Sanyo ICC-1141 Electronic Calculator

This machine turned out to be a bit of a surprise. This machine was offered for sale, and was almost bypassed as it seemed like it was perhaps "too new" to be included in the Old Calculator Museum's collection at first glace of the photos. It was decided to take a risk that perhaps its appearance might be deceiving, and the calculator was purchased. When it arrived, it was in absolutely beautiful condition. When the calculator was given its usual inspection inside and out before powering it up, it was found to be from Sanyo's third-generation of electronic calculators, utilizing Japanases-made small and medium scale Metal Oxide Semiconductor (MOS) integrated circuit technology from the latter part of the 1960's.

Sanyo, until 2009 when a majority stake in the company was acquired by Panasonic, was in the business of providing a wide range of products since its founding in 1947. The company was created somewhat as a spin-off of Matsushita Electric Industrial Co., Ltd. As part of the post-war agreements with the allies, some of the larger Japanese corporations were required to be broken up, creating smaller corporate structures. The founder of Sanyo, Mr. Toshio Iue, was the brother-in-law of one of the original founders of Matsushita, and as a result of the break-up orders, Mr. Iue was provided an old Matsushita manufacturing plant in order to form his own company, which he named Sanyo. Sanyo's initial product was a small generator that was powered by the rotation of a bicycle's front tire. The generator would the headlight on the bicycle, providing a path of light for night-time riders, as long as they were moving at a moderate speed. Sanyo initially experienced modest, but steady growth, moving into tremendous growth in the late 1960's and 1970's as it vastly diversified its businesses. Sanyo became engaged in a large number of consumer businesses, starting with home appliances, then moving into consumer electronics. The company then added industrial products, such as power systems, components for transportation systems, and then into services, including logistics, engineering services, and much more.

It appears that Sanyo began its push to develop an electronic calculator in the 1965 timeframe, though details of its early history in this realm are sketchy. At this point, it is assumed that the first Sanyo electronic calculator created for general sale was the Sanyo SACOM DK-102, which came to market sometime in the 1966-1967 timeframe. This was a mostly discrete-transistor calculator, although it did use a number of early Mitsubishi DTL/TTL small-scale integrated circuits on one of its circuit boards. The design of the DK-102 is very reminiscent of some of Sharp's second-generation transistorized calculators, though it is certain that it was not a copy or adaptation of one of Sharp's calculators.

Sanyo First-Generation Electronic Calculator, the SACOM DK-102
Image Courtesy of Serge Devidts, Calcuseum

At this time, it isn't known if there were any electronic calculators marketed by Sanyo prior to the DK-102, nor if there were any other machines that were marketed using this first-generation design. If anyone reading this happens to know anything about Sanyo's early electronic calclators, please contact the Old Calculator Museum by clicking the Email button at the top of this exhibit page. One thing is certain, though; Sanyo very quickly moved away from discrete transistor circuitry to early Japanese-made MOS integrated circuit-based designs. The use of ICs made the machines less complex to build, more reliable, and consumed less power, not to mention making the machines smaller and lighter.

An example of a Sanyo Second-Generation Electronic Calculator, the ICC-162
Image Courtesy of Rainer Fredrich

Examples of the Sanyo's second-generation MOS IC-based electronic calculators were machines such as the ICC-121 (with ICC standing for "Integrated Circuit Calculator), ICC-141, and ICC-162. A hallmark of the second-generation calculators was that they were constructed in a modular fashion, with multiple circuit boards that plug into backplane with the interconnections between the boards. Another characteristic of the second-generation calculators was the use of unique seven-segment displays that are lit with incandescent lamps. Because of the use of incandescent lamps, the display system is static, each digit has its own decoding and drive circuitry rather than using a multiplexed display technology that shares the decoding and drive circuitry, and time-shares it among each of the digits of the display, scanning the digits one at a time at a rate fast enough that the human eye perceives the display as continuous.

The third-generation calculators tended to aggregate the circuitry onto a couple of larger circuit boards that are interconnected by a small backplane, with a less-modular design that gave the machines a sleeker appearance, but in general used much the same IC technology as the second-generation machines. Also, though not exclusively, the non-multiplexed incandescent seven-segment display was replaced by individual multiplexed seven-segment gas-discharge display tubes.

Profile view of Sanyo ICC-1141

This ICC-1141 appears to have been manufactured in the mid-1968 timeframe (based on somewhat cryptic date codes on some of the integrated circuits), and uses early Japanese-design MOS IC technology, with the majority of the ICs in the calculator manufactured by NEC. The ICC-1141 is virtually identical to another machine in Sanyo's long line of electronic calculators, the ICC-1411, with the only visible difference being that the ICC-1411 has a "display test" button that lights all segments in all of the digits to verify that all of the display tubes are working properly. This feature is a carry-over from Sanyo's second-generation of electronic calculators.

