AUTOMATIC TRANSMISSION CONTROL SYSTEMS byO.K.KELLEY andM.S.ROSENBERGER General Motors Corp. THIS description ofthehydraulic control used with thehydra-matic transmission reveals how thecontrol operates tochange ratios under powerwithout direction from thedriver. Thecontrol's pattern ofautomatic shifting for ordinary, high-range driving hasbeen selected asthebest compromise between topperform ance andlowratio ofengine noise towind noise. Thecontrol's lowrange shifts gears according to performance dictates alone, furnishing greater power forextreme conditions atlowspeeds andenabling thedriver tousehisengine asabrake onsteep descents. Heart ofthecontrol system isadouble hy draulic governor, sensitive bothtocarspeed and throttle opening. THIS paper, aswellasthetwothatfollow, one byMessrs. Nutt andSmirl andtheother by Mr. Kimberly, make upasymposium onauto matic transmission components presented atthe 1947 SAESummer Meeting. INouropinion thecontrol ofanautomatic trans mission should provide automatic ratio changing without anyconscious effort onthepart ofthe driver. Wefeelthatthere should benothing new forthedriver tolearn —that heshould beallowed merely tohold hisfoot ontheaccelerator tothe degree ofperformance hewishes andtheauto matic control should dotherest. Idonotwish toimply any criticism ofthe designs that donotwork this way. Iknow that they have been produced andsold andwell liked bytheAmerican public andcertainly have been commercially successful. Neither doIwish tobe- little ourreasoning fordoing itourway toaccom plish continuous automatic performance. Iam just personally glad that there issufficient dis agreement infundamental reasoning tokeep thecompetition inautomatic-transmission design lively. The keener theengineering competition, thegreater willbethepublic benefit.Allourbackground andexperience isonthe type ofcontrols that change transmission ratios under power. Ishall limit mypresentation, accord ingly, totheproblems ofproviding acontrol sys temforthehydra-matic type ofautomatic trans mission. Compromise Needed Assuming a4-speed gear boxofatotal ratio coverage of3.56 to1andarather fast rear-axle ratio of3.4to1,ourfirst problem istodecide at what speed thetransmission should shift from low tosecond, tothird, andtofourth. Acompromise must bemade between maximum performance and permissible engine noise. Fig. 1illustrates our final choice ofthis compromise. Theengine speed istaken nohigher than 2000 rpm inlowgear before shifting into second atfullthrottle. At lower carspeeds where thewind noises have not reached anyappreciable value, theengine noise stands outvery prominently andthebest compro[This paper was presented attheSAE Summer Meeting, French Lick,Ind., June 3,1947.] October 1947 Vol. 1No.4 559 Downloaded from SAE International by University of British Columbia, Tuesday, September 25, 2018The Authors O.K.KELLEY (M'26), head oftheProduct Study Section ofGeneral Motors Corp. tocon tinue the development ofautomatic transmis sions, and M.S.ROSENBERGER (M'45), assis tant engineer incharge ofthesame section, have followed similar paths intheir engineering ca reers. Associated with General Motors for over 20years, both were members ofthe early GM Product Study Section that developed thehydra matic drive. They both served aschief engineers ofthe Detroit Transmission Division manufac turing thehydra-matic drive. Mr.Kelley wasap pointed head ofhis section in1940, and Mr. Rosenberger joined him in1945 asassistant engi neer incharge. mise seemed toustoberidofthelowgear rather early, atapproximately 15mph. Second gear is continued to30mph atfullthrottle, letting the engine speed reach 2800 rpm. This isnomore objectionable fornoticeable engine sound level than lowgear's 2000 rpm. Additional wind noises at 30mph help tocover upthegreater engine noise. Alarger step ingear ratio between second and third permits theuseofthethird gear toahigher speed, upto65mphwith theengine speed reaching 4100 rpmatfullthrottle before shifting