wcf@psuhcx.psu.edu (Bill Fenner) (07/12/88)
All right, y'all, I've got a problem for someone who's smarter than I to solve. I have, at this moment, an IBM full-ht. 10 meg drive and controller in my cheap snap-together clone. :-) The controller has no identifying marks other than IBM and a serial number. It has no jumpers that are accesible (it has spots for jumpers but there are no pins, just holes in the board filled with solder. I didn't feel like trying any of them.) The drive says Model No: WD12 on it, so I get the feeling that it's a Western Digital drive with IBM's name on it, so that may be the case with the controller, although the controller doesn't seem to have a controller in rom at any of the usual addresses. Of course, no, I don't have any documentation with it. Now the problem. I just bought a second drive, an ST212. It works fine as drive C, it formats correctly and everything, boots up faster and quieter than the IBM drive, etc. But, the 2 drives will NOT work together. I have tried so many things... C and D indicate middle and end of the (twisted) 34-pin cable, c and d indicate the c and d 20 pin drive connectors. DSx indicates that drive select x is active on whatever drive, and (T) indicates the terminating resistor. IBM Cc DS1 ST212 Dd DS1 (T) With this combination, both hard drives are accessed right after it tries for the floppy. (both at once.) and it says "Boot disk failure. Press any key to retry." IBM Cc DS1 ST212 Dd DS2 (T) Like this, before (yes, before) it goes to the floppy, it lights the HD light for about a minute on the ST212, then goes and boots fine but ignores drive D. IBM Dd DS1 (T) same as above, except there's no drive C for it to access so it just hangs. If you switch the IBM and the ST212 around, it does no good. oh, one other thing IBM Cd DS1 (T) works fine. so I tried IBM Cd DS1 ST212 Dc DS1 and this does the same as the last 2. I am starting to think that it's my controller. I haven't managed to find someone who is wwilling to lend me their controller yet, so I can't test this assumption, but everyone I ask tells me that it should work. It's not the 34 pin cable, I've tried 2 different ones and they did the same thing. Please respond via E-mail, I'll summarize if anything interesting comes along.
berger@clio.las.uiuc.edu (07/13/88)
I didn't translate all the alphabet soup you listed, but you mentioned
using a "twisted" 34 pin cable. That's what IBM supplies for their
floppy disk drives, but it should not be used with the hard disk. Use
a straight cable, and set the disk address select jumper on the
drive appropriately.
If you didn't find any ROM at c800:0000, then you may have one of the
very old dump XT controllers. I don't know what you have to do to
get that to recognize a second drive.
Mike Berger
Department of Statistics
Science, Technology, and Society
University of Illinois
berger@clio.las.uiuc.edu
{ihnp4 | convex | pur-ee}!uiucuxc!clio!bergerNanook@cup.portal.com (07/14/88)
Try setting your drives to Select 2. If I remember right on IBM XT's (& clones)
with the twisted cable, the drive should be select 2. See if setting the drive
jumpers on both of them to 2 will fix the problem...
The reason the system hangs in some configrations is that it's seeing responces
across the command lines but nothing on the data lines (two different
connectors). If you want, configur your system so the two drives hang, then
swop the SMALL cables and see if it helps.
Just groping in the dark... Bring it to Florida & I'll be glad to help!
Nanook of the South
uunet!sundc!sun!cup.portal.com!nanook
-------------------------------------------
"How can you expect to run the world if you
can't answer a few unsolvable questions???"