Sanyo ICC-1141 Circuitry

The IC's used in the machine contain from between a few logic gates or flip-flops, to higher-level integration functions such as multi-bit shift-registers. Compared to earlier all-transistor technology, these early IC's made it possible to shrink down the size of a calculator while providing the same functionality as the larger all-transistor machines. Most of the logic of the ICC-1141 is implemented using IC devices, but there is still a bit in the way of discrete diode/resistor gates and transistor-based logic in the machine. At the time the machine was made, a careful balance had to be struck between the cost of new technology like integrated circuits, versus the extra space and power required by discrete technology. As integrated circuit technology advanced, the cost of IC's came down, and by later in the 1960's, it was possible to build a calculator with a minimum of discrete component logic, with the discrete devices relegated to functions such as power supply regulation and display drivers. By 1970, Large Scale Integration (LSI) IC devices appeared on the scene, and the whole world of calculator technology completely changed... what used to take from between 70 to 150 small-scale IC's could be crammed into four large-scale devices. The age of the large, heavy, A/C-power-hungry desktop calculator was nearing its end.

The Sanyo ICC-1141 Model/Serial Tag

The 1141 is a very well-built machine. The logic of the ICC-1141 is contained on two circuit boards that are oriented horizontally in the machine, stacked atop each other. The boards plug into dual edge-connectors with a hand-wired backplane located at the rear of the chassis to provide interconnection between them. The boards are secured in the chassis by a fairly complex plastic casting that provides spacing between the boards, as well as stiffening for the fairly large circuit boards. The majority of the integrated circuits are made by NEC, from their early MOS µPD10-series small-scale integration(SSI) devices in TO-100 and TO-101 packages, and µPD100-series MSI in dual-inline packages. There are also a few Mitsubishi small-scale bipolar TTL (Transistor-Transistor Logic) IC's sprinkled about in the circuitry. The pheonolic circuit boards have etch on both sides with jumper wires to provide connections on the component side of the board that couldn't be made via etched connections because of real-estate constraints. Feed-thrus are made between matching pads on each side of the board, with snippets of wire soldered through the holes to assure reliable connections between the component and wire sides of the boards.

The Early Seven-Segment, Gas Discharge Display used in the Sanyo ICC-1141

The 1141 uses fourteen seven-segment gas-discharge display tubes for its display. The display board connects to the main electronics via two edge connectors and cables. The display on this machine is unusual for the time. Many machines of the late 1960's used Nixie tubes for display simply for lack of other relatively low-cost display technologies. It took some time, but in the early 1970's other technologies became available at low-cost (such as vacuum-fluorescent and Burroughs Panaplex planar gas-discharge displays, as well as Light-Emitting Diodes (LED), Nixie tubes eventually fell from favor. One of the first technologies to challenge Nixie tubes was the seven-segment gas-discharge display tube. This type of display tube worked on a similar principle as the Nixie Tube. Nixie tubes worked by having cathodes shaped in the form of the individual digits, arranged in a stack. Placing a high voltage (between 170 and 200 Volts) across one of the digit-shaped cathodes, and a see-through mesh anode situated in front of the stack of digits caused a mixture of Neon and other gases inside the tube to ionize around the cathode, causing the gas around the cathode to glow an orange-red color. The tubes used in the Sanyo ICC-1141, instead of having stacked digit-shaped cathodes, used seven cathodes arranged on a black surface, with a similar mesh screen in front of the segment array acting as the anode. The segments were arranged in the form of a squared-off 8, such that lighting a combination of the segments could form any digit from zero through nine. This type of display was significantly easier to manufacture, and though it resulted in a more "electronic" looking display than the natural-looking digits of Nixie tubes, the cost advantage was well-worth it.

The gas-discharge seven-segment display tubes work in a similar fashion to a Nixie tube, but rather than having ten individual electrodes shaped as digits, seven electrodes are arranged in a segmented '8' pattern, allowing combinations of the segments to be lit to form any digit from zero through nine. The tubes are filled with a similar gas to that used in Nixie tubes, that, when an appropriate voltage was delivered to a segment electrode, will cause the segment to glow. Lighting the appropriate pattern of segments forms any digit from zero through nine. Each tube also contains an electrode for a decimal point, situated below and ever so slightly to the right of each digit.

