Document aJ9YRdRQboyma4KDQbEQwrDnN

104 ROSS A. MC FARLAND a knob with axis pointing toward the operator and in line with his forearm, a diameter of 1 to 3V2 inches is satisfactory.84 One of the primary determinants of the size of a knob is the torque that is to be exerted upon it. This introduces one exception to the superiority of the fourpointed star shape mentioned above: when small torques are desired (65 to 110 pound-inches), a cylindrical knob 2 inches in diameter is preferable. If it is to be placed facing the operator, a circular knob with a heavy rim should be used since this can be easily grasped with the finger tips instead of the entire hand. To de velop a torque between 130 and 175 pound-incihes, the diameter should be 4 . inches.82 In continuous-adjustment tasks involving the use of handwheel controls, the results of tests with two wheels were compared, one with a radius of 21/* inches ' and a second with a radius of 4Vs inches. Relatively better scores were obtained with the wheel of larger radius at slow speeds, whereas the reverse was true at high speeds. Even a handwheel with an 8-inch radius can be used with slight advantage up to 60 r.p.m., although the maximum rate of turning is lower than with the smaller handwheels. This effect of handwheel size is small, but it appears from these studies that an optimum radius would be about 3 to 41/- inches provided the friction is small (2 to 3 pound-inches). With handwheels in this size range, accu racy in use tends to increase with increasing handwheel speeds up to near the" human limit of turning speed. Maximum accuracy for different handwheels varies. from 140 to 200 r.p.m.85 ...................... ! rf e. Gear Ratio. The optimum gear ratio for tasks in which an operator setsi| a pointer by turning a knob, such as in radio tuning, is approximately IV2 inchesfv of pointer movement for every revolution of the knob.83'88 4, When a lever-type, nonlinear control (variable gear ratio) was tried, it wasf found that subjects responded as though they expected a linear response to theiFcontrol movements. In a continuous-control task, no consistent differences in ac4| curacy were found between tracking with linear and nonlinear levers.78 However,": when large discontinuous movements were being made, and the nonlinear relation ;' ship had not been explained to the operators, there was a marked advantage in' favor of the linear lever. When the nature of the nonlinear lever had been explained and-largdiscontinuous responses were madeTthe errors were reduces However) there was little improvement with practice and a difference remaineddSt favor of the-linear control. J In continuous-control tasks, low handwheel-gear ratios and high handwheel- " K. J. W. Craik and M. A. Vince, A note on the design and manipulation of instrument): knobs, Kept. 46/172, Med. Research Council Unit in Applied Psychol., Cambridge University,;:. England, January 14, 1945. :){, "The Foxboro Company, Foxboro, Mass., Inertia, friction and diameter in handwhedf;! tracking, OSRD Report No. 3464, 1943. " A. Chapanis, Psychology and the instrument panel, Scientific American, 188, 74-82* (1953). ,fl HUMAN ENGINEERING AND INDUSTRIAL SAFETY 105 fjjare advantageous when accuracy is important. This is particularly true for 'nSfe'd'personnel.85 &Rriction and Inertia. With continuous-winding tasks, viscous friction and l|s from heavy handwheels or flywheels, seem to have a beneficial effect on mms" ai$y of control. In a lever type of control, however, friction has a negative ffiThis advantage of inertia tapers off more rapidly for small-radius handli|t!han for large ones because of the force necessary to accelerate loads when Bp-small-radius controls.85 Viscous friction and inertia are particularly !*tfor accurate control under bumpy conditions or where vibration is pres- IMitibn up to 4 pounds at the: handle has been found beneficial, but it does not *m'ltely that the use of higher values would result in any improvement.87 Continuous-control tasks using handcranks, coulomb friction, or "stricfesvbeen found to have a deleterious effect on the smoothness and accuracy jp|ponse. Values up to 7 pounds are increasingly detrimental; higher values to be much more so. This effect is less apparent for high handwheel ^b'and heavy handwheels.85 Although this type of friction is practically ^present to some extent in control systems, its reduction is apparently worth nfre accurate control is required. r-type controls for continuous-control tasks it is feasible to build ||y8tem a pressure feedback cue that increases with displacement. This cue Implement the position cue of the control and would probably enable the `T(5rknuw"the"pasitioirof'hisT:ontroTbetter arfd't'hereby attain'more accuiftion. The pressure, however, should not be so great as to interfere with J%pid movement required in the task. Further, this use of pressure should Jto:situations where the control need not be operated over intervals long mtfesult in undue fatigue.2 stable Platform for Operations. The operator of any type of machine ||||nb should be able to maintain an upright posture on a stable working Twhether seated or standing. Deviations from an upright position, such " he bending of the trunk or body or in the extreme movement of an arm ^in the nonproductive use of energy and hence reduce the efficiency of further, bending or rotating of the head or trunk inevitably alters the ^thatrthe`view~Of"the'area may`be modifiM^flrfedueedt'Airopferator i to move freely, shift his weight, and adjust to a more comfortable |^Kpj|never he so desires. He should not be required to caijry out repeated ilfgg's*^P^mpvements for the performance of his task because'df deficiencies in " eiwork-space. Rwc'Ti^' '''SSt some tasks may require fairly extensive body movements. In such can, by movement of his trunk, adjust his center.of gravity,to his base in many different locations. He can also extend his base in ;cv; f Friction in manual controls with special reference to its effect (on accuracy movements in conditions simulating jolting, Rept. 47/386, Med. Research Council Mte.V `,, , ..mvi')'ligd'Psychol., Cambridge University, England, June 18, 1945. `