RECOLLECTIONS ON THE MADAS CALCULATOR
(Manufactured by H.W.Egli, of Switzerland)

by
Ray Mackay

History


MADAS JUNIORDisclaimer: The following article is not intended as an all encompassing guide to the MADAS models 'L' through '20BTZG' but to provide enthusiasts with a guide to 'where to look for what'. The various illustrations are taken from copies of original drawings from the archives of H.W.Egli. The original drawings would be held by the descendents of Hans W Egli. I expect copies will also be held by Peacock Bros., of Melbourne. I do not have the complete set but do have twenty-three out of a known thirty-six pages; undoubtedly there are more. Any assistance that can be given in obtaining the missing pages would be appreciated.

THE AUTHOR'S STORY

The author began working in the calculating machine industry after emigrating from the U.K. in 1954. His first occupation in the U.K. being as a trainee in a Radar and Signals Establishment in the county of Lancashire (now Cheshire). As with all trainees in the 1950's the apprentices were really lackies given the more mundane and dirty tasks to complete. The more interesting tasks being left to those with the supposed skills. Experience was gleaned from watching the experts at work when the opportunities came. The author's dreams were of an electronics nature not of becoming a 'mechanic'.

In Australia he was first employed as a scales apprentice, a job he hated and which lasted only months. From there he sought employment as a trainee mechanic at Peacock Bros. in Bourke Street, Melbourne. Peacock Bros. being the Australian Importer of the Felt and Tarrant Comptometer, the MADAS and the Walther range of office machines.

As far as mechanical training was concerned one got much of the same treatment as in the U.K. and trainees were given the less interesting jobs until a situation came about where one was promoted up a notch.In the author's case he was brought up by the proverbial boot straps by an unlikely string of events. Firstly a skilled technician left to take up another position. Simultaneously a senior trainee decided to return to England, thus a couple of vacancies required filling promptly. The Head technician, or mechanic, was quite elderly and only had a couple of years of working life to retirement and an opportunity was born. The author was forced into a rapid learning curve 'to learn the ropes from the ground up'. Mr. Fitzhenry was given the task of knocking all his knowledge into the head of an eighteen year old whipper snapper. A task he accomplished in a short time to the relief of the management. When Mr. Fizthenry retired the author became the "service manager' for Peacock Bros., at a very young age.


THE VARIOUS MACHINES

The three main products sold by the office machines section of Peacock Bros. were the Comptometer, MADAS and Walther, although there were other machines at times. For the purpose of this exercise we concentrate on the MADAS from the above three indicated Brands. Some details of the other units being treated elsewhere. The MADAS was manufactured by Hans.W.Egli of Switzerland.

H.W.EGLI, SEESTRASSE 356 / 58, ZURICK 2 / 38, SWITZERLAND.

MADAS - 1908The author knows little of the Egli factory in Switzerland. In spite of frequent ambitions to visit the establishment he never satisfied this desire. His first experience and knowledge of H.W.Egli came from the metal name plates of Millionaire Calculators that the author started working on in 1954. Although I had long forgotten these monsters I understand from Erez Kaplan's site (http://www.webcom.com/calc) that these machines were first produced in 1895. Egli name plates on many MADAS machines we worked on indicate a starting date earlier than 1920.

MADAS - 1921The machines made in 1920 were not as large as the Millionaire however, they were still large, but quite advanced. Even in 1954 these machines were considered monsters and were picked up from their owners by van (Karrier Bantam Van services) and brought in for service. In the earlier parts of the author's career the MADAS were considered sophisticated machines and only selected personnel were allowed to work on them.
CALCULATING MACHINE TYPE

MADAS BTGThe first simple 'L' machines had all four functions but with the control of carriage shifts and various operations being a composition of individual manual operations. One can trace the evolution of the various MADAS models by the mechanism. The 'L' model was simple in concept and mechanically. A machine providing semi-automatic division came later in the evolutionary process. This evolutionary process culminating in the the model 20BTZG. The 'B' merely indicating a later Automatic multiplication mechanism. All the machines work in cycles timed by a master cam shaft. This cam shaft was first driven by a hand crank and then later by an electric motor installed underneath the unit. The motor having a clutch mechanism that also allowed operation by hand crank in the event of electrical failure.

