Understand the issue with the 1053 not fully moving to rotate 5 positions; dealing with carrier return tweaking

ROTATE MECHANISM REFRESHER

The type element (golf ball) rotates to bring a particular column of the characters to the front, choosing one of eleven positions know as rotate -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, and 5. Mechanically this is accomplished by tightening or loosening the tension of the metal rotate tape, pivoting an arm on the left side of the machine inwards towards the center or outwards to the left side. 

To trigger a print cycle, some of the selection magnets are activated. These push a particular linkage away from the moving bail such that the linkage is NOT pulled down as the bail is moved down and up in the cycle. Any magnet that is not activated does not push the linkage, thus it fits under the bail and is pulled down. 

The rotation choices are made using the R1, R2A, R2 and R5 selector magnets. A mechanism hooked to those four linkages converts the binary activation of the four links into a pivot of the left arm to one of the eleven rotation positions for the ball. In the early part of the print cycle, the arm pivots and the golf ball is rotated due to the change in tape tension.

The tape runs from the right side of carrier to the right side, reverses direction around a pulley on an arm, runs underneath the carrier to the pivot arm and pulley on the left side, then back to the carrier where it pulls on a spring loaded wheel inside. The pulleys on both sides allow the carrier to move left and right on the machine without altering the tension of the metal tape; only movement of the left and right arms changes the tension. 

The right side arm pivots to one of two positions, which twists the type ball to either the upper case or lower case side. Each ball has twenty-two columns, eleven per side. A shift cycle is executed to turn the ball 180 degrees prior to a print cycle for a character that is on the other hemisphere of the ball. 

When the left arm has moved to turn the ball to its approximate column position, a cam releases a spring loaded detent arm that slides into the triangular teeth on the bottom of the ball, locking it into a precise position before the ball is swung into the ribbon and the paper behind that. 

CHECKING TAPE TENSION AGAINST MY WORKING 1053

I found that the tension of the tape was far too loose on this machine compared to a properly working typewriter. I adjusted the rotary spring inside the carrier until I had the proper tension. Without enough tension, the selections that move the left arm inward, releasing tension, won't turn the ball far enough. 

CORRECT TENSION, NOW OTHER ADJUSTMENTS NECESSARY

Once the tension was correct I worked on other adjustments, I set up the linkage from the mechanism that determines the desired column to the pivot arm so that it was in the right spot for a rotation 0 character, one that has no change in rotate tape tension. 

linkage from mechanism to pivot arm

pivot arm to add/remove tension

I then adjusted the attachment of that linkage to the pivot arm so that it produced sufficient swing of the arm to move the type ball all the way to the +5 and -5 positions. Initially the ball did not rotate quite far enough for any of the negative or positive rotation columns. 

It was still not locking in the correct column every time, because the detent wasn't reliably sliding into the proper tooth of the ball. To work on this, I have to adjust the ball on the rotary wheel, from underneath the carrier, to achieve this. There is a spot where the detent tip should touch the typeball just before it engages, high on one side of the tooth of the ball, which this rotary wheel adjustment will help to set correctly.

Setscrew underneath rotate pulley on carrier

I will be making those adjustments next time I am in shop and hope to have the character selection dialed in just where I want it 

OTHER ADJUSTMENTS I NEED TO MAKE - CARRIER RETURN AND CR/INDEX CLUTCH

I was not satisfied with how the carrier return was latching on and how it was terminating when the carrier reached the left margin of the machine. There are two pushrods involved in this, a sliding margin bar and quite a few links and latches behind and on the left side of the machine. 

When the operational magnet for CR is activated, or the operator pushes the Return button on the front of the 1053, it releases the latch which was locked in the forward idle position. As the latch is pulled to the rear of the unit by a spring, it trips the right hand operational clutch which takes one full rotation. 

As the clutch rotates, cams on it drive a bail downwards at the rear of the machine in the operational bracket area. The latch which had just tripped the clutch is also pushing a lever under the bottom of the bail, so that as the clutch turns in pulls that lever down and then releases it. As the clutch cycle nears its end, a restoring cam causes the latch to be pushed forward where it will lock again in the idle position. 

The lever that is pulled down by the bail rotates or pulls on a number of levers and linkages related to carrier return. At the start of the CR lever movement, it pulls a pushrod to release the margin bar in front, which is spring loaded to lock in its most rightward position. 

A mechanism on the right side of the machine latches in its downward position, which maintains the machine in return mode until the carrier actually reaches the left margin of the machine. The CR clutch cycle completes relatively quickly but its bail latched the machine into return mode. 

Other linkages pivot around as the machine is put in return mode, ultimately causing a nylon shoe to be pressed against the unconnected end of a coil spring that is driven at its other end by the operational shift which runs continuously because of the motor. A cylinder that sits under the coil spring is attached to the gears that will wind the escapement cord drums in the return direction. When the shoe presses down on the spring, it winds up around the cylinder so that it grips and turns the cylinder and gears with the motor power. This spring plus shoe is a kind of clutch to drive the return operation. 

The CR operational clutch finishes its one rotation but has engaged the return mechanism to continue pulling the carrier to the left until the left edge of the carrier strikes the left margin lever. That forces the margin bar to the left. At the start of the return operation we armed the margin bar by the action of one pushrod; striking the left margin pushes it the other direction which is the trigger that activates a second pushrod.

That second pushrod releases the latched up return operation, pulls the nylon shoe up away from the coil spring, and thus completes the carrier return at exactly the spot we wanted it to stop. There are quite a few adjustments to all the parts necessary to properly arm the margin bar, release the latched return operation, engage the shoe for return motor power but ensure it does not drag on the carrier at any other time, engage the gears with the escapement drums, and so forth. 

I was not satisfied with how the return operation latched into place, nor with the arm and trigger of the margin bar. I began adjusting screws and moving parts in a way that made sense to improve the latching, for example, but discovered unintended consequences where the nylon shoe was partially engaging the clutch at all times. 

Adjustments in a Selectric mechanism tend to be inter-related; changing one will throw off other adjustments quickly. There is typically a sequence by which the adjustments must be made to achieve satisfactory results. I will study that sequence and the individual adjustments before I return to work on the typewriter.

Finally, as I reported last time, I have to adjust the CR and Index operational clutch so that its check pawl drops into the notch when the clutch comes to a stop at the end of a cycle. The computer and the typewriter are interlocked to avoid driving any operations on the typewriter while it is in the midst of prior operations. The check pawl must fully enter its notch or the microswitch that holds off the CPU will remain activated and block any further output to the typewriter. 

The two indefinite length operations on a Selectric are carrier return and tab. There are two microswitches on the operational magnet assembly that are activated as soon as the CR operational clutch or the tab/space/backspace clutch are triggered. These hold the contacts until the associated clutch has completed its rotation and its check pawl is engaged. I am waiting on thin wrenches in order to make this adjustment. 

In addition to the clutch time microswitch, return and tab each have a second switch that is activated as long as the function is active even though the clutch itself finished; the clutch only starts the operation it does not power the entire function to its end. For return, the mechanism that latched in return mode has a microswitch and is only released when the pushrod unlatches it based on the margin bar being triggered. For tab, the tab torque bar is held in its latched position until the carrier strikes a set tab in a column which released the torque bar; there is a microswitch detecting when the torque bar is latched (twisted).