15 May
2006
15 May
'06
8:14 p.m.
I got one for $5 at a garage sale. Everything but the hard drive (CP3044
40MB) seems to work.
I cracked it open and replaced the hard drive with one of about the same
vintage and still no go. Further research indicates that it only works
with 3 very specific drives, although I don't know what those are other
than the CP3044.
The two ROM sockets are empty. I think they could contain ROMs with a
floppy disk image and allow the machine to boot without a disk. Anybody
know if this is right? Do you need both sockets populated for an
even/odd byte thing or is it one ROM making up a disk and the two
sockets making two disks?
This leads to an OT MSDOS question: Can MSDOS be told about a hard disk
that the BIOS does not know about? You can probably see my line of
thinking. Make a floppy that boots DOS and can talk to the new HD. Burn
that onto a ROM and then make the machine boot from the ROM socket.
Now what I would do with this 286-10 after that is unclear.
15 May
15 May
8:31 p.m.
C. H. Dickman wrote:
I got one for $5 at a garage sale. Everything but the
hard drive (CP3044
40MB) seems to work.
I cracked it open and replaced the hard drive with one of about the same
vintage and still no go. Further research indicates that it only works
with 3 very specific drives, although I don't know what those are other
than the CP3044.
Chase down the interrupt table to find the Disk Parameter Table
(or whatever it's called) using DEBUG. See what disk geometry
the machine wants (i.e. find the entry corresponding to the
disk "type" in the BIOS settings).
A hack I have done in the past (other machines) for old drives
with "wrong" geometries is to take some bogus "DPT" entry and
modify it to match the disk that you want to install. Then,
take still another bogus entry and "- modify" it -- to ensure the
checksum stays unchanged. Burn new PROMs to replace the
originals, etc.
The two ROM sockets are empty. I think they could
contain ROMs with a
floppy disk image and allow the machine to boot without a disk. Anybody
know if this is right? Do you need both sockets populated for an
even/odd byte thing or is it one ROM making up a disk and the two
sockets making two disks?
If it is a 286, chances are the two ROMs are hi/lo bytes
of a 16 bit word.
This leads to an OT MSDOS question: Can MSDOS be told
about a hard disk
that the BIOS does not know about? You can probably see my line of
thinking. Make a floppy that boots DOS and can talk to the new HD. Burn
that onto a ROM and then make the machine boot from the ROM socket.
See above. :>
Now what I would do with this 286-10 after that is
unclear.
Let me know since I have a Compaq Portable III that I am equally
pondering... :-/
16 May
16 May
7:31 a.m.
At 05:31 PM 5/15/06 -0700, you wrote:
Let me know since I have a Compaq Portable III that I am equally
pondering... :-/
If your P III has the backpack option it will take regular ISA cards.
The P -IIIs with the backpack make nice controllers for EPROM burners,
HP-IB and HP-IL controllers and similar uses that require a slow system and
an ISA slot that you plug in an interface card into.
Joe
12:02 p.m.
Joe R. wrote:
At 05:31 PM 5/15/06 -0700, you wrote:
Yes, I currently use mine as "console" for my Unisite.
I've hacked it to support a 600MB drive which gives me
plenty of room for ROM images, etc. (hence my previous
comments about how to support "unsupported" disk drives
in older equipment)
Let me know since I have a Compaq Portable III
that I am equally
pondering... :-/
If your P III has the backpack option it will take regular ISA cards.
The P -IIIs with the backpack make nice controllers for EPROM burners, HP-IB and HP-IL controllers and similar uses that
require a slow system and
an ISA slot that you plug in an interface card into.
I assume the "backpack" is the "expansion chassis"?
No, I haven't (yet) acquired one. I had tentatively
been planning on using it to host an OPUS PM -- since
I believe the expansion chassis will handle *two*
(full length?) ISA cards (the second card being a
NIC for the PM). But, I am not sure if:
- it supports full length cards (the PM is such a card)
- power supply can support the PM
- the PM's daughter card will encounter mechanical interference
- NIC will fit in the case *with* the PM's daughter card
Since it hasn't been a priority and it doesn't take up
much *space*, I haven't pursued it...