The cam shaft determined the timing of each operation during each cycle of the drive gear. Each standard operation was completed in this one cycle. At first each operation was selected manually by the operator as in the case of early and less complex models. With the introduction of machines with semi-automatic division the machines performed several rotations of the cam shaft before coming to a stop which required further operator intervention. The operator then moved the carriage and repeated the operation until the process was completed. Thus the number of cycles was dictated by the complexity and existing conditions in the calculation.


EVOLUTION OF THE MACHINE

The forgoing paragraph indicated an evolutionary process by H.W.Egli that commenced with their earliest machines and progressed through to their most sophisticated and complex model the BTZG. Following through the mechanics of the machines this evolutionary process becomes most evident as controls and levers became interwoven in a complex manner around a basic concept. Levers took on weird and wonderful shapes dictated by where they could be located to interlink through the existing framework, mechanism and construction. You will observe this as we progress. The first two big changes to machines by H.W.Egli will be ignored, these being the change from the Millionaire to a MADAS with the carriage built into its box like structure. The Millionaire was the only mechanical machine I know of that actually multiplied by a single action multiplication mechanism. The change from the large box like structure of the first MADAS machines into the models with a moveable carriage at the top of the machine, this was not such a radical change as the mechanical principles were similar and the general mechanical form remained.


DEFINITION OF REGISTERS

MADAS 20BTZGThe method of defining keys and registers in this work will adhere to the protocols used by H.W.Egli as delineated in the illustration on the right. The registers in reference to the illustration on the right will be as follows...

0 Multiplication Register
III Keyboard or Entry Register
II Counter or dividend register
I Main or Product Register
IV Triplex or Accumulation Register

It should be noted that not all models will have every register and you need to use the hyperlinks throughout the text to jump to illustrations indicating particular features and particular mechanisms. In relation to levers and operations performed by such levers reference to the detailed picture will be essential; at least if you wish to follow all the details.

DEFINITION OF KEY NAMES

The down arrow key transfers data from the Main Register to the multiplication register
The key with two dots, one above the other transfers data to the left of the main register, ready for division.
The keys marked I II and III clear the Main register, the counter register, the dividend register and the keyboard register respectively
The 'DIV' key is self explanatory
+ and - are also self explanatory
Carriage shift keys indicate direction of movement indicated by arrows.
The long Bar starts an automatic multiplication.
The key marked 0 clears the multiplication register.
The reason for the small bar to the left of the long bar escapes my recollection.

PinionsMETHOD OF ENTERING INFORMATION

All information concerning data for operations is entered into the keyboard. The keys had a latching mechanism that prohibited more than one key being locked down in the same column. The key marked III allowed for rapid clearance of numbers before a new entry was put into the keyboard. Under the keytops are metal keystems that slide into flat metal strips, two to each column; one for odd keys and one for even keys; the arrow points to the pair of slots. These flat strips in turn move a sliding pinion, mounted on a square shaft, into a position adjacent to the equivalent number of teeth on a Leibniz (stepped) drum.

POSITIONING KEY SELECTED PINIONS

Leibniz WheelReference to the illustration on the left will indicate, to those not familiar with such things, just what a stepped drum looks like. This stepped drum or Leibniz wheel is similar to a fixed pinion gear excepting its design incorporates teeth that are extended. Each step being incremental in length but stepping so as to make less pinion teeth accessible to operate the gear to be driven. A somewhat clumsy description but it fits the function quite suitably. Although the Leibniz wheel in the illustration is not in the context we desire, it is not hard to imagine a pinion positioned on the square shaft above our Leibniz wheel sliding over the top of various spur teeth and turning the square shaft on which the pinion is mounted by a required 36 degrees of movement representing 1/10th turn for each tooth. This surely is a description of the required motion of such a pair. Indeed have we not described the use of a stepped drum, or Leibniz wheel, to decide the number of degrees turn of a shaft in increments of one tenth.