-Flynn [TRON]wcf@psuhcx.psu.edu (Bill Fenner) (07/15/88)
In article <16800327@clio> berger@clio.las.uiuc.edu writes: | |I didn't translate all the alphabet soup you listed, but you mentioned |using a "twisted" 34 pin cable. That's what IBM supplies for their |floppy disk drives, but it should not be used with the hard disk. Use |a straight cable, and set the disk address select jumper on the |drive appropriately. | |If you didn't find any ROM at c800:0000, then you may have one of the |very old dump XT controllers. I don't know what you have to do to |get that to recognize a second drive. | Well, I took the cable apart, removed the twist, set DS2, and bagoomba! 20 megs. Nobody that I talked to, including Seagate Tech support and all the computer stores around town, thought anything about the fact that the cable was twisted. I know that at least 2 other people are using twisted cables for their hard drives... maybe it's the old dumb XT controller that makes it need a straight cable. Anyway, thanks for suggesting this, it would have been a long time before I had thought of it. Bill -- Bitnet: wcf@psuhcx.bitnet Bill Fenner | Internet: wcf@hcx.psu.edu | This space UUCP: {gatech,rutgers}!psuvax1!psuhcx!wcf | for rent Fido: Sysop at 263/42 |
mobo@sphinx.uchicago.edu (Samuel Wilson) (07/15/88)
In article <230@psuhcx.psu.edu> wcf@psuhcx (Bill Fenner) writes: >In article <16800327@clio> berger@clio.las.uiuc.edu writes: >| >|using a "twisted" 34 pin cable. That's what IBM supplies for their >|floppy disk drives, but it should not be used with the hard disk. Use >|a straight cable, and set the disk address select jumper on the >|drive appropriately. >| >Well, I took the cable apart, removed the twist, set DS2, and bagoomba! >20 megs. Nobody that I talked to, including Seagate Tech support and all >the computer stores around town, thought anything about the fact that the >cable was twisted. I know that at least 2 other people are using twisted &c... I put a second ST225 in an ATT6300. I was using a Western Dig. WX2 controller. I was also using a cable with a twist between the connection for drive c and d (or a and b, as the case probably was). the controller could talk to one drive, or the other, but never both at once. I called Everex (who, for some reason known only to them had marketed the WD-Seagate combination), and the (very clued-in) tech guy said, "well, you have both drives set to drive 2 don't you?" (silence on my end). I finally said "Yea, sure I do, course I do, waddya think, I'm a dummy, that's the ticket" and went and set both seagates to be drive 2 and the thing works great. I have some idea how controllers work, and some idea how drives work, but no idea why *that* worked. -- Samuel Wilson ..ihnp4!gargoyle!sphinx!mobo mobo@sphinx.uchicago.edu University of Chicago, Division of Social Sciences
bkliewer@iuvax.cs.indiana.edu (Bradley Dyck Kliewer) (07/15/88)
>controller. I was also using a cable with a twist between the >connection for drive c and d (or a and b, as the case probably was). >the controller could talk to one drive, or the other, but never both >at once. I called Everex (who, for some reason known only to them had >marketed the WD-Seagate combination), and the (very clued-in) tech >guy said, "well, you have both drives set to drive 2 don't you?" >(silence on my end). I finally said "Yea, sure I do, course I do, >waddya think, I'm a dummy, that's the ticket" and went and set both >seagates to be drive 2 and the thing works great. > >I have some idea how controllers work, and some idea how drives work, >but no idea why *that* worked. Since there seems to be a lot of discussion on the topic of twisted wires on the controller cables, you might as well know what the twist does -- it swaps the drive ID lines from between the two connectors. This is done so that all drives are set to the same ID (who knows why, I think it's pretty silly myself -- especially since they're all set to 2). So untwisted wires work the way any normal person would think, drive 1 set to drive 1, drive 2 set to drive 2; but not twisted wires. And, as has been mentioned, this is also done on some floppies. Even worse, I had to install a second drive on a Tandy 1000SX. The edge connector on the card is upside-down from standard, and the cable is extremely short. So, I had to buy some ribbon cable and connectors, make a longer cable, put in a few twists to invert the second connector, and install the drive. Isn't it wonderful how manufacturer's make little changes here and there to prevent you from buying other vendor's parts :-)? Bradley Dyck Kliewer Hacking... bkliewer@iuvax.cs.indiana.edu It's not just an adventure It's my job!
