipment flew on all Apollo missions involving the Grumman Lunar Module, including all six successful moon landings and the fateful aborted mission, Apollo 13.

My recollections will not duplicate the technical details covered by my paper. Instead, I will attempt to recall the atmosphere and some unique aspects of the Apollo program and the cooperative relationships needed between Grumman and Arma and also between different engineering disciplines on the project.

NASA selected Grumman Aircraft Engineering Co. of Bethpage, NY as the prime contractor for the Lunar Excursion Module (LEM) which, along with the Service Module (SM) and the Command Module (CM) connected together, constituted the spacecraft to the moon and the payload atop the huge Saturn 5 rocket. Grumman selected subcontractors for various subsystems and major elements of the LM. Grumman did the LM system design and fabricated major structural elements.

Grumman selected Arma Division of American Bosch Arma Corp. to design and build two electronic “boxes” which were part of the LM Instrumentation System. Arma was located on Roosevelt Field, about 5 miles from Grumman. A good omen for a historic voyage: Lindbergh took off from Roosevelt Field in 1927 on his transatlantic flight to Paris.

Arma was no stranger to major military contracts and complex systems during WWII and thereafter. Arma had won a competition to design and build the tail turret fire control system which was on early models of the B-52 Bomber in the 1950’s. In 1957, Arma was selected by USAF to design and build the inertial guidance system for the Atlas ICBM (world’s first intercontinental ballistic missile). This (SECRET) program was fully ongoing in the 1960s. In 1962, Arma employed about 3000 people including about 400 engineers. The two LM contracts, original value about 3 million dollars, employed less than 100 people at any one time, including about 40 engineers of all kinds, from 1962 to 1972. By the time the program ended, the Arma contracts had grown in value to about 30 million dollars. Many former Arma engineers now worked for Grumman, the major employer on Long Island. Arma engineers and Grumman engineers were very often neighbors. Both companies enjoyed a reputation for quality engineering and superior products. Considering all these factors, working together on LM was an easy and natural fit, and so it proved to be. The watchwords on the program were simple: number one was crew safety, number two was mission success. Where tradeoffs could be made, weight reduction was sacred. For many reasons power reduction was to be aggressively pursued. Any reduction in power requirements also translated into a weight reduction. Every one pound reduction in the payload results in a 200 pound reduction in booster power needed at Earth lift-off.

Grumman defined the standard package envelopes and electrical, mechanical, and thermal interfaces. Within these outlines, we had design freedom. The SCEA is a signal processor which accepts simultaneously 300 different input signals from almost every sensor or critical source aboard the LM and produces 500 electrically isolated outputs to various data display, warning, and telemetry equipment in the LM cabin. The SCEA consists of 2 assemblies weighing 70 pounds total. Total power input to SCEA is 30 watts. Our biggest design challenge was not to just make a product that would function as intended. We had to make it small and very light in weight. It had to be very, very reliable and consume very little power. It had to be rugged enough to survive rocket launch, and the extreme temperatures and total vacuum of outer space. Electrical circuit parts had to be of proven reliability, which meant no integrated circuits back then. Resistors, capacitors, transistors, diodes, etc. were 100% “burned-in” before use to weed out “infant-mortality” failures. Piece-part leads had to be of materials suitable for electronic welding connections to nickel ribbon interconnects. Electronic welding permitted us to weld close to the piece-part’s body without damaging the part. As a result, we attained high part packaging density and weight reduction due to shorter interconnects and no solder added. With 17,000 parts per SCEA this made for significant weight reduction.

Any piece part failure after assembly into a circuit module required detailed failure recording and analysis to isolate the cause of the failure. Part failures occurring in an encapsulated module required excavating encapsulant with a dental drill to reach and retrieve the suspect part. At a certain advanced level of assembly, every failure analysis required formal written failure reports for Grumman and NASA. Every piece-part had a paper trail, from its location in a specific circuit module, all the way back to its supplier and a specific production lot number. Every part producer retained a representative sample quantity from each lot for possible future testing to identify suspect lots. I do recall one particular transistor failure. The metal cap was removed to examine the chip material under a microscope. “Purple plague” (an aging deterioration) was suspected. We took our bad part back to Sprague Electric in Manchester, NH, where they had retained samples from the same lot a few years back. Some of the sample units were opened up and underwent detailed examination and comparison with our failed item. “Purple plague” was not confirmed. We had a normal “random part failure”. The entire lot was therefore suitable for use. I really never knew exactly what “purple plague” was, but we knew it was something bad, and we did not have it.

On this project, dramatic advances were made in the design and production of micro-miniature signal transformers. These parts were specifically produced for our LM equipment by OECO Inc. of Milwaukie, Oregon. Our transformers, typically, were contained in a sealed 3/8th of an inch cube with 8-9 lead wires emerging from one face. They contained 3 separate electromagnetic coils, some with center taps. The coils were wound with delicate #46, #48, even #50 AWG coated wire using a special process and skilled personnel. We needed more than 500 transformers per SCEA. This all involved a highly cooperative process, and OECO, our supplier and our partner, was outstanding.

Being so near to Grumman meant that it was a fairly simple matter to visit the building where the Lunar Modules were being assembled. I think there were 15 Apollo missions, including 6 successful moon landings, so maybe 14 complete LMs were built over all. It was pretty exciting to see 3 or 4 LMs in various stages of assembly. I was able to climb up and go inside the crew compartment of the spaceship. A Grumman mechanic revealed that any worker on a LM had to do an inventory of his toolbox, before and after entering any LM.

During each mission involving a working LM, Grumman required major subcontractors to designate qualified technical support personnel to be available by phone for inquiries that might be needed. I recall 2 specific incidents where we were involved. On one mission, while in lunar orbit and prior to lunar descent, it was noted by the crew that one 28 VDC ship’s power bus was occasionally as low as 18 VDC. The specified voltage for this bus could be anywhere from 20 to 32 VDC. We were asked if our equipment would be affected by operating at 18 VDC. We were able to respond that we had operated without problems as low as 16 VDC. The other inquiry involved a specific signal channel in a specific module. Did we have any history of failure and/or repair for this hardware? That took some time and some digging. The reply: no.

In July 1969, I was on vacation with my family in upstate NY, when Apollo 11 was in progress. I was on a tour boat in the Thousand Islands section of the St. Lawrence River when over a loudspeaker came the news that “The Eagle has landed”. I could barely contain an impulse to sob and my eyes filled with tears. So the dream had been real after all, this voyage of the century was nearly completed. And late that night, on a motel TV screen, I watched Neil Armstrong take “one giant leap for mankind”. This time, the emotion was euphoria.


Back to Human Space Travel Primary Sources