n138ct@tamuts.tamu.edu (Brent Burton) (02/27/91)
I just got my 7300 today. It has an intermittant problem, but first the
system:
7300, 2Meg Ram, no cards, WD2010 (no other fix, just the chip exchange)
and a 20Meg HD with "3.52jmC" operating system installed.
The machine CAN work fine. However, the problem is when it decides to work.
The machine came to me diskless. I had the 20Meg sitting around and so I
installed it. The machine came with the monitor and it's top case piece
separate from the body of the 7300. To install the drive, I had to take it
apart to see how it all fits together. OK, I finally got the drive installed
and I check everything (all wires, etc.) and then I turn it on and it boots
up fine. I login as root and issue a 'shutdown'. When things stop, I
assemble everything and move it to my desk (I was careful and didn't bang it).
I turn it on and all I get are the "marching boxes". Puzzled, I disassemble
it again on the floor and turn it on there -- it boots fine. I reassemble
it, put it on the desk and no boot action. (This is crazy, isn't it??)
I had talked with the original owner and the machine has worked fine for the
past year and I know it can work, however, something is going on here.
My ideas:
* I ran the HD cables through those rectangular slots in the upper mother-
board RF shield. Also, the power wire bundle for the HD was ran to the
back of the RF shield and up to the disk. Is this the correct path of
the cables?
* The owner mentioned something might be putting excess pressure on the
motherboard, however I have not seen anything that obviously would. I
was careful about the HD cables.
When it doesn't boot, the floppy light stays on, the boxes slowly progress
to the right on the screen, and all the LED's on the left are fully lit.
One time when it was disassembled and didn't boot, I wiggled the AC wires
from the power cable to the PS and the machine started up. I wiggled them
later on another no-boot and they had no effect (I really don't think they
are a cause).
What do the marching boxes mean and why doesn't it see the hard drive to start
loading? I have two sets of disk cables and both sets result in intermittant
boots.
Does anyone have any ideas? This is driving me crazy.
Thanks in advance!
+----------------------+--------------------------+
| Brent P. Burton | n138ct@tamuts.tamu.edu |
| Texas A&M University | Computer Science/Physics |
+----------------------+--------------------------+dnichols@ceilidh.beartrack.com (DoN Nichols) (02/28/91)
In article <12744@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >I just got my 7300 today. It has an intermittant problem, but first the >system: > 7300, 2Meg Ram, no cards, WD2010 (no other fix, just the chip exchange) > and a 20Meg HD with "3.52jmC" operating system installed. What is "3.52jmC" ?? Where did you get it ??? What does it do different ???? >The machine CAN work fine. However, the problem is when it decides to work. > >The machine came to me diskless. I had the 20Meg sitting around and so I >installed it. The machine came with the monitor and it's top case piece Is this the standard Miniscribe 20MB with the slow-boat-to-china seek time? If not, is it in the shielding can which came with the 7300? If it is, is it possible that the pressure of the grounding spring from the center of the monitor pivot is pressing the can in such a way as to short something? If not in the can, is it possible that the grounding spring is shorting something directly? >separate from the body of the 7300. To install the drive, I had to take it [...] >My ideas: > * I ran the HD cables through those rectangular slots in the upper mother- > board RF shield. Also, the power wire bundle for the HD was ran to the > back of the RF shield and up to the disk. Is this the correct path of > the cables? The power cable should also come through the rectangular slot, if it is coming from the motherboard. If it is coming from the power supply, it must have had a 3B1 power supply installed. (Come to think of it, the 2MB on mother board also sounds more like a 3B1) Does it have the extra hump in the center of the top to clear a full-height drive? Another possibility is that you have too much stress on the power cables because the hard disk or floppy are slid too far back in the range of possible positions. (This is usually more of a problem for the floppy.) If the drive is not the standard Miniscribe, the position of the power connector may result in unusual stresses also. If the sheild can is canned, I also remove the grounding fingers that touch the back of the shield can. (They may be shorting something when the case is in place.) In the 3B1 that I have modified, they are held on the modem-shield by pop-rivits, in the 7300, the connection was made by melting the two together at the edges of the holes where the pop-rivits would normally go. In either