Jump to content

IBM PS/2 Model 80 Troubleshooting Guide: Difference between revisions

From RetroTechCollection
โ† Older edit

Josh (talk | contribs)

5 user groups
10,226 edits
undo
VisualWikitext
Add representative photo (Wikimedia Commons, attributed on file page)
Rebuild against Ardent Tool's three 8580 planar pages: there are three planar types not two, and the -111/-311 models were on the wrong one; 'seven 32-bit MCA slots' and 'one 8-bit MCA slot' do not exist (Types 1/2 have three 32-bit and five 16-bit, Type 3 four and four); the 70 ns memory requirement is unsourced (both documented kits are 80 ns); replace the three invented ECAs with the real ECA 051 (8580 memory parity errors, FCT841 modules at U24/U34) in IBM's own words; '161 - replace DS12...
ย 
Line 1: Line 1:
<templatestyles src="Template:StyledTable/styles.css" />
[[File:IBM PS-2 Model 80 (photo).jpg|thumb|right|300px|IBM PS/2 Model 80. Source: Wikimedia Commons.]]
[[File:IBM PS-2 Model 80 (photo).jpg|thumb|right|300px|IBM PS/2 Model 80. Source: Wikimedia Commons.]]
This guide documents fault diagnosis for the '''[[IBM PS/2 Model 80]]''' (machine type 8580, all submodels). The Model 80 shares its POST architecture with the [[IBM PS/2 Model 70]] (same planar family) but has 8-slot MCA, server-class ESDI configurations and Type 1 / Type 2 planar differences that affect troubleshooting.


== Reference Diskette and Diagnostics ==
Fault diagnosis for the '''[[IBM PS/2 Model 80]]''' (machine type 8580). The Model 80 shares its POST architecture with the [[IBM PS/2 Model 50]].


The Model 80 requires the '''Model 80-specific Reference Diskette'''. The Type 1 planar (8580-041 / 071 / 111 / 311) uses one Reference Diskette image; the Type 2 planar (8580-A21 / A31) uses a different Reference Diskette image. Use of the wrong Reference Diskette will refuse to boot or will mis-configure the planar.
== Which planar do you have ==


Boot Options:
'''There are three documented 8580 planars, not two''', and an earlier version of this page assigned the models to the wrong ones:<ref name="ard_t1" /><ref name="ard_t2" /><ref name="ard_t3" />


* '''F1''' โ€” boot the Reference Diskette.
{| class="wikitable styled-table" style="width:100%; text-align:left;"
* '''Ctrl-A''' from the Reference Diskette menu โ€” Advanced Diagnostics.
|+'''8580 planars'''
ย 
! Type !! Speed !! Models !! Memory kit !! Slots
== POST Sequence ==
ย 
The Model 80 POST runs in this order:
ย 
# Reset; CPU register check (80386 includes paging and protected mode tests).
# ROM checksum.
# CMOS / RTC battery check.
# Planar RAM count.
# Onboard VGA initialisation.
# Floppy controller and drive seek.
# MCA adapter ID scan (each card returns a 16-bit ID; up to 8 cards).
# ADF lookup against CMOS configuration record.
# ESDI controller initialisation.
# Boot device selection.
ย 
== Beep Codes ==
ย 
{| class="wikitable styled-table" style="width:100%; text-align:center;"
|+'''Model 80 beep codes'''
! Beeps !! Meaning
|-
|-
| 1 short || POST passed; normal boot.
| '''Type 1''' || '''16 MHz''' || -041, -071 || 1 MB '''80 ns''' kit 72X6670 or 6450375 || J1, J2, J4 32-bit with MME; J3, J5, J7, J8 16-bit; J6 16-bit with AVE
|-
|-
| 2 short || Configuration error; numeric error on screen.
| '''Type 2''' || '''20 MHz''' || -081, -111, -121, -161, -311, -321 || 2 MB '''80 ns''' kit 92F0669 or 6450379 || Same as Type 1
|-
|-
| 1 long, 1 short || Planar fault.
| '''Type 3''' || '''25 MHz''' || -A16, -A21, -A31 || Up to 8 MB on the planar using 4 MB System Memory Boards || J10โ€“J13 32-bit with MME; J6, J7 16-bit with AVE; J14, J15 16-bit
|-
| 1 long, 2 short || Display adapter fault.
|-
| Continuous || Power supply or planar fault.
|-
| None, no display || Planar or PSU fault before video init.
|}
|}


== Numeric POST Codes ==
Planar FRUs: Type 1 33F8416 / 72X7492; Type 2 33F8506 / 92F0756 / 92F0767; Type 3 34F0022 / 33F5571 / 85F0046 / 91F7513.<ref name="ard_t1" /><ref name="ard_t2" /><ref name="ard_t3" />


The Model 80 shares the same 1xxโ€“104xx code families with the Model 70. Codes that are identical to the Model 70 are summarised here; refer to [[IBM PS/2 Model 70 Troubleshooting Guide]] for the complete tables.
'''There is no 8-bit Micro Channel slot on any Model 80''', and there are never seven 32-bit slots. Types 1 and 2 have '''three''' 32-bit and five 16-bit; Type 3 has '''four''' 32-bit and four 16-bit, two of the 16-bit ones carrying the AVE extension and one normally occupied by the 16-bit SCSI controller.<ref name="ard_t1" /><ref name="ard_t3" />