2:43 a.m.
On 5/15/2006 at 8:14 PM C. H. Dickman wrote:
I cracked it open and replaced the hard drive with one
of about the same
vintage and still no go. Further research indicates that it only works
with 3 very specific drives, although I don't know what those are other
than the CP3044.
My Grid 1040 shows the following 3 DPT entries:
612 cylinders, 2 heads, 17 sectors/head
612 cylinders, 4 heads, 17 sectors/head
612 cylinders, 2 heads, 34 sectors/head
IOW, 20 MB, 40 MB and 40 MB.
Is is possible to load a driver to access other types of disks? Maybe, but
Grid hard disk controllers are in a world of their own. At least they're
not at the standard port addresses.
Cheers,
Chuck
6:05 p.m.
Chuck Guzis wrote:
My Grid 1040 shows the following 3 DPT entries:
612 cylinders, 2 heads, 17 sectors/head
612 cylinders, 4 heads, 17 sectors/head
612 cylinders, 2 heads, 34 sectors/head
IOW, 20 MB, 40 MB and 40 MB.
Interesting. The information I found said 20MB, 40MB and 120MB, although
it was specifically referring to the 1520. I didn't try to dig around
the BIOS with debug to see what was there though.
Is is possible to load a driver to access other types
of disks? Maybe, but
Grid hard disk controllers are in a world of their own. At least they're
not at the standard port addresses.
So then I will have no luck trying to use any PC disk diagnostics
software? I tried something that I had in a box of old DOS disks and it
didn't work...
-chuck
12:18 p.m.
C. H. Dickman wrote:
I got one for $5 at a garage sale. Everything but the
hard drive
(CP3044 40MB) seems to work.
I cracked it open and replaced the hard drive with one of about the
same vintage and still no go. Further research indicates that it only
works with 3 very specific drives, although I don't know what those
are other than the CP3044.
The two ROM sockets are empty. I think they could contain ROMs with a
floppy disk image and allow the machine to boot without a disk.
Anybody know if this is right? Do you need both sockets populated for
an even/odd byte thing or is it one ROM making up a disk and the two
sockets making two disks?
This leads to an OT MSDOS question: Can MSDOS be told about a hard
disk that the BIOS does not know about? You can probably see my line
of thinking. Make a floppy that boots DOS and can talk to the new HD.
Burn that onto a ROM and then make the machine boot from the ROM socket.
Now what I would do with this 286-10 after that is unclear.
That part number, CP3044, sounds like the same drive that Zenith used in
the SupersPort series. If so, it uses a precursor to the standard IDE
interface and is unique.
It is not likely that hacking the drive CHS table will do anything.
The ROM sockets are for a read-only extension to the floppy drive. The
BIOS supports it that way. The ROMs can contain a bootable image of DOS
and/or any files, including exe's, com's, bat's, etc. There is a
program from GRiD called "rombuild" that would take a file or set of
files and/or the boot portion of a floppy and create images for the
ROM's that you would burn, plug into the 1520, and the BIOS would then
boot from them or just see the files on them.
There is a GRiD group on yahoo groups that has the program rombuild.exe
and a short doc file on how to use it. I put it there and have used it
a lot in the past.
The GRiD machines were very cool for the time. Quite rugged and built
to last a long time.
Dave
1:12 p.m.
On 5/16/2006 at 12:18 PM Dave Mabry wrote:
That part number, CP3044, sounds like the same drive
that Zenith used in
the SupersPort series. If so, it uses a precursor to the standard IDE
interface and is unique.
AFAIK, the CP3044 is a standard ATA drive--but don't take my word for it.