case, they may be removed by careful drilling, with a drill just a little bigger than the hole with the melted join, or a little smaller with the pop-rivits. DON'T do it in the machine!!! And BE SURE to use a vise to control all that sharp sheet metal. [...] >One time when it was disassembled and didn't boot, I wiggled the AC wires >from the power cable to the PS and the machine started up. I wiggled them >later on another no-boot and they had no effect (I really don't think they >are a cause). You might check the voltages from the power supply during a no-boot condition. >What do the marching boxes mean and why doesn't it see the hard drive to start They mean that it can't find anything to read on either disk. (If the +12V is shorted, neither disk drive will spin -- but the monitor shouldn't work either -- it is powered by the +12V. One other possibility is that the ribbon cable and connector to the power supply is not making good contact when the side of the case presses the bow of the cable inward. (You might try pressing it in with your hand while the system is open and powered up.) If this makes a difference, you might need to clean the connector pins with something like an eraser. First, look for signs of overheating, which can point to a poor connection. >loading? I have two sets of disk cables and both sets result in intermittant >boots. > >Does anyone have any ideas? This is driving me crazy. > >Thanks in advance! Hope one of the things that I have mentioned helps! Good Luck DoN. -- Donald Nichols (DoN.) | Voice (Days): (703) 664-1585 D&D Data | Voice (Eves): (703) 938-4564 Disclaimer: from here - None | Email: <dnichols@ceilidh.beartrack.com> --- Black Holes are where God is dividing by zero ---
n138ct@tamuts.tamu.edu (Brent Burton) (03/01/91)
I've been messing with the system trying to figure out the problem and
the one thing I have learned to do is disassemble the system in record time!
:-) I've owned the machine two days and I bet I've opened it up more times
than most of you...
As far as the system problem goes, it was able to boot off the disk SOMETIMES
when the disk was outside the 7300. Eventually, I decided to reformat the
drive from the diagnostics menu (what could I lose?).
I reformatted it OK, even installed 3.51 with no problems. The 7300 would
also boot from the disk and run perfectly.
I then parked the heads (after shutting the system down) and transferred
the disk to the 7300 drive bay. Now it would not boot from it.
Diagnostics eventually said it needed to be re-initialized, so I selected
the format drive option under the menu. The drive (20M miniscribe, 612 c and
4 h) formatted OK. Now the fun begins...
The next messages for me were:
Error: Can't recal:response=10
Bad block table write failed. Disk needs to be re-initialized.
Test: Surface Test
ERROR:Can't recal:response=10
Test:Park Disk Heads
ERROR:Can't recal:response=10
Then, the drive parks the heads OK.
Any clues? What is a 'recal'? Could it possibly be a problem
with the drive?
Thanks for any comments.
+----------------------+--------------------------+
| Brent P. Burton | n138ct@tamuts.tamu.edu |
| Texas A&M University | Computer Science/Physics |
+----------------------+--------------------------+botton@i88.isc.com (Brian D. Botton) (03/01/91)
In article <12744@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: Lots of stuff skipped over. >What do the marching boxes mean ... The story of the marching boxes is this: After the boot ROM initializes the hardware it jumps to an infinate loop and draws the first square. It then tries to boot from the floppy. If it cannot, it tries to boot from the hard disk. If it cannot, it goes back to trying the floppy. Each time it starts the loop over it draws a new square. The floppy's LED stays on because it gets "select"ed at the beginning of the infinate loop and thus says on. When the loader is finally loaded into RAM, both the floppy and hard disk are "deselect"ed. BTW, if you look in /usr/include/sys/machine.m you will see that there are 3 disk types, winchester (HD), floppy (FD) and syquest (SY). It is fairly obvious from the boot ROM and loader code that support for more than the HD and FD was intended and easy to add. Does anyone know what a "syquest" is? > and why doesn't it see the hard drive to start >loading? I have two sets of disk cables and both sets result in intermittant >boots. The hard disk must go into a ready state before it can be accessed. Is the circuit card on the hard disk shorting to the cabinet when all the screws are tight? Look for cables that are being stretched or something shorting the hard disk. -- ... ___ *** _][_n_n___i_i ________ ******* Brian D. Botton (____________I_I______I_I_______I laidbak!botton or /ooOOOO OOOOoo oo oooo oo oo laidbak!bilbo!brian
n138ct@tamuts.tamu.edu (Brent Burton) (03/01/91)
Well, another update to the adventure.