THE POSITION SO FAR

Sketch 1A selected key is entered on the keyboard. A stud on the lower end of the keystem slides down the keybar ramp into a notch in the keybar. This action positions a sliding pinion of a square shaft over the selected position equating with the key on a 'Leibniz Wheel'. As the drive shaft turns the Leibniz wheel actuates the pinion turning it by 36 degrees for each spur in the selected position (36 degrees being 1/10 of a turn). On the top end of the square shaft a fixed pinion drives the mechanism that couples the carriage. This mechanism consists of another square shaft with a double edge pinion as indicated on the left. This pinion/s can slide upwards or downwards as actuated by a lever.

In the upward position the motion turns a numeral wheel pinion in one direction, in the downward position in the opposite direction. We now have a situation where a gear can increment or decrement a numeral wheel by a selected number. Indeed we have accumulation from one to nine and must now consider transferring a carry to the next column as our wheel completes a cycle.
TRANSFERRING DATA FROM PINION SHAFT TO STORAGE REGISTER

The process of transferring data from the pinion shaft to the carriage is illustrated on the left, as is the coupling to the numeral wheel. Behind the numeral wheel you will note a clearance rack and also a cam marked 13. This cam triggers a carry as the numeral wheel goes from nine to zero or zero to nine. It is important to note the carry can be a positive or negative carry as this becomes important later on when we get to the division process.

REMOVAL OF CARRIAGE

Some readers may have a machine whereby they wish to remove the carriage to examine the workings described in this text. The carriage is removed by tripping a small latch to the left of register three. The Carriage then lifts out with a slight tilt upwards towards the back. Replacing the carriage requires an understanding of a few points. If the machine is a 'Z' model the carry over mechanism is first lifted out of the counter register gears. Using a piece of string looped over the carriage clearance racks the racks are pulled into the clearance position. This means that the tens transfer cam followers are held close to the cams and will not foul the carry over arms. The Z-mechanism is then rested just on top of the notch in the negative carry gears and the carriage placed on in the extreme left-hand position (all the while holding the clearance racks out with the string). "T" models require far more attention as one must place the two markers on the carry over cam shaft adjacent to each other and make certain the long actuating drum is in its bias notch prior to refitting the carriage. The move really needs practice however once learnt is never forgotten.

THE TENS TRANSFER MECHANISM

Carry MechanismWe earlier mentioned a carry lever that was actuated by the 'cam follower' tracking the position of the numeral wheel. As the wheel went from nine to zero or visa versa the cam follower pushes in a lever as in fig 6. This lever in turn moves a sliding pinion on the square shaft that drives the carriage numeral wheel mechanism. When the cam follower pushes in the carry lever this pinion moves into the path of an oscillating carry arm, the movement timing of which is controlled by the main cam shaft via a train of gears. When this pinion moves it to the path of the carrying detent as the detent oscillates it transfers the carry into the next higher order. Note :- this movement takes place at a time when all other action cycles to the carriage have been completed and thus there is no possibility of conflict. Fig 5 shows the lever and cam shaft. The carry lever is then restored as a 'V' shaped projection of the carry lever and is kicked out by a pin in the cam shaft drum 112a. The lever position is then retained by the bias ball clip at the top of the carry lever.

To recapitulate. We now have a process whereby we can enter data into a keyboard. Transfer this data from the keyboard to the display on a movable carriage. As the display turns from zero to nine or nine to zero we can time a carry into the next higher order. This process giving us a mechanism by which we can add up to ten in one column and take care of a carry. It take no stretch of the imagination to realise if we placed ten such mechanisms (building blocks) side by side we would have available a ten column by ten digit adding machine capable of displaying the result. We must further recognise that as multiplication is merely repeated addition then, by moving the carriage into various positions we actually perform a multiplication, the result depending on the number of machine cycles in each position.

