On the Calculation of a Magnetic Tape Storage Device for a Digital Recording System
Abstract Generated abstract
This paper addresses the design calculation of a magnetic tape storage device for a digital recording system intended to prepare continuous experimental electrical signals for input to a digital computer without modifying the computer itself. It formulates relations for selecting tape length, tape speed, recording density, and the number of tape zones as functions of sampling frequency, channel count, signal bandwidth, and experiment duration, using the sampling condition associated with Kotelnikov’s theorem. The authors apply these expressions to a system including a two-channel switch, an analog-to-digital converter, and a magnetic tape storage unit, and present calculated parameter values for several sampling frequencies and recording durations.
Full Text
ON THE DESIGN OF A MAGNETIC TAPE STORAGE DEVICE FOR A DIGITAL RECORDING SYSTEM
A. V. Trikhanov, A. N. Osokin, P. P. Grigoriev
(Presented by the scientific seminar of the Department of Computer Engineering)
At present, great importance is attached to the use of electronic digital computers for processing the results of physical experiments [2]. For the processing of large data arrays arriving in continuous form, it is necessary to automate the preparation of information for direct input into a digital computer. Recording experimental data on punched tape and punched cards can be carried out only for slowly proceeding processes, owing to the low speed of perforators. Recording experimental data on standard magnetic-tape storage devices (MTS) is a bulky and expensive means; it is possible only at the operating rate of the MTS. There are various ways of solving this problem, described in [3–5]. The existence of many variants is due to the different circumstances and requirements imposed on storage devices by experiments.
For the broad and successful use of universal electronic computers in the study of various kinds of analog signals, in addition to fixing the signals on one or another carrier, it is necessary to have convenient means for inputting them into a digital computer. The information-input facilities available in domestic digital computers do not always satisfy the requirements determined by the specific nature of the signals being processed. This applies, in particular, to electrical signals with a wide frequency and dynamic range.
In addition to the purely technical characteristics of the input apparatus, it is necessary to ensure the “inviolability” of the digital computer. This is especially important in those cases when the laboratory or department conducting the research does not have its own digital computer, but uses the services of a computing center. Therefore, signals can be input into a digital computer through external memory, namely through a magnetic-tape storage device, since in this case no modifications of the digital computer are required.
The digital recording system must provide all necessary conversions and record the signal in digital form on magnetic tape (MT), which can then be mounted in the machine’s magnetic-tape storage device, read, and processed according to a specified program. A block diagram of the digital recording system for continuous electrical signals, developed at the Department of Computer Engineering of TPI, is presented in Fig. 1. This system includes
included are: a two-channel switch for continuous electrical signals, an analog-to-digital converter (ADC), and a magnetic-tape storage device (MTSD).
Fig. 1. Block diagram of the digital recording system
When designing the system, it is necessary to determine the length \(l\) and the tape travel speed \(v\), the recording density \(P\) on the tape, and the maximum number of zones as a function of the parameters of the signals fed to the ADC and of the experiment time \(t_{\mathrm{e}}\). The following expressions may be used to calculate the MTSD:
\[ v \leq \frac{l}{t_{\mathrm{e}}}, \]
\[ v \leq \frac{f_{\mathrm{k}}}{P}, \]
where \(f_{\mathrm{k}}\) is the sampling frequency in the ADC.
In accordance with Kotelnikov’s theorem [1], \(f_{\mathrm{k}} = 2 f_{\mathrm{c}}\), i.e., twice the limiting frequency of the signal taken from the sensor. Taking into account the number \(N_{\mathrm{k}}\) of switch channels, \(f_{\mathrm{k}} = 2 N_{\mathrm{k}} f_{\mathrm{c}}\).
It follows from the preceding that
\[ l \geq \frac{2 N_{\mathrm{k}} f_{\mathrm{c}} t_{\mathrm{e}}}{P}. \]
With a constant initial zone number, the maximum zone number \(N\) can be determined from the following relation:
\[ t = \Delta t_{\mathrm{r}} + \Delta T N + \Delta t_{\mathrm{t}}, \]
where \(t\) is the tape travel time from beginning to end;
\(\Delta t_{\mathrm{r}}, \Delta t_{\mathrm{t}}\) are, respectively, the tape acceleration time and braking time.
\(\Delta T\) is the time of travel of a zone section.
The time \(\Delta T\) is equal to:
\[ \Delta T = \Delta T_{\mathrm{zn}} + \frac{E}{f_{\mathrm{k}}} + 2 \Delta T_{\mathrm{p}}, \]
where \(\Delta T_{\mathrm{zn}}\) is the time for passing the zone number;
\(\Delta T_{\mathrm{p}}\) is the time for passing the interval between the end and the beginning of adjacent zones, and the interval between the zone number and the beginning of information recording;
\(E\) is the number of lines in a zone (for the M-220 M computer, \(E = 24576\)).
Because the quantities \(\Delta t_{\mathrm{r}}, \Delta t_{\mathrm{t}}, \Delta T_{\mathrm{zn}}\), and \(\Delta T_{\mathrm{p}}\) are small in comparison with \(\Delta T N\), they may be neglected in the expressions given above. Taking this into account,
\[ N = \frac{t}{\Delta T} = \frac{f_{\mathrm{k}} t}{E}. \]
It is clear that such an estimate is somewhat too high. Let \(t = 100 k\), where \(k = 1, 2, 3, \ldots\); then
\[ N = 4.15\, k f_{\mathrm{k}}. \]
Table 1 gives the calculated data for \(N, v, P, \Delta T\), the length of a magnetic-tape zone section \(\Delta l\), the experiment time \(t_{\mathrm{e}}\), and the tape length \(l\) for three values of \(k\) and ten values of \(f_{\mathrm{k}}\). Tape travel speeds of \(0.06\ \mathrm{m/s}\) and \(0.7\ \mathrm{m/s}\) were selected with allowance for the maximum recording density \(\left(14.3\ \frac{\mathrm{pulses}}{\mathrm{mm}}\right)\).
Table 1
| \(f_k\), kHz | \(\kappa\) | \(N\) | \(v\), m/s | \(P\), μm/mm | \(\Delta T\), s | \(\Delta l\), m | \(t_e\), s | \(l\), m |
|---|---|---|---|---|---|---|---|---|
| 0.2 | 1 | 0.8 | 0.06 | 3.3 | 123 | 7.4 | 123 | 8 |
| 0.2 | 2 | 1.7 | 0.06 | 3.3 | 123 | 7.4 | 246 | 15 |
| 0.2 | 3 | 2.5 | 0.06 | 3.3 | 123 | 7.4 | 369 | 23 |
| 0.4 | 1 | 1.7 | 0.06 | 6.7 | 61.5 | 3.7 | 123 | 8 |
| 0.4 | 2 | 3.3 | 0.06 | 6.7 | 61.5 | 3.7 | 246 | 15 |
| 0.4 | 3 | 5 | 0.06 | 6.7 | 61.5 | 3.7 | 308 | 19 |
| 0.8 | 1 | 3.3 | 0.06 | 13.7 | 30.8 | 1.8 | 123 | 8 |
| 0.8 | 2 | 6.6 | 0.06 | 13.7 | 30.8 | 1.8 | 216 | 13 |
| 0.8 | 3 | 10 | 0.06 | 13.7 | 30.8 | 1.8 | 308 | 18 |
| 1.2 | 1 | 5 | 0.7 | 1.7 | 20.5 | 14.5 | 103 | 73 |
| 1.2 | 2 | 10 | 0.7 | 1.7 | 20.5 | 14.5 | 205 | 145 |
| 1.2 | 3 | 15 | 0.7 | 1.7 | 20.5 | 14.5 | 308 | 218 |
| 1.6 | 1 | 6.6 | 0.7 | 2.3 | 15.4 | 10.4 | 108 | 75 |
| 1.6 | 2 | 13.5 | 0.7 | 2.3 | 15.4 | 10.4 | 215 | 150 |
| 1.6 | 3 | 19.8 | 0.7 | 2.3 | 15.4 | 10.4 | 308 | 214 |
| 2 | 1 | 8.3 | 0.7 | 2.9 | 12.3 | 8.6 | 111 | 78 |
| 2 | 2 | 16.6 | 0.7 | 2.9 | 12.3 | 8.6 | 209 | 146 |
| 2 | 3 | 24.9 | 0.7 | 2.9 | 12.3 | 8.6 | 308 | 216 |
| 4 | 1 | 16.6 | 0.7 | 5.7 | 6.2 | 4.3 | 106 | 74 |
| 4 | 2 | 33.2 | 0.7 | 5.7 | 6.2 | 4.3 | 211 | 146 |
| 4 | 3 | 49.8 | 0.7 | 5.7 | 6.2 | 4.3 | 308 | 215 |
| 6 | 1 | 24.9 | 0.7 | 8.6 | 4.1 | 2.9 | 102 | 73 |
| 6 | 2 | 49.8 | 0.7 | 8.6 | 4.1 | 2.9 | 205 | 145 |
| 6 | 3 | 74.7 | 0.7 | 8.6 | 4.1 | 2.9 | 308 | 218 |
| 8 | 1 | 33.2 | 0.7 | 11.4 | 3.1 | 2.2 | 106 | 75 |
| 8 | 2 | 66.4 | 0.7 | 11.4 | 3.1 | 2.2 | 208 | 148 |
| 8 | 3 | 99.6 | 0.7 | 11.4 | 3.1 | 2.2 | 310 | 220 |
| 10 | 1 | 41.5 | 0.7 | 14.3 | 2.5 | 1.7 | 105 | 72 |
| 10 | 2 | 83.0 | 0.7 | 14.3 | 2.5 | 1.7 | 208 | 141 |
| 10 | 3 | 124.5 | 0.7 | 14.3 | 2.5 | 1.7 | 313 | 211 |
References
-
B. V. Anisimov, V. N. Chetverikov. Fundamentals of the Theory and Design of Electronic Digital Computers. Moscow, “Vysshaya shkola,” 1970.
-
V. T. Glushkov. Prospects for the application of computer-technology equipment. In: Interorgtekhnika—66. Moscow, Ontipribor, 1966.
-
N. M. Kazachuk et al. A device for recording experimental data on magnetic tape. In the collection Computing Systems, issue 35, Novosibirsk, “Nauka,” 1969.
-
V. M. Kevlishvili et al. An input device for the M-20 computer. In the collection Problems of Technical Operation of Computers, issue 2. Publishing House of the Academy of Sciences of the USSR, 1969.
-
A. I. Stapilovsky et al. A digital spectrum storage unit on magnetic tape with processing on a computer. In the collection Problems of Technical Operation of Computers, issue 2. Publishing House of the Academy of Sciences of the USSR, 1969.