A friend, Daryl, brought his 3b1 over and removed his drive; we
installed my 20M into his machine, started it up with the diagnostics
and eventually were able to add the foundation set OK. His 3b1 booted
from my disk no problem -- we know the disk is truly not bad.
We *quickly* moved the 20M drive to my system, where it booted first.
We shutdown, reassembled the machine, tried to reboot and it failed again
(marching blocks).
We tried 2 other times with failure. Then, we decided to check the power
supply voltages and the machine booted (this is insane). Two other
attempts at rebooting worked fine and it's running right now!!!
The power supply is providing the following:
spec'd voltage actual
------------- ------
+5.0V 4.97
+12.0V 11.95
-12.0V -12.00 (taken with new digital meter)
Where should these values lie and when I need to adjust them, which
pots on the ps board do I mess with? (so many pots, so little patience)
Help.
+----------------------+--------------------------+
| Brent P. Burton | n138ct@tamuts.tamu.edu |
| Texas A&M University | Computer Science/Physics |
+----------------------+--------------------------+n138ct@tamuts.tamu.edu (Brent Burton) (03/01/91)
Well, I shutdown the system again (have you heard this phrase from me or what?)
and reassembled it. Turned it on and no boot. We then quickly ripped the
top off to measure the power supply levels and we found a problem.
The voltages were the same as I mentioned in my last posting.
EXCEPT the -12V (pin 18 on the PS-->MB cable) was +0.79V. This is
a bad sign. What hints are there?
+----------------------+--------------------------+
| Brent P. Burton | n138ct@tamuts.tamu.edu |
| Texas A&M University | Computer Science/Physics |
+----------------------+--------------------------+thad@btr.BTR.COM (Thaddeus P. Floryan) (03/01/91)
In article <12843@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: > >Error: Can't recal:response=10 > Bad block table write failed. Disk needs to be re-initialized. > >Test: Surface Test >ERROR:Can't recal:response=10 > >Test:Park Disk Heads >ERROR:Can't recal:response=10 > >Then, the drive parks the heads OK. > >Any clues? What is a 'recal'? Could it possibly be a problem >with the drive? Two suggestions: 1) check if the disk controller chip is a WD2010 instead of a WD1010. This is a 40-pin chip (mfd by Western Digital) and is located several inches toward the left rear of the motherboard from the CPU. Your system "may" require the "DRUN Patch" I posted back in December (assuming the HD itself is OK and is spinning) 2) check if the HD is spinning when you power up. Many "older" HDs and most HDs from a certain manufacturer (who shall remain nameless but whose name begins with the letters "Seagate" :-) will exhibit a phenomenom termed stiction (from static friction). Essentially, the heads are "glued" to the platters in the park zone. Often the condition can be temporarily "fixed" by twisting (using one's finger or the eraser end of a pencil) the shaft on the head-stepper-motor a few degrees, then powering up the drive and getting one's data off ASAP. I have over 3MB worth of data on this subject already (which I started collecting after all 11 of my Seagate drives (ST-251 and ST-157) encountered the problem only weeks after their warranty periods expired). And by twisting the spindle, I do NOT mean to open up the drive! Simply hold the drive upside down so its electronics are facing up, and you should see a central main-spindle and, off to one side, the stepper motor shaft. You do NOT want to attempt the 2" drop-kick start as suggested by others recently! :-) Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]
thad@btr.BTR.COM (Thaddeus P. Floryan) (03/01/91)
In article <1991Mar1.045750.19952@i88.isc.com> botton@i88.isc.com (Brian D. Botton) writes: > BTW, if you look in /usr/include/sys/machine.m you will see that there are >3 disk types, winchester (HD), floppy (FD) and syquest (SY). It is fairly >obvious from the boot ROM and loader code that support for more than the >HD and FD was intended and easy to add. Does anyone know what a "syquest" is? A "syquest" is a hard drive with a removable platter larger than but resembling the 3.5" floppies which are quite popular on many systems. Believe it or not, I was running a Syquest as the primary HD on one 3B1 on which I was doing a lot of testing several years ago ... it was easier just to pop in a new cartridge with a different kernel and/or other files than it was to de-/re-cable different HDs. I don't recommend anyone doing this (using a Syquest) unless you're fully aware of the consequences: the kernel doesn't like to have the pager disk removed while UNIX is running! :-) Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]