COUNTER NUMBER OF CYCLES IN EACH OPERATION.

CounterThe counter register is quite a simple device in comparison with other areas. It consists of a slotted disk attached to the numeral wheel assembly. A daisy chain rack assembly oscillates to and from and tries to go forward and back each cycle. In the first position, or extreme right hand position the daisy chain rack is adjusted so it always toggles the extreme left wheel. If the position of the slot on the disk of assemblies to the left is open these other wheels can also be toggled.If the slot in the numeral wheel disk is available then the pawl of the daisy chain rack assembly is allowed to drop in to this position thus actuating this position also. This allows for carries in operations of more than ten cycles in any one position. It also provides a means for entering negative carries across the whole counter register. In other words. If the register was clear and you entered a minus operation all counter positions would change to nines.

Carriage moveMANUAL CARRIAGE MOVEMENTS

We have possibly noticed previously how a pair of pinions tied together on a square shaft can be moved upwards or downwards to engage the drive pinion on another mechanism. By so doing the direction of the turn is forward or reverse depending on the gear engaged. This is best illustrated in the lower illustration to the right of "the position so far".

ClutchA similar method can be used with a clutch mechanism to cause a shaft to rotate to the right or left. An illustration of this clutch mechanism is to the left. If one or the other of these clutch positions is engaged the rollers on the cam to the right can be made to change position. As these rollers are engaged with a large tooth rack mounted on the carriage the carriage can be made to move to the right or to the left. To do so whilst any other operation was in progress would be a disaster. Preventative measures are taken all over the MADAS to safeguard actions likely to cause conflict. In the case of carriage shift you will note in the illustrations various cam followers are made to lock out all other operations during carriage shifts.

MULTIPLICATION BY REPEATED ADDITION

MADAS 10RI guess every person would know the method of Multiplication by repeated addition however for the sake of completeness I will run through it briefly. When we multiply 12 by 12 we first multiply the first twelve by two, the answer being stored. We step across one position as we are multiplying by tens and multiply by one. The accumulated result being 144. On a mechanical machine such as the MADAS the same logical process is completed mechanically. By moving the carriage one position to the right we effectively multiply by a ten. Two positions to the right and we are multiplying by a hundred and so on. On early machines the process requires one routine process to be completed by an operator, this remained the situation with manual machines such as the 10R (see right-hand inset). Later mechanisms were introduced to the existing manual machines to make automatic multiplication possible

AUTOMATIC MULTIPLICATION

Stepped DrumTo make automatic multiplication position one must first have a storage register to store the number by which you wished to multiply. A sensor mechanism then becomes necessary so that the information in such a register can be 'read'. This is not an insignificant task as we have to not only sense the number we have to also sense the order or carriage position. H.W.Egli dropped back on the old faithful, the Leibniz wheel however in a different form.This new form, which I will call a stepped drum, actually uses the stepped ends of the spurs of the drum to correlate the numeric values in the register '0' with the position of the stepped drum.

Pin DrumA pin drum was also used to sense the number of cycles the machine needed to perform to complete a multiplication. This pin drum has multiple pins. At each carriage shift a different pin is made to protrude. It only being possible for one to protrude at any one time.

