PDA

View Full Version : 5.25" floppy drive problem



Arkady
November 13th, 2017, 06:36 AM
Hello,

I've got Magnetic Peripherals Model BR8B1A floppy drive (full height, 360K) that doesn't work.


When the drive is alone (A), then the heads don't move at all when booting and after. They stay stuck at track 00.
If i type the DIR command, it can read the directory of a floppy.
If i use ANADISK or DiskCopyFast, i can only read the whole track 00 (face 0 & 1). Nothing else, the heads don't move.


When i use it with a second drive (A or B), then the heads go to track 39 or 40 when booting and stay there.
Even stranger : if i use ANADISK and try to read a track (i.e track 10), then i hear a nasty noise because the drive tries to move the heads. But as they already are at the track 40 position, they bump into the mechanichal stop (-> noise). They never go back, even when i try to read track 00.

Short story : when the drive is alone the heads stays at track 00 and never move (it can read track 00). When there is a second drive, the heads go to track 40, stay there (like if it was track 00). The drive try to move them beyond track 40 (if read command is sent).
Last, i made these tests with an PC/XT and an IBM controller AND a 286 clone with a ISA 16bit controller

This is a picture of the electronic board of the drive :

41930

As it's not completly dead, so where i should have a look first ?

Thank you.

Chuck(G)
November 13th, 2017, 07:24 AM
First question: Is your terminator installed when you try to use it alone? Your photo seems to show a terminator installed, but that's a guess.

SomeGuy
November 13th, 2017, 07:47 AM
As it's not completly dead, so where i should have a look first ?
The cables! It's always the dern cables! :)

When the drive is alone as a:, what happens if the heads are not already at track zero? Do they stay there or do they move to track zero and then stay there?

Another tool to try might be ImageDisk. It has a calibration mode that lets you manually step the heads in and out.

It almost sounds like the head step Direction signal is getting messed up.

Arkady
November 13th, 2017, 10:16 AM
Thank you for your replies.


First question: Is your terminator installed when you try to use it alone? Your photo seems to show a terminator installed, but that's a guess.

The terminator was/is installed when i use(d) it alone as drive A


The cables! It's always the dern cables! :)
Cable tested. Same behavior. Tried another cable. Same. The connector (gold finger) is ok, i cleaned it.



When the drive is alone as a:, what happens if the heads are not already at track zero? Do they stay there or do they move to track zero and then stay there?
They don't move at all (floppy drive alone) or they go to track 40 (Two floppy drives).


Another tool to try might be ImageDisk. It has a calibration mode that lets you manually step the heads in and out.
The heads don't move (one floppy drive) or move a bit - about two tracks - (two floppy drives).

I tried something else (two floppy drives mode) : i managed to format 2 tracks using ANADISK, physical tracks 40 and 41. But the drive numbered them as track 00 and track 01.
So with the two drives configuration, the drive sends the heads to track 40 position and "beleive" that the heads are at track 00.
I wrote the contents of a track 00 onto track 40 (using the second drive) and the "bad" drive managed to read and list the directory (reading the physical track 40, thinking it's track 00). I tried to launch a command and the drive made a nasty noise : it tried to step from the track 40, beleiving it's heads were on track 00.
It can read/write/format/step correctly, but only from track 40 (two floppy drive mode). It doesn't know where are really located its heads.


It almost sounds like the head step Direction signal is getting messed up.
This sounds interesting but i have no idea how to go further...

Chuck(G)
November 13th, 2017, 10:31 AM
I wonder if the track 0 sensor is bad. Can you check logic levels at the drive connector? If so, with the drive selected, pin 26 should read high (>3V) unless the carriage is at track 0, which would cause it to read low (~0V). If the drive is not selected, the pin will always read high.

I believe that the CDC/MPI drive schematic is in the IBM PC XT refereence manual (available on MZD) as the "type 2" drive.

Arkady
November 15th, 2017, 10:23 AM
Bad news, the track 0 sensor (an optical switch) seems to be dead : i always get +5v at pin 26.
The optical switch is connected to the board through a 4 pin connector (the first light blue on the picture). When i unplug it, the floppy drive behave the same way.

Really a bad news because i didn't manage to find informations about it. I found these markings on it :
SPX 4063-18
8343
Mexico

If someone has got informations about this optical interrupter, that would be great... Thank you

Chuck(G)
November 15th, 2017, 10:46 AM
From other CDC/MPI schematics, it's just a simple IR LED+NPN phototransistor affair (Slot-type photointerrupter). Get some dimensions and start looking through datasheets. CDC was not in the habit of giving commodity equivalent part numbers for much of its stuff.

ldkraemer
November 18th, 2017, 02:41 AM
The H21B3 should work fine as long as there is a series resistor for the LED to keep the current to less than 20 mA,
and assuming this style case is needed. (I've cut the LED (TX) and PhotoDarlington out of this plastic case and used
the TX & RX to repair several small Mice.)

https://www.ebay.com/itm/Fairchild-H21B3-Optical-Interrupter-Lot-of-10-H21B3/161006854008?epid=673603093&hash=item257cc19f78:g:dn4AAOxy43FRaEYS

Larry