thad@btr.BTR.COM (Thaddeus P. Floryan) (03/01/91)
In article <12877@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >[...] >The power supply is providing the following: > >spec'd voltage actual >------------- ------ > +5.0V 4.97 > +12.0V 11.95 > -12.0V -12.00 (taken with new digital meter) > >Where should these values lie and when I need to adjust them, which >pots on the ps board do I mess with? (so many pots, so little patience) Your voltages are well within spec (+/- 5%). The ranges are: 4.75 <= 5.00 <= 5.25 11.4 <= 12.00 <= 12.6 I prefer to run mine slightly high (5.1 and 12.2) to help offset any drops in PC trace runs, esp. to the expansion boards. The adjustment points (ONLY FOR +5 and +12 (the -12 is provided by a fixed, non-adjustable IC regulator)) are silk-screened on the power supply per "+12 ADJ." and "+5 ADJ." alongside their respective pots. The +12 ADJ point is about 2" in from the front of the power supply and about 1" inwards from the 18-pin connector to which the ribbon cable from the motherboard connects. The +5 ADJ point is about 2" towards the rear from the +12 ADJ point. Be SURE to use a plastic or non-conductive shaft screwdriver when making any adjustments, and be VERY careful. Counter-clockwise reduces the voltage, and clockwise increases the voltage. The power supply MUST be loaded-down when making adjustments, and what I recommend is a "Y"-cable temporarily attached to the HD with one set of wires going to the power supply and the remaining set being available for your DVM probes. Such "Y"-cables (with three 4-pin Molex-style connectors at each end of the "Y") can be found at most electronics and computer stores, swap meets, ham fests, surplus houses, etc. and "should" not cost more than about $1. Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]
thad@btr.BTR.COM (Thaddeus P. Floryan) (03/01/91)
In article <12880@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >[...] >The voltages were the same as I mentioned in my last posting. >EXCEPT the -12V (pin 18 on the PS-->MB cable) was +0.79V. This is >a bad sign. What hints are there? Sounds like a bad -12V IC regulator. As I recall, the -12V regulator is right at the front of the power supply, so replacing it should be straightforward. Not too much on the system uses -12VDC, and I don't believe anything could have been damaged by the voltage you measured, but what concerns me is what caused the IC regulator to go belly-up. If it's a problem with other circuitry in your system, then you gotta problem. Maybe temporarily using an external power supply may help isolate the trouble area if there is a problem on the motherboard. If you need the power supply pinouts (part of one of my regular postings in this regards), send email. Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]
pusateri@duke.cs.duke.edu (Thomas J. Pusateri) (03/02/91)
In article <12877@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >We *quickly* moved the 20M drive to my system, where it booted first. >We shutdown, reassembled the machine, tried to reboot and it failed again >(marching blocks). > >We tried 2 other times with failure. Then, we decided to check the power >supply voltages and the machine booted (this is insane). Two other >attempts at rebooting worked fine and it's running right now!!! What does it do when it doesn't boot again? I missed that part. I had a problem that I open the case, jingle some cables and get it to boot and then when I put the case back on, it quick booting. (Mine displayed this alternating pattern across the screen). Anyway, I finally figured it out thanks to some help from the net. By reseating the 68010 cpu, I have never had this problem again. Tom pusateri@nbsr.duke.edu pusateri@cs.duke.edu
guy@cbnewsc.att.com (guy.r.berentsen) (03/02/91)
In article <1991Feb28.020612.11229@ceilidh.beartrack.com>, dnichols@ceilidh.beartrack.com (DoN Nichols) writes: > In article <12744@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: > >I just got my 7300 today. It has an intermittant problem, but first the > >system: > > 7300, 2Meg Ram, no cards, WD2010 (no other fix, just the chip exchange) > > and a 20Meg HD with "3.52jmC" operating system installed. > > What is "3.52jmC" ?? Where did you get it ??? What does it do > different ???? > The machine on my desk at work has been running 3.52jmC for more than 2 years (It was installed by the previous owner.) I was told that this kernel was someone's private build, including the same fixes that were released in the 3.51 fix disk. I don't think that this kernel was ever officialy distributed. I believe the letters "jmC" are the initials of the person that developed or built the fixes.