It is now essential if I am to convey the process to the uninitiated that a scenario, or picture be painted in your mind. Starting with the carriage in the extreme left hand position. When the Multiplication Key is depressed,this initiates a string of mechanical processes. The part we are interested in at this stage being the pin & pin drum. The Pin drum rotates forward with pin one high. The pin hits the step drum on the step correlated with the number in the multiplier register '0' position one. If the number happens to be a two then the drum drops two tooth positions of the step drum rack. The machine cycles five times and at each cycle the pin drum ratchets back to its starting point by one position. Drum RatchetAt the completion of the two cycles a carriage shift is tripped in and the carriage moves one position to the right. Pin one drops out of play and pin two comes up. In this case we have the one in multiplier register 'O' position 2 (the tens position). The pin drum drops again until the newly raised pin strikes the step drum. The position being only one tooth down. The machine takes one cycle actuating the pin drum back one position. The mechanism senses via a sensing mechanism that no further numeric values are in the multiplier register and either the carriage is returned to its starting point or stopped depending on operator selectable options. The mechanism on the left illustrates the means by which the step drum is ratcheted back to its starting position. Notice the pawl and the detent which latches to stop over operation. Also note the stud behind a pendulum like lever that trips (pointed to by arrow) in the next operation at the end of each recovery back to the zero position.It should be noted many side issues take place however in consideration of size and descriptive skill I have ignored these side issues.

DIVISION BY REPEATED SUBTRACTION

From the above it is obvious that a number of individual processes can be automated mechanically. We illustrated that multiplication could be automated. Historically, on the MADAS automatic division came before automatic multiplication. It was however necessary to first paint a picture of automation in mechanics. Division is merely repeated subtraction using a 9's compliment. The quotient needs to be thrown up into the top left of register one. This is accomplished by the key with the two dots. Pressing this key down instigates a clearance action on all registers followed by a carriage shift to the extreme right. In this position the contents of the keyboard are thrown up into register one. .The keyboard is cleared. The operator now enters a divisor. On pressing the division key, the number in the keyboard is added to register one, however there is a difference. Under the subheading 'the position so far' we described how a pinion engaged at 90 degrees from the top of a gear would turn one direction whereas if this pinion was attached to a pinion that could slide up to the bottom of the gear the direction would be reversed. In other words, how a beveled gear engages from top to bottom of a pinion will decide the direction of the drive. In division this mechanism is reversed and thus the increment becomes a decrement. We take away from the extreme left hand position until a negative carry takes place across the machine. A special division cam then institutes an add back by one followed by a carriage shift. The routine is repeated until such time as the carriage reaches the extreme left position. The counting register now displays the number of cycles that existed in each column; this being the dividend or answer and the remainder remains in register one.

THE USE OF 'Z' MECHANISM TO TRANSFER NEGATIVE CARRY ACROSS WHOLE TWENTY COLUMNS

Later machines introduced a second counter register to the left of register (see 20BTZG illustration) This mechanism extended beyond the normal mechanical control of the main section of the machine. A 'Z' mechanism was introduced so that carries could be executed across the section of carriage beyond the counting and carry levers of the main counter mechanism. An examination of this mechanism is self explanatory and can be viewed under the left front of the carriage whilst operating the machine manually. Its description will be skirted.

AN OVERVIEW

An overview of the carcass mechanism less control parts is enclosed to permit an understanding of the main frame structure of the unit.

Also, several pictures showing the internals of the MADAS 20BTZG have been included.


THANKS AND ACKNOWLEDGMENT

This document is a draft document. It undoubtedly contains materials of a proprietary nature. As far as the author is aware H.W.Egli no longer make mechanical calculating machines and never made any electronic calculating machines. It is not even known if H.W.Egli are still in business as the author lost contact in the 1970's. Copyright exists for some fifty years after a death of the inventor or writer. This document acknowledges the rights of H.W.Egli to all drawings. They are provided merely to enhance an understanding of this brief for educational purposes. The record may also assist in preserving these historic machines
.

The four color MADAS photos inserted at the beginning of this article are courtesy of Eusebio Huélamo.


This document is an historic record and may be copied and released for educational purposes. It may be copied and given to enthusiasts interested in the mechanics and remembered history of the product. It may not be reproduced in whole or part for the purpose of profit.

  Copyright © January, 1998 Ray Mackay All rights reserved ® 1998
HISTORY ALBUM MUSEUMS COLLECTORS RESOURCES

X-Number World