The model 51R controller that I own has four possible combinations of disk devices it supports. One 1.2MB floppy (A), Two 1.2MB floppy drives (B), one 1.2MB and a 2.4MB floppy drive (C), or a 1.2MB floppy plus a 20MB hard drive (D).
The base configuration which I have has the cabling in place to support two 1.2MB drives although only a single drive is installed at the factory. This supports either (A) or (B) configurations above. The cable has a 40 pin IDE style connector onto the motherboard and discrete wiring to a pair of 34 contact card edge connectors (IDC style).
Adding a 2.4MB second drive, configuration (C), requires a different cable which implies that the wiring for the two drive types is different. The microcode can determine based on the wiring of the cable which of the two densities is installed for the second drive.
Configuration (D) uses a different 56 pin connector to the motherboard and has discrete wiring to three connectors. One goes to the first 1.2MB floppy drive and therefore has the same wiring as the cable for (A) and (B). The other two cables go to a pair of connectors on the hard disk.
The motherboard has two connectors side by side, a 40 pin position used for a floppy only installation and the 56 pin connector necessary for attaching a hard disk plus a floppy.
THREE CABLES SPECIFIED IN THE PARTS LIST
As a consequence of the two different motherboard connectors and the wiring differences that inform the microcode about the density of each attached floppy drive, there are three different part numbers for the cable. A 1.2/1.2 cable, a 1.2/2.4 cable and a 1.2/20 cable are the three choices. I only have the 1.2/1.2 cable which precludes me from attaching one of my Gotek/FlashFloppy drives as a high density drive allowing me to IML from the B5 microcode version.
WIRING INDICATES THE DRIVE DENSITY INSTALLED
I was fortunate enough to be sent a wiring diagram of the cable for configuration (D), the floppy plus hard disk combination as installed on the larger controller models that only offer 2.4MB drives. Theoretically, if I compare the wiring from this diagram I have with the cable installed in my 3174 I should be able to figure out what wiring differences exist.
This was complicated because IBM has their own numbering scheme for connectors. This means that the signals that are common between the two motherboard connectors, the one for dual floppy drives and the one for hard disk plus floppy, are on different pin numbers on the two connectors. I then have to map the IBM numbering for the card edge connector against the way the entire rest of the world numbers it, before I know the logical signal associated with each motherboard pin
The existing cable I have, for dual 1.2MB drives, is properly mapped to motherboard pins on the 40 pin connector it uses. The same connector would be used for the 1.2/2.4 drive pair cable, but I don't know the differences in wiring for that cable. I thought I might be able to interpolate this from the floppy cable on the 2.4 plus hard disk cable for which I have a diagram.
ISSUES WITH THE DIAGRAM FOR THE 2.4MB PLUS 20MB HARD DRIVE CABLE
The diagram as it sits has a floppy drive that cannot work. It is missing connections to the floppies ReadData, WriteData and Track00 signals. Based on that I have to assess this diagram as inaccurate to the point that I can't judge how the 2.4 differs.
WILD GUESSES BASED ON A FEW CLUES
The signal lines from the floppy plus hard disk cable have 36 total pins going to the floppy drive, 18 of which are ground. Looking at my dual floppy cable, which has 40 pin positions but four of them are unpopulated. Thus, I presume it carries the same 36 wires with 18 on the ground side.
From my dual floppy cable, I see that all 17 signals of the IDC connector are used, including pins 4 and 6 that are shown as reserved in modern diagrams. I know that the hardware modification I made to the Gotek as to present a signal on pin 4 when the drive was selected, claimed to be a "ready" status signal.
A partial description of the pins used for the PC/AT diskette interface lists pin 6 as the "drive type" signal sent to the FDC adapter. I extracted the 1.2MB floppy unit from my 3174 (it is a Hitachi HFD516C12U drive) and looked a bit at the use of pin 6 in the drive.
The drive has a jumper set up for pin 6, but no jumper in place. The other end of the jumper takes the signal to the input of a 74LS04 inverter gate therefore with the jumper connected, this drive takes the signal as an input from the motherboard. Since it is not connected, pin 6 is left to float open.
The 2.4MB drive can read and write media written for 1.2MB drives, according to various posts I have found online. A hobbyist claimed their drive only operated in 1.2MB mode when used with a PC. Also, a post suggests that the first track is written at 1.2MB and then subsequent tracks are written at 2.4MB for high density mode.
This suggests to me that the signal on pin 6 might be an output from the 3174 that switches the drive between standard and high density modes. This is confusing since pin 2 is a density select signal and should be able to accomplish this role.
If the pin is used to switch the drive to the higher density, it still leaves the question open about how the 3174 knows it has a 2.4MB capable drive attached. Further, since pin 6 is wired to the 3174 even in my dual 1.2MB cable, there would be no need for a different cable to support the second drive at 2.4MB unless it changes wiring somehow.
If only there was a copy of the data sheet for either the Hitachi HFD532EIU or the Y-E Data YD-803 I might be able to figure out more. On the other hand, given the existence of a distinct cable wiring, I might have a problem that can't be resolved without access to a 2.4MB cable.
At this time, therefore, I have to accept the 3174 as a dual 1.2MB system.
No comments:
Post a Comment