dnichols@ceilidh.beartrack.com (DoN Nichols) (03/02/91)
In article <1991Mar1.045750.19952@i88.isc.com> botton@i88.isc.com (Brian D. Botton) writes: [ ... ] > BTW, if you look in /usr/include/sys/machine.m you will see that there are >3 disk types, winchester (HD), floppy (FD) and syquest (SY). It is fairly >obvious from the boot ROM and loader code that support for more than the >HD and FD was intended and easy to add. Does anyone know what a "syquest" is? Yep! It is/was a removable-media 5.25" hard disk. I think that it was ST-506 interface, but the stuff in /usr/include/sys/machine.m ? (I don't find 'machine.[mh]' in my /usr/include/sys, but it may have been blown away in some backup/restore operation. I'll have to check the machine with the STOCK 3.51 on it.) I've seen them at hamfests, but avoided purchasing one since I understand that the media are made of hens' teeth. :-) (I once bought some 3.25" NOT 3.5" floppy drives, and found ONE(1) floppy for them in the next two years. :-) If media were available, these would make nice backup/transfer media, since I think that they were 20MB or so. (I think that the media were outrageously expensive, which may have been a factor leading to their decline. Another was a reputation for poor interchangability of media between drives.) If I'm wrong about this, let me know, and I'll pick up one at the next hamfest, and install it as a second drive in place of the floppy (with ICUS mod in place). Good Luck DoN. -- Donald Nichols (DoN.) | Voice (Days): (703) 664-1585 D&D Data | Voice (Eves): (703) 938-4564 Disclaimer: from here - None | Email: <dnichols@ceilidh.beartrack.com> --- Black Holes are where God is dividing by zero ---
dnichols@ceilidh.beartrack.com (DoN Nichols) (03/02/91)
In article <12877@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >Well, another update to the adventure. > [ ... ] >We tried 2 other times with failure. Then, we decided to check the power >supply voltages and the machine booted (this is insane). Two other >attempts at rebooting worked fine and it's running right now!!! > >The power supply is providing the following: > >spec'd voltage actual >------------- ------ > +5.0V 4.97 > +12.0V 11.95 > -12.0V -12.00 (taken with new digital meter) > >Where should these values lie and when I need to adjust them, which >pots on the ps board do I mess with? (so many pots, so little patience) Only the +5V and +12V are adjustable. Probably you should aim to be a "lettle grey cells" (oops, a little high - watching too much Poirot :-). Probably +5.1 and +12.2. The pots ARE labeled, but the labels tend to be hard to see from any reasonable angle of view. Looking at a 3B1 power supply, +12V is near the center of the multi-pin output power connector, and +5V (R-108) is near the inboard end of the same connector, with its label under a yellow and a black wire, and the curve of U1 (whatever that is - maybe an opto-isolator :-). If you ever HAVE to adjust the -12V, do it by replacing U101 (the regulator chip). Good Luck DoN. -- Donald Nichols (DoN.) | Voice (Days): (703) 664-1585 D&D Data | Voice (Eves): (703) 938-4564 Disclaimer: from here - None | Email: <dnichols@ceilidh.beartrack.com> --- Black Holes are where God is dividing by zero ---
dnichols@ceilidh.beartrack.com (DoN Nichols) (03/02/91)
