Using CE Cartridge to align heads on IBM 1130 internal disk (13SD factor name, 2310 when in external cabinet)

THEORY OF THE ALIGNMENT TRACK AT CYLINDER 100

IBM recorded a 5 mil wide track that is all zero bits, but offset the recording mechanism so that the track is eccentric compared to all normal data tracks. Each track/cylinder is 10 mils apart, with the data ideally centered in that ring as the surface rotates under the head. The eccentric track moves about 2.5 mils either side of center, touching the outer edges of the ring at two points and being perfectly centered at two other points, all points being 90 degrees from each other. 

The head on the 13SD disk (the internal disk drive of the 1130, also called a 2310) has a read/write coil and an erase coil. The read/write head reaches each point on the track first, with the erase coil trailing behind. The width of the recorded pattern is wider than 5 mils initially, so the erase coil trims off the edges to leave the 5mil center. 

When using the CE cartridge to read the alignment track at cylinder 100, the read/write heads are disconnected and the erase coil is instead connected to the read preamplifier. IBM uses a special adapter that is a cable extension to swap the wiring. This means that the coil is only seeing magnetic flux at the two small poles that make up the tunnel that the normal signal passes under. 

Thus when the eccentric track moves towards one extreme or the other, it is mainly under one of the two poles. This produces a strong signal of one polarity or the other depending on which side of the coil the track is under. As the eccentric track returns to the nominal path in the center, the signal is in the tunnel and produces a minimal signal.

Thus, using an oscilloscope, we can watch as the heads are moved outward from before cylinder 100, gradually seeing a strong peaked pattern from the eccentric track. Initially peaks in one polarity are much higher than the other polarity but as the head is moved to the central position in the track, the two peaks should be equal heights. 

Theory

Directly on target position

About 60% offset from correct position

Vendors who built drives derived from the 13SD, such as DEC, Pertec, Diablo and others, usually moved to a higher density recording head. However, they still use the same alignment philosophy. Their versions of the CE packs have a high density eccentric track at cylinder 105 in addition to the standard density track at 100. 

They also worked out an interesting way to use the standard density track at 100 from an IBM CE cartridge to align their high density heads, by inserting a 5mil shim at a precise location on the cartridge to increase the eccentricity of track 100 (and all other tracks as well). This increased width made the signal peaks easier to read with high density heads.  

I can see that this might be a way to write an eccentric track on a regular cartridge to use in a pinch if one doesn't have an IBM CE cartridge available. 

SWAPPING LEADS ON DISK DRIVE ELECTRONICS GATE TO USE ERASE COIL

The leads from the disk heads are connected to pins on the backplane (SLT board) on the rear of the disk drive. By moving the leads around, the erase coil can be connected to the read preamplifier rather than the read/write head. This matches the function of the IBM alignment adapter tool.

HOOKING UP SCOPE TO TEST POINTS ON THE DISK DRIVE

The preamplifier circuits have an output at slot D3, pin B07 (also named D2 G07) on the backplane, which is hooked to an oscilloscope. It is set with the input at 500mv per division and the timebase at 5ms per division. The pattern displayed is used to adjust the arm.

LOOSENING THE CLAMP SCREWS ON A HEAD

The top head is 0, the bottom is 1. Working with one head at a time until it is aligned, we loosen its clamp screw and pull the arm out towards the spindle of the disk cartridge a bit. The adjusting screw for that head is preset to a depth of .188" from where it exits the bracket to its point. This is a good starting point for the head to be definitely to the left of the proper track position. The arm is pushed back to touch the adjusting screw and the clamp set to very light pressure. 

SPINNING UP CE CARTRIDGE AND ALIGNING THE HEAD

While watching the oscilloscope, the adjusting screw is screwed slowly clockwise which pushes the arm outward towards the cartridge spindle. The clamp should allow this to move the arm but not be so looses that the arm moves beyond where the screw tip would push it. 

When the signal looks correct on the scope, tighten the clamp screw while checking that the signal is still good. This may take a couple of iterations as the clamp screw can sometimes move the arm a bit as it is tightened which must be anticipated with the adjusting screw positioning of the arm. 

Turn off the drive. Switch to the lower head (1), loosening its clamp screw, pulling it out a bit and presetting the adjustment screw for the lower head to 0.188" depth. Push the arm back in and apply very light clamping. 

Spin up the drive and adjust the screw for head 1 until the signal will be evenly balanced when the head clamp screw is fully tightened. This finishes the alignment.