=== 1xx โ€” Planar / System Board ===
=== What is distinctive about the Type 3 ===


Same as Model 70. Key codes:
The 25 MHz planar is a Japanese design, shared with the PS/55 Model 5570-Vxx, and is the only one with:<ref name="ard_t3" />


* '''104''' โ€” Protected-mode failure. Often planar SMD cap leakage near the CPU.
* A '''386DX-25''' at U96, a socket for a 387DX-25 at U165, and '''64 KB of L2 cache''' behind an '''82385-25 cache controller''' at U54 with eight M5M5178P-33 8K ร— 8 SRAMs at U99โ€“U106. This is the planar on which IBM's POST code '''0001 91XX, "82385 cache test failed"''', can actually apply.<ref name="ard_err" />
* '''114''' โ€” Paging test failure (80386-specific).
* '''DMA support on the parallel port''' โ€” the A21 and A31 parallel ports support DMA parallel communications.
* '''161 / 162 / 163''' โ€” CMOS / RTC battery cluster. Replace DS12887.
* '''Real fuses on the planar''': F2 1 A, F3 1.5 A, F4 2.0 A, with solder pads at F1. On the Type 1 and Type 2 planars IBM's own note for the keyboard voltage error still applies โ€” if there is no fuse, replace the system board.<ref name="ard_err" />
* '''164''' โ€” Memory size mis-match with CMOS. Run SETUP.
* A socketed 8042 keyboard controller at U59, an MC146818AFN RTC at U132, and a 90X8941 VGA at U16.
* '''165''' โ€” MCA adapter ID mismatch. Run Auto Configuration. '''More common on Model 80 than Model 70 due to higher card count.'''
* '''166''' โ€” MCA arbitration failure.


=== 2xx โ€” RAM ===
== Reference Diskette and diagnostics ==


Same as Model 70. The Type 2 planar (8580-A21 / A31) is more particular about SIMM speed and rejects slower SIMMs with a 201 / 225 cluster.
Use the Reference Diskette matching the planar type. A mismatched diskette will misreport the memory configuration and may refuse to complete configuration.


=== 3xx โ€” Keyboard ===
* '''F1''' boots the Reference Diskette; '''Ctrl-A''' from its menu enters Advanced Diagnostics.
* A four-digit numeric code at the top left of the screen is a fault code from the test block that was running, and is the FRU lookup key.
== POST beep codes ==


Same as Model 70.
IBM's PS/2 audio codes are five patterns, and three of the five mean the same thing:<ref name="ard_err" />


=== 6xx โ€” Floppy Drive ===
{| class="wikitable styled-table" style="width:85%; text-align:left;"
|+'''PS/2 beep codes''' (* = short, _ = long)
! Pattern !! Meaning
|-
| <code>*</code> || Normal operation
|-
| <code>* *</code> || A problem exists โ€” check the configuration. Run diagnostics if no error code is displayed
|-
| <code>_ *</code>, <code>_ * *</code>, <code>_ * * *</code> || Display or display adapter problem
|-
| Continuous short beeps || Check the keyboard
|-
| Continuous solid beep || System board
|}


Same as Model 70.
An earlier version of this page gave a different table in which the long-plus-short patterns were split between "planar fault" and "display adapter fault", and added rows for "continuous โ€” power supply or planar" and "none, no display". IBM's list does not make those distinctions; the no-beep case is dealt with under [[#Dead system|Dead system]].


=== 8xx โ€” Math Coprocessor ===
== Numeric POST error codes ==


The Model 80 supports the optional '''80387DX''' coprocessor.
'''This wiki keeps one copy of IBM's PS/2 numeric error code list, on the [[IBM PS/2 Model 50 Troubleshooting Guide#Numeric POST error codes|Model 50 guide]].''' It is transcribed from the Ardent Tool of Capitalism's reproduction of IBM's own symptom-and-FRU lists and keeps IBM's FRU / action column.<ref name="ard_err" />


* '''801''' โ€” Coprocessor test failed.
'''Do not work from a per-machine summary of these codes.''' The tables previously on this page were hand-transcribed, mixed PC/AT-era meanings into PS/2 codes, and โ€” in the ESDI family in particular โ€” contained a consecutive run of codes of which ten do not exist in IBM's list and several more had the wrong meaning. They have been replaced by the link above rather than corrected in place, so that there is one copy to keep right.


=== 104xx โ€” ESDI Fixed Disk Adapter/A ===
Three specific corrections worth carrying in your head:


The Model 80 uses the '''IBM ESDI Fixed Disk Adapter/A''' (or its later revision for the larger 314 MB drives) for the hard drive subsystem. The 104xx error family is the same as on the Model 70.
* '''The math coprocessor family is 7xx, not 8xx.''' This page previously headed it "8xx โ€” Math Coprocessor" with a single 801 code. IBM's coprocessor family is 0007 XXXX; 0008 does not exist.
* '''104 is not a protected-mode failure.''' IBM's protected-mode code is 115. 104 appears only grouped with 102, 103 and 107 against a system board or processor board FRU.
* '''162 is not a checksum error.''' It tells you to check that devices are powered on and whether the configuration has changed. The bad-checksum code is 173, and IBM's note there is "possibly a weak battery".


See [[IBM PS/2 Model 60 Troubleshooting Guide]] for the complete 104xx code table.
== Dead system ==


Server configurations