Here's the product manual for the 3044:
http://www-user.tu-chemnitz.de/~kzs/hdd.dat/conner/cp3044.asc
I do recall that in the early ATA days, the format of the response from the
IDENTIFY command had some variation between manufacturers, but Conner
pretty much set the standard for others to follow.
The old Prarietek drives were a different kettle of fish, however.
Cheers,
Chuck
2:03 p.m.
Chuck Guzis wrote:
On 5/16/2006 at 12:18 PM Dave Mabry wrote:
You are correct. Thanks for the link to the reference material. I was
thinking of the predecessors from Connor that had three digit model
numbers (343 perhaps?). So the 3044 is a "normal" ATA drive. It would
be interesting to see if something can be done to make the GRiD see a
different model number.
Dave
That part number, CP3044, sounds like the same
drive that Zenith used in
the SupersPort series. If so, it uses a precursor to the standard IDE
interface and is unique.
AFAIK, the CP3044 is a standard ATA drive--but don't take my word for it.
Here's the product manual for the 3044:
http://www-user.tu-chemnitz.de/~kzs/hdd.dat/conner/cp3044.asc
I do recall that in the early ATA days, the format of the response from the
IDENTIFY command had some variation between manufacturers, but Conner
pretty much set the standard for others to follow.
The old Prarietek drives were a different kettle of fish, however.
Cheers,
Chuck
6:10 p.m.
Dave Mabry wrote:
Chuck Guzis wrote:
> IDENTIFY command had some variation between manufacturers, but Conner
I tested
the drive by installing it in a NetBSD machine and it showed
the name of the drive during kernel startup. I couldn't get it to read
any sectors though.
It would be interesting to see if something can be done to make the
GRiD see a different model number.
That was a thought that I had too. Perhaps replace the name of the drive
in the BIOS with the name of a drive I have available. I wouldn't even
have to change the geometry if I was willing to waste some (most) of the
disk.
-chuck
6:23 p.m.
Dave Mabry wrote:
Chuck Guzis wrote:
>
> AFAIK, the CP3044 is a standard ATA drive--but don't take my word for
> it.
There is one thing that is not typical about the CP3044 and that is the
relative placement of the 40pin header and the power connector. The 1520
uses an adapter board to connect the floppy and HD to the motherboard.
This adapter board fixes the relative location of the floppy header, HD
header, and HD power connectors. The HD is actually installed inverted
in the computer. (I wonder if that orientation is acceptable for all
drives.) The CP3044 aligns the power pins with the row of header pins
furthest from the disk platters. Newer drives seem to align the power
connector pins with the mid-line between the header pins.
-chuck
6:50 p.m.
C. H. Dickman wrote:
Dave Mabry wrote:
In the machines I have worked on, I have found (IDE) drives installed
"normally", upside down, on either *side* edge as well as pointing
straight up/down (connector on top/bottom).
I have (so far) only encountered SCSI drives oriented "normally"
or on their side edges. (though I suspect no differences between
them and their IDE counterparts).
Most manufacturers describe allowed mounting orientations in their
data sheets for the drives.
Chuck Guzis wrote:
>
> AFAIK, the CP3044 is a standard ATA drive--but don't take my word for
> it.
There is one thing that is not typical about the CP3044 and that is the
relative placement of the 40pin header and the power connector. The 1520
uses an adapter board to connect the floppy and HD to the motherboard.
This adapter board fixes the relative location of the floppy header, HD
header, and HD power connectors. The HD is actually installed inverted
in the computer. (I wonder if that orientation is acceptable for all drives.) The CP3044 aligns the power pins with the row
of header pins
furthest from the disk platters. Newer drives seem to align the power
connector pins with the mid-line between the header pins.
6643
days inactive
6643
days old
11 comments
5 participants
participants (5)
-
cclist＠sydex.com -
chd_1＠nktelco.net -
dgy＠DakotaCom.Net -
dmabry＠mich.com -
rigdonj＠cfl.rr.com