mjg@ecsvax.uncecs.edu (Michael Gingell) (07/15/88)
In article <4452@sphinx.uchicago.edu>, mobo@sphinx.uchicago.edu (Samuel Wilson) writes: > > I put a second ST225 in an ATT6300. I was using a Western Dig. WX2 > controller. I was also using a cable with a twist between the > connection for drive c and d (or a and b, as the case probably was). > the controller could talk to one drive, or the other, but never both > at once. I called Everex (who, for some reason known only to them had > marketed the WD-Seagate combination), and the (very clued-in) tech > guy said, "well, you have both drives set to drive 2 don't you?" > (silence on my end). I finally said "Yea, sure I do, course I do, > waddya think, I'm a dummy, that's the ticket" and went and set both > seagates to be drive 2 and the thing works great. > > I have some idea how controllers work, and some idea how drives work, > but no idea why *that* worked. > On both floppy and hard disk controllers the cables have separate control lines for each of two drives. There is a drive select and a motor on line for each drive. The lines are so placed so that the first connector is for drive two while the lines that go through to the end connector for drive one correspond to unused pins on the first drive. The twist reverses this so that now drive one has its line connected and drive two's wires are unconnected. The only way this works is if you choose drive two in the first place. IBMs wiring scheme is a clever solution to how to get separate Motor Control and Drive select lines to each drive using the existing Shugart drive connector on the drive end and their own "proprietary ?" wiring scheme on the controller end of the cable. What's the betting IBM has this on its long list of patents it has threatened to sock the competition with. Mike ..ecsvax!mjg
karl@ddsw1.UUCP (Karl Denninger) (07/16/88)
In article <4452@sphinx.uchicago.edu> mobo@sphinx.uchicago.edu.UUCP (Samuel Wilson) writes: >In article <230@psuhcx.psu.edu> wcf@psuhcx (Bill Fenner) writes: >>In article <16800327@clio> berger@clio.las.uiuc.edu writes: >>| >>|......using a "twisted" 34 pin cable. >>| >>Well, I took the cable apart, removed the twist, set DS2, and bagoomba! >>20 megs. Nobody that I talked to, including Seagate Tech support and all >>the computer stores around town, thought anything about the fact that the >>cable was twisted. I know that at least 2 other people are using twisted >&c... > >I put a second ST225 in an ATT6300. I was using a Western Dig. WX2 >controller. I was also using a cable with a twist between the >connection for drive c and d (or a and b, as the case probably was). >the controller could talk to one drive, or the other, but never both >at once. > >I have some idea how controllers work, and some idea how drives work, >but no idea why *that* worked. It's simple! Here's the poop: ATs (and some XTs) have been made with a "twist" in the cable. These hard disk control cables (the 34-pin is called the control cable) are *NOT* the same as a twisted floppy cable, although they perform exactly the same function with the twist in both cases. The reason for the twist is to change drive select lines (select, that is). This allows people to slam in drives without understanding that they have to set the little drive-select jumper first -- everyone is on DS 2 and all works fine (base = DS 1). However, with the clones being out there, there is no longer a "standard". As a result, you will often get a drive with the jumper wherever it was during final factory test (usually DS 1). If you have *no* twist, then drive 1 is set as Select 1 (base = 1 again), and drive 2 is set Select 2. IF there *IS* a twist, set BOTH drives to Select 2. The drive right after the twist in the cable (usually the end connector) will be drive #1, and the other one will be drive #2. All this "magic" is done by simply swapping the select lines in the wire.... (A note: It's always a good idea to check the jumpers on a drive before installation. Sometimes there are a few that aren't supposed to be there present -- I've even seen 'em come out of the factory packaging with TWO OR MORE select jumpers installed! This will make for one non-functional system, and if it's a secondary you're installing may scramble the primary on boot). Floppy drives work exactly the same way, as do the twists. See, when all the world was IBM all the techs just set the drives to Select 2 and off you went. This has the additional advantage in that the drive select is position independant -- you can swap selects just be changing the position on the cable. Note that floppy cables w/twists, while very similar in appearance, twist a different number of wires. DO NOT USE a twisted floppy cable for a HD -- all kinds of bad things can happen, although all you usually end up with is a drive that doesn't work. -- Karl Denninger (ddsw1!karl) Data: (312) 566-8912, Voice: (312) 566-8910 Macro Computer Solutions, Inc. "Quality solutions at a fair price"