In article <12880@helios.TAMU.EDU> n138ct@tamuts.tamu.edu (Brent Burton) writes: >Well, I shutdown the system again (have you heard this phrase from me or what?) >and reassembled it. Turned it on and no boot. We then quickly ripped the >top off to measure the power supply levels and we found a problem. > >The voltages were the same as I mentioned in my last posting. >EXCEPT the -12V (pin 18 on the PS-->MB cable) was +0.79V. This is >a bad sign. What hints are there? Either the -12V is shorted somewhere, or the regulator chip is defective/damaged. I mentioned the location in my previous article, but who knows which will reach you first. At the end of the power supply board, near the output connector, is a small heat-sink labled on the board U101. It is a -12V regulator chip, probably a 7912 or moral equivalent. This regulator/heat-sink combination can be fairly easily hit by the case when re-installing, especially if the hard drive is not present, and if it bends in one direction, it can break the leads, or pull them out by the roots. Try gently pressing the top of this heat-sink one way and the other. If this kills/restores your -12V, you'll need to unsolder the heat sink and the chip, replace the chip, and re-solder. (I had this happen, but the leads broke at the pc-board, so I was able to wrap buss-wire around the terminals, and make new leads. It works fine, now. The failure was catastrophic, rather than intermittent, so the remedy was more apparent.) Hope that this cures your problems. DoN. -- Donald Nichols (DoN.) | Voice (Days): (703) 664-1585 D&D Data | Voice (Eves): (703) 938-4564 Disclaimer: from here - None | Email: <dnichols@ceilidh.beartrack.com> --- Black Holes are where God is dividing by zero ---
dlb5404@tamuts.tamu.edu (Daryl Biberdorf) (03/08/91)
In article <1972@public.BTR.COM> thad@btr.BTR.COM (Thaddeus P. Floryan) writes: > >A "syquest" is a hard drive with a removable platter larger than but resembling >the 3.5" floppies which are quite popular on many systems. > >Believe it or not, I was running a Syquest as the primary HD on one 3B1 >on which I was doing a lot of testing several years ago ... it was easier Are the Syquest drives available in the ST-506/MFM models? The only ones I've seen are SCSI models. These would be a NEAT alternative for those of us who are contemplating the HD2 upgrade. I think this would have nice uses for backups, considering the scarcity and expense (and low storage capacity) of the 3b1 tape setup. Man, when is that Mondo Combo SCSI card gonna be ready? I want to put some of these guys on my 3B1! --Daryl Biberdorf, dlb5404@{rigel,tamuts}.tamu.edu Texas A&M University
thad@public.BTR.COM (Thaddeus P. Floryan) (03/11/91)
In article <13156@helios.TAMU.EDU> dlb5404@tamuts.tamu.edu (Daryl Biberdorf) writes: >Are the Syquest drives available in the ST-506/MFM models? The only ones >I've seen are SCSI models. All "modern" Syquest drives I've seen are SCSI; the one I have is an oldie whose platter holds only 10MB. The drive itself is ST-506 and formats and operates just fine on the 3B1. I bought it used from Weird Stuff Warehouse several ago and don't know if they have any more; the 10 MB media was a pain to find and the most I was ever able to locate were approx. 30 disks, so I don't use this drive for anything "serious." >Man, when is that Mondo Combo SCSI card gonna be ready? I want to put >some of these guys on my 3B1! One guy in our Users' Group apparently has a prototype card using a 68HC11 and some new WD chip on a daughterboard ... the whole assembly plugs into the WD1010 socket on the 3B1 motherboard, and apparently supports 2,7 RLL and ESDI. He claims to be adding SCSI and Ethernet support. He didn't have any updates at the February meeting, so I'll ask him again later this month. Thad Floryan [ thad@btr.com (OR) {decwrl, mips, fernwood}!btr!thad ]