LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN 510.84 Digitized by the Internet Archive in 2013 http://archive.org/details/comparisonofread30ceda SBR0B8C I)IOI?/I. COKRT XSTCRTL repot by '-- ltheugh :.li:«w IV, the h? digit mm..iy, can bo coded to t«sst «^d rint o r?wad«firoiajd-»riiti ,..-e JPos a; : has at r acnpry a&dreeees in less then one-half hoar* it is suspected •atiae conditions are not close eno £h to Identical all ;• • . - r Oils as a and the fact that on* vastus e*:aputer tine testing nlif a few tubes at a tina, it seen® desirable to have a tester i h ons am test nic •**& tub* at a tlna for read* and v* : ill give results that can s*redi- this t .be 3 :*?rfanttnco in Villlssjs IV. Williams II. the one digit wrvt-vy slailar in operation ":. liana IV, was modified t --ide a resA~are*sad-rntU test f JIM cathode rr *» new eircr ' -marates all nenwy positions at tine rate of .-no every ; n ds, then st:res a d>a or aaeh frost to 102^ time* at any given single address as specified* After storing et the rate of one every ?<5 sdcroeec nds a given nuntber ot tines vhieh can be set by the operator, the elm repeats thts cycle. The read-ar ■und-ratlr- as defined for this experiment It the largest n- after of tines ont can store a dash, the whole cycle boim wj U Bisd for 15 seconds, vith^rfc clanging sny f the ? rrounding 9 to dashrc. (Circuit appears or* draw* •r feature >f this ci s that c f beinff al alternate the infornati n stored In a raster approximately once every -d. That is changing dots and dashes to dashes and dot- . '?an be done over the whole rar the raster while the at statically. Thus • the Ofaratie© of a oat'ao&e ray tube, { ••.■„.? b«s »tor©4 at every add (..:) Doe: ) Can I infomation ":*? alternated sat la the raad-Arcuod- ■AjNMMft Of tfftaMtlty, f Cua Ml Mtl0Mfct'5?i. ■' t-;.:..M'.- r ~;y/.-;.' :dv;Yxl .;.• . good fO m and : are ad 'us ted for clattu signals undc Mia | 150 J^L-Jj ^ I 2*tt) . UL 7 1*8 The raadHwroov' able 1 are for the wmm tralbe and in all tane e e as e s the foeoa aad w that roquiwawrta (l)| (2) f aat (~) were net, tnrt raad-aro-ral-^ "jrent. For a good read anuria ratio at all adavaaaaa a mora axaet for adVfftwent Is necessary., at leer high read-ar ttabl© 2a lose r/ and V!lliaw daggered *»■ Ttm awe addressee wsms used for i. £i . Vfcuix bit ::. «L tta.iv 7 .- ua • 3 M kfl b CGHt«r I*QJ. :ad-*ro .;.d-r lose IV end Btagg&ftA *•»& -. : . 1 A »*. II 5 7 Htau IV 7 -fc- Aa examination of table 2 docs not give any exact correlation between the read-around-ratios determined In Uilllaas IV and those determined in Williams II, b t some wM correspondence can be seen* Tube 60j vhicb ban the highest rea4-«ro\aid-ratloa of the four according to Williams II, has the highest according I llama IV. Tube ?**• baa a very non-uniform read-around -ratio by both tests and would be proclaimed geor by either. Tube f$k would almo be called po:->r by bath tests* Tube 108, however, hair a relatively uniform read-aro>ind- ratio according to William* II and the trend is high, but Williams IV people want to thr-jw it out. Very oseibly It could have been In better focus alth there is no quick way of knowing vhen the better focus 1® obtained in Williams IV as there Is liams II. There are two significant differences between the t- Tbe Williams II test requires that the addresses surrounding the at spot do not fall to dashes while the process cycles for 15 sec nd. The Williams IV test essentially cycles only once for each determination of the reed-around-rp lo reduce the robabi * an abnormally high read-around -ratio In Williams IV due to a regeneration of t storing spot, the process !e carried out h times, the lowest of the k trials being called the read-ar>und ratio. Th'f? alt partly isates for the fact that the operator of William II is at choose the lowest of several 15 second trials. This accounts for the generally smaller magnitude of the Williams II read-ar und-ratlo but does a lain the ■uniform correlation between the two figures. The other big difference is the unoertftlmly In the .oiiformlty of i acus for Williams IV. Tubes la William* II were adjusted (net gmatlem cad focus) for beat and approximately equal r««l-arouad-ratlo at the center and c Bj vhar reason for a difference In nagalt ude is that wh(*u Williams II is storing, It stores every conds wit iny interspersed regeneratioms, while Williams IV stores only 2 la every Ikk seconds. For a read-around- rat i be d rati ;he store period would be . V 0S second and .007 second fe Haas II and Williams IV; respectively. If ..lae is sna!" compared to the tim f t a charge t< leak off, the difference la the two tests caused by thJs effect will he small* Other causes for a magnitude difference might be different twitch or stagger factor or raster size and different beam-on- time caused by differing Ises. 1 of these departures fr m exa llcatlon of ons should cause approximately a constant factor difference in magnitude all ad dr es s e s of the same tube with the exce tion of the afore- mentioned fbM criterion* But s'ich is not the case* A better sr * adjusting the focus me use better correlatioa between the results of the read-around tests in the two machines, hut mare important, w* I be using our cathode ray tubes more effectively under s-ich adjusts* It can be temcluied that if one eomsldere the general magnitude of the read-around- rat l. nd-rsr I Villi aw IV doea not seem feasible with the present set u ,