wcf@psuhcx.psu.edu (Bill Fenner) (07/17/88)
In article <1393@ddsw1.UUCP> karl@ddsw1.UUCP (Karl Denninger) writes: | |Note that floppy cables w/twists, while very similar in appearance, twist a |different number of wires. DO NOT USE a twisted floppy cable for a HD -- |all kinds of bad things can happen, although all you usually end up with is |a drive that doesn't work. | Actually, I think this may be what happened to me. From a quick count, it looks like 10-16 are the lines that are twisted. Is this a floppy cable? If so, that would definitely explain my problems, no? :-) -- Bitnet: wcf@psuhcx.bitnet Bill Fenner | Internet: wcf@hcx.psu.edu | This space UUCP: {gatech,rutgers}!psuvax1!psuhcx!wcf | for rent Fido: Sysop at 263/42 |
karl@ddsw1.UUCP (Karl Denninger) (07/19/88)
In article <248@psuhcx.psu.edu> wcf@psuhcx (Bill Fenner) writes: >In article <1393@ddsw1.UUCP> karl@ddsw1.UUCP (Karl Denninger) writes: >| >|Note that floppy cables w/twists, while very similar in appearance, twist a >|different number of wires. DO NOT USE a twisted floppy cable for a HD -- >|all kinds of bad things can happen, although all you usually end up with is >|a drive that doesn't work. > >Actually, I think this may be what happened to me. From a quick count, >it looks like 10-16 are the lines that are twisted. Is this a floppy >cable? If so, that would definitely explain my problems, no? :-) I've got a dual floppy cable right here in front of me -- and it's twisted on pins 10-16.... This will definately NOT work for 2 hard disks..... The HD cables, from what I remember, only twist 4 lines instead of 6, although I don't remember the exact pins (we don't use them anymore if we can help it... although once in a while our third-party service runs into one of them). Use a straight cable (or make one from Radio Shack parts -- yep, they have everything you need including the 34-pin connectors!) and just set the drive selects -- it's much less hassle. -- Karl Denninger (ddsw1!karl) Data: (312) 566-8912, Voice: (312) 566-8910 Macro Computer Solutions, Inc. "Quality solutions at a fair price"
mikes@mntgfx.mentor.com (Mike Stanbro) (07/20/88)
From article <16800327@clio>, by berger@clio.las.uiuc.edu: > > I didn't translate all the alphabet soup you listed, but you mentioned > using a "twisted" 34 pin cable. That's what IBM supplies for their > floppy disk drives, but it should not be used with the hard disk. Use > a straight cable, and set the disk address select jumper on the > drive appropriately. > > If you didn't find any ROM at c800:0000, then you may have one of the > very old dump XT controllers. I don't know what you have to do to > get that to recognize a second drive. > > Mike Berger > Department of Statistics > Science, Technology, and Society > University of Illinois > > berger@clio.las.uiuc.edu > {ihnp4 | convex | pur-ee}!uiucuxc!clio!berger I experienced a similar problem the other night while trying to configure an HD for a friend. I was trying to get two identical HDs to run off of one controller so that I could do an XCOPY of all the data from drive C to drive D. I expected everything to plug together and be up and running in a matter of minutes ... two hours later (like you) I had tried every combination and could only get one drive at a time to operate. What really made it frustrating was that I had done this before and had things work just fine ... I was ready to spit fire!!! Then it dawned on me ... I was using a different 34 pin cable. I dug up the previous cable and compared it to the one I had been trying to use; they were different! There is a difference between the signals that are twisted for a floppy cable and an HD cable. I suspect that you are using a "twisted" type cable and that it is the floppy type instead of the HD type. Try to locate an HD type twisted cable and configure both drives as "DS1" or, better yet, use a non-twisted cable and configure the first drive as "DS1" and the second drive as "DS2". Either way, the terminator should be at the second drive (the end of the cable). I too recommend the non-twisted type of cable because it eliminates the confusion and it is the "normal" way of connecting two drives together. Some smart guy at IBM apparently conceived the idea of twisting the cable so that all the drives could be setup as DS1 and they didn't have to reconfigure the drives during manufacturing when there were two drives connected to one controller. By the way, you can get away with a non-twisted cable for two HDs but you are forced to use a twisted cable for two floppies; the floppy controller is specifically designed to work with the twisted cable. Good luck ... let me know if it works. -- Mike Stanbro, Research Engineer (503) 626-1437 Mentor Graphics Corp., 8500 SW Creekside Place, Beaverton OR 97005 ...!{sequent,tessi,apollo}!mntgfx!mikes OR mikes@pdx.MENTOR.COM These are my opinions, & not necessarily those of Mentor Graphics.