Two incandescent indicators are situated at the left end of the display panel, one lights to indicate a negative number (when all digit positions are occupied), and the other lights when an overflow condition exists. An unusual feature of the machine is an audible alert that sounds when an overflow condition occurs. In an office environment, this seems like it would be an extremely annoying feature. Three incandescent indicators are used to display the status of the machine. These indicators are located on the keyboard panel, behind red jeweled panels. One indicator, labeled "M", lights up when the memory register has non-zero content. The other two indicators light up after the first operand of multiply or divide calculations have been entered, to remind the user of the operation in progress. An "X"-labeled indicator lights after the [X] key has been pressed, and a ÷ symbol denotes the indicator that lights after the [÷] key has been pressed. After the operation is completed, the indicator for the funcion is extinguished. Some early Sharp electronic calculators provided a similar function, but integrated the lamps into the key assembly such that the X or ÷ symbol in the keycap would glow when the operation was pending.

Along with using an unusual (for the time) display technology, the 1141 also provided a number of other functions that were ahead of its time. The display system provides leading-zero suppression, which was quite uncommon at the time. Leading-zero suppression really only became a common feature once large scale IC technology was available -- a feature that was relatively easy to squeeze into a calculator chipset, but required some additional (and cost-increasing) circuitry in less-highly integrated calculators. The blanking of leading zeroes makes reading numbers in the display easier, as the eye isn't distracted by all the extra zeros. Another unusual feature is that, if the number on the display is negative, the "-" is displayed using the digit display tubes rather than with a separate indicator, as was very common on calculators of this vintage. The machine also provides an automatic summation mode, activated by a push-on/push-off key on the keyboard (with a 'sigma' nomenclature) that automatically accumulates the results of multiplication and division operations into the memory register. The 1141 has a single memory register that operates as an accumulator, with [M+] and [M-] keys adding or subtracting the number in the display from the memory register, and placing the result back into the memory register. The [CM] key clears the memory register, and the [RM] key recalls the content of the memory register to the display. The Sanyo ICC-1141 also has a constant function, activated by a push-on/push-off key labeled [K]. When the constant function is turned on, a constant multiplier or divisor is retained, and can be used for repeated operations with differing multiplicands or dividends, or the constant can be repeatedly applied for use in power sequences.

The Keyboard Assembly of the Sanyo ICC-1141

The keyboard panel of the machine is populated with the expected operator keys, with a few minor deviations from the norm. The keyboard uses magnetic reed switches, making for extremely smooth and reliable operation. The Clear Entry function, usually [CE] on many calculators, is [CK] on the 1141, for "Clear Keyboard". This key clears any number that has been entered thus far, allowing for correction of input errors. The [CA] key (for Clear All) does the obvious, clearing the machine (with exception of the memory register), making it ready for calculations. The [RC] key swaps operands on multiplication and division operations. A series of five pushbuttons occupy the left-hand side of the keyboard panel. These buttons, labeled "0", "1", "2", "4", and "8", select the fixed decimal point location in the display. Pushing one of these keys will cause any previously selected key to be deselected, and the new key to be 'latched' down. A slide switch selects the rounding mode of the machine, with truncate, and 5/4 positions.

The ICC-1141 has a number of quirks which seem to be common on early small-scale integrated circuit calculator implementations. First off, any multiplication or division problems will give false overflow conditions if either operand contains 14 significant digits. For example, 99999999999999 X 1 will result in an overflow condition, even though it is within the fourteen digit range of the machine. Also, pressing keys on the keyboard while the machine is busy calculating will cause an overflow condition to occur most of the time. Sometimes, though, the result will be an incorrect answer, which is a somewhat serious bug in the machine's logic. The machine also gets somewhat confused about the value of zero, allowing the existence of -0. Performing 1 [-=] 1 [+=] will result in -0. An odd quirk that I've not noted in other calculators is that calculations using zero as the multiplicand causes the machine to go into a very strange state, that can only be cleared by power-cycling the machine. As with many other machines of the era, division by zero causes the machine to get confused, but in this case, pressing the [CA] key will return the machine to normal operation. During calculations, the 1141 lights all decimal points, and leaves the displays active, causing quite a light show on longer calculations. The machine is somewhat slow compared to other machines of similar vintage, with thirteen 9's divided by 1 (remember, the machine can't handle multiplication or division with 14-digit operands) takes a little over 3/4 second to perform.

Sanyo marketed another machine called the ICC-1121 that was identical in all aspects to the 1141 except (as expected) that it provided 12 digits of capacity versus the 14 digit capacity of the ICC-1141.