Direct-Current Eraser Record and Playback System

Page 1

DIRECT-CURRENT ERASER RECORD AND SA DIA SYSTEMATIC DECLAISEIICATION RI•:`/IEW The direct-current eraser record and playback system is designed primarily 1 recording information in a high-radiation field. The system was used on Operation Plurnbbob for recording close-in information from shots Boltzmann and Fizeau. Of the types of recording used in these two shots, the erase system was least affected by electrical transients and radiation. Work performed under AEC Contract AT-(29-1)-789.

Page 3

TABLE OF CONTENTS Comparison of FM and Erase Recording in High Radiation Fields 11

Page 4

LIST OF ILLUSTRATIONS Fig. 2 -- Circuit for plug-in unit -- Modified Fastax camera recording scope-trace 12 -- Neon bulb driver unit 13 -- Circuit for driver unit 14 Fig. 13 -- Transient effect erase-type channels terminating with a resistance at the Fig. 14 -- Erase-type channels at zero time using the plug-in units 8 -- Trigger system for Fastax camera 15 10 -- FM information channels at zero time 17 12 -- Timing and oscillator power supply package 19

Page 5

DIRECT-CURRENT ERASER RECORD AND PLAYBACK SYSTEM The need for a method of recording information in a region where FM and carrier systems are adversely affected by radiation forced us to look at the erase system as i'means of gaining data inside the fireball. To obtain maximum quality in the information recorded, a high-speed playback system was designed which used a Tektronix Model 535 oscilloscope and a modified 35-mm Fastax camera. With this type of playback, maximum frequency response and time resolution can be obtained. A carrier erase playback system is used for playing back , records on an oscillograph recorder at slower speeds. This playback system was designed by Edward Ames, Ampex Model S-3439 tape transports are used as recorders for this system. The transport is a 14-channel recorder using 1-inch tape. The recorder runs at 60 inches per second and is driven with 28 volts of direct current. Tapes are pre-recorded, by use of a HewlettPackard oscillator Model 205AG to record to a 20-kc carrier at 10 milliamperes. A junction panel with monitoring points is used to adjust the carrier level on each channel being used for erase-type recording. Channels using FM recording and timing channels are not pre-recorded Several types of information can be recorded by the use of the erase sys On th#' Nevada shots, the erase system was used to study the effects , of radiation on the following corn ponents: 2 Zipper units, choppers, gaps, condensers, transformers, firing sets4,ineèhanical timers, and baroswitches. In addition, the erase system was used for fireball iiicT radiation Melt plugs, thermocouples, photocell radiation measurements, and pressure transducers inside the fireball were instrumented by this method. (However, pressures were measured with an MR-52 tape transport because the transducer output was insufficient to record on an 2, ITR 1520: Vulnerability of Weapons to Nuclear Countermeasures 3, ITR 1517: Fireball Studies 4. ITR 1510: Vulnerability of Weapons Components

Page 6

A quick-change plug-in unit was designed to furnish a DC voltage for melt studies, mechanical timers, baroswitches, and other components not having a voltage output that could be used as a monitor. This plug-in unit (Fig. 1) and circuit (Fig. 2) delivered at least 10 milliamperes across the head. A 1-ohm resistor, which acted like a short to the rest of the circuit, waa shunted across the head to protect it from transient effects; any transient voltage induced in the cable at time zero works directly into the resistor in such a situation. When longer signal cables are used, a 2- or 3-ohm resistor can be substituted in order to get 10 milliamperes through the head with the same driving voltage. However, this would mean less protection for the head, and transient effects will be noticed if the resistance is increased too Mercury-cell batteries (RM-42R type) were used in the unit under discussion not only because of their size and their ability to furnish the required current to the circuit, but also because they are less affected by radiation than are Ni-Cad or silver-cell batteries. Mercurycell batteries also can be changed rapidly, and elimination of the necessity of charging Ni-Cads or silver-cell batteries is a great advantage in the field. Other erase-type channels that were monitoring components were fed through blank • plug-in units. Through voltage dividers, it was relatively easy to record battery voltages, Zipper charge-rates, gap breakdowns, and characteristics of other components which furnish their own voltages. Some components such as condensers were operated into a cathode follower in order to, make them match impedances and to give enough signal to record on the head. Photocell outputs measuring gamma radiation and light intensity were sufficient to record directly across For thermocouples, the regular 8-ohm head of the Ampex recorder was replaced with a 1-ohm head built for the purpose. The thermocouple output was enough to drive a 1-ohm head directly for erase recording. Channels requiring no allowable current drain were monitored through an FM oscillator and were used to record reactions of photomultipliers and thermal batteries. An FM oscillator, modulated with a 1000-cycle timing oscillator, was used to record timing channels (channel 1 of each recorder). Channel 14 of each recorder had a backuptiming signal of 50 kc fed directly to the head. Seven of these 14-channel tape recorders, with all their plug-in units, FM oscillators, cathode followers, and remote-control timingt, for starting and stopping the transports, were built into a single recording packs:* (see Fit. 3) in order to use the restricted shelter space to best advantage. The package's springsuspension shock-mounting system was designed to protect the eqnipment from damage or * recording interruption during the tremendous shock to which it is subjected. Channels recorded on the direct-record package and scope package recorders were also protected with Playback (See Fig. 4 - 1 laybaak Station) An Ampex Model 307-14 tape transport is used for playback. All head cables are fed into a selector junction panel. At slow-speed playback, the selected channels are fed through the carrier erase playback system to the oscillograph galvanometers. At the same time the timing signal is fed through an EMR discriminator and the 1000-cycle output to a galvanometer. If the paper speed on the oscillograph is too slow, and the 1000-cycle timing cannot be resolved, then the signal from the discriminator is fed through a modified Berkeley counter network and is subdivided to a readable timing mark. During high-speed playback the signal from the selector panel is fed to a Tektronix 112 amplifier and then the output is fed to the horizontal plates of a Tektronix Model 535 scope.

Page 11

If it is desirable to rotate the scope tube 90 degrees, the 112 amplifier can be eliminated and the vertical amplifier can be used to deflect the beam horizontally. A Fastax camera modified into a ribbon camera records the scope trace on 35-mm film (see Fig. 5). The film is moved past the scope face at 100-120 feet per second. The 20-kc signal is separated so that each cycle can be read individually. From the selector panel the timing signal is run to the EIvIR discriminator, and the output is run to a neon bulb driver. This driver unit (Fig. 6) and circuit (Fig. 7) will amplify a 5- to 10-millivolt signal to 100-125 volts. For each cycle a neon bulb (NE-66) in the Fastax camera will light, making marks on the edge of the film (Fig. 13B), thus giving an accurate time base. Credit for final design of this neon bulb driver goes to John McMillin, Division The Fastax, which pulls film at 100-125 feet per second, has a running time of approximately 1-1/4 seconds, which means that it takes about 1/2 second for the camera to get up to recording speed. To start the camera at the proper time, a trigger system with yery_good _ response and accurate timing was designed (see Figs. 8 and 9). The trigger system works as,, follows: By checking the signal with the scope, a place on the record tape is selected to be used as a trigger signal. A 2-inch piece of white splicing tape is placed on the back side of the record tape at this spot. (Some organizations cut the tapes and insert trigger splices, birt ' damage to the tapes and the possibility of losing channels by this method can be eliminated by using the photocell trigger unit.) A light source (GE-222) furnishes a beam of light. As the white trigger tape moves past this beam, the light is reflected through a lens to a small photocel(ClairexType CL-1). The photocell output is amplified to drive a sensitive relay (Sigma 10,000 sil). The sensitive relay contacts are used to actuate a latching relay that starts a Goose timer, which can be adjusted to give the proper running time and voltage for the Fastax When it is desired to play back and record with an oscillograph recorder, the Goose timer is eliminated and the contacts of the rotary latching relay are used to start the oscillograph. A 2-inch strip of tape is placed on the record tape at the place selected for stopping Comparison of FM Recording and Erase Recording in High Radiation Fields Recording a signal during a nuclear explosion presents several problems. On previous operations in which an FM system was used transients were noticed on the records at zero time. These transients have at times ruined the transducer and their effect on the FM oscillator makes it useless if the desired information occurs at zero time or several milliseconds thereafter. At zero time the FM oscillator is completely blocked, and the time needed for its' recovery leaves a gap in the recorded information when recording is done at close-in stations. Figure 10 is a record of four FM channels used for information on Fizeau. These channels were blocked for 25 milliseconds or more after zero time. Information occurring after this blocked interval is possibly good data. Figure 11 is a record of a blank FM channel with no information recorded. The FM oscillator quit oscillating at zero time, began to oscillate intermittently 6 or 7 milliseconds after zero, but did not oscillate normally for 30 milliseconds. From 30 milliseconds to 134 milliseconds after zero, the FM oscillator operated normally, but from 134 milliseconds to approximately 300 milliseconds the signal was again highly It is assumed that the power supply to this FM oscillator was jarred by the shock wave, causing the later loss of signal. The power supplies for the FM oscillators were mounted on another package 20 feet from the recording package. This package (Fig. 12) was suspended on springs, but the possibility of it hitting the cement floor was so great that a Styrofoam pad about 6 inches thick was placed under the suspended package to cushion it. However, the shelter sustained 6 to 7 inches of displacement during the shot, and the entire shock-mounting system was found to be insufficient.

Page 14

1A( CA u txrwit man' AT 1014. Fig. 7 -- Circuit for driver unit

Page 19

Fig. 12 -- Timing and oscillator power supply package Transients are present on erase-type channels, but the transient-effect time on the erase system amounts only to a maximum of 10 per cent of that which arises in the FM system. Channels terminating with a resistance at the transducer end and not using* 1-ohm resistor shunted across the recorder head are affected by transients. Figure, 13A is a record made during zero time of four erase-type information channels used on Fizeau aiid, terminating with a resistor at the transducer end. -4<o: These channels monitored firing sets which were triggered at zero time minus 5 seconds to check discharge, charge time, and voltage just before the shot (see Fig. 13B). These channels also furnished proofs that the recording system was working properly just before shot time. Note that the transient effect at zero time on the FM channels (Fig. 11) has been reduced on the erase channels (Fig. 13). Erase-type channels using the plug-in units (Figs. 1 and 2) have no transient effects. Figure 14 is a record at zero time of four channels using plug-in units: Fig. 14E - Timer contacts operating normally at zero time plus 13 secohds The records from Fizeau indicate that: 1. The FM system (Fig. 11) is affected by transients, but will recover; the recorded information after zero time, plus 25 milliseconds will probably be good

Page 20

Fig. 13 -- Transient effect erase-type channels terminating with a resistance at the transducer end

Page 22

2. The erase system with a resistance at the transducer end is affected for approximately 1/2 millisecond after zero time. 3. The erase system using the plug-in unit with a 1-ohm resistor shunting the head is unaffected by transients. 1. The chances of obtaining data in the fireball or its vicinity are much more favorable with the erase system than with the FM system. 2. Before attempting to use the FM system for close-in recording, the system should be studied for component vulnerability, and the system should be designed to work in the fireball environment. This would be a development project which would require time not presently 3. Transducers commercially available for the FM system will not adapt to -the erase system, nor will they operate in the environment of close-in recording. Development of transducers for the erase system will require time; and if the FM system is to be made practical for close-in recording, transducers for this system would also have to be developed.