Very low voltage negative logic

[epitaxial]

Hello all. I'm documenting the hard drive interface IBM used on two models in their System/36 line, models 5360 and 5362 specifically. This is a model 21ED drive. There are a number of pins with 5v or 3.3 TTL logic but two pins have this strange negative logic that swings from slightly below ground to negative 0.4 volts. These signals are always present even when the drive is idle. Anyone have a name for this or is this something that only IBM could dream up?

[langwadt]

some kind of ECL?

[amyk]

Looks like raw (amplified?) data from the disk head.

[epitaxial]

Data from the disk head makes sense. I'm going to see if this is active during a spin up. Maybe a drive signature or check that's always running to gauge health? This machine was introduced in 1984 and there is no published information or specs about the interface.

[ejeffrey]

NIM (nuclear instrumentation module) logic used negative current steering logic (some ECL derivative) with similar logic levels.

[Roehrenonkel]

Hi epitaxial,
 
what probe was used?
Maybe ringing in the gnd-wire??
Is there a negative supply on the board anyway???
 
Good luck

[TimFox]

As I remember from 1975, the fast outputs from a NIM module were something like 30 mA (negative) pulse, designed to drive two 50 loads in parallel, usually one at the driver and one at the receiver, to get roughly -700 mV pulse voltage.
The driver was a long-tailed pair from -12 V, where the NPNs did not saturate.

[amyk]

Data from the disk head makes sense. I'm going to see if this is active during a spin up. Maybe a drive signature or check that's always running to gauge health? This machine was introduced in 1984 and there is no published information or specs about the interface.

It's probably the raw data, from whatever track the head happens to be sitting over while the disk is spinning.

How many pins does the interface have? Look up ST-506, which was used in PCs of the 80s. This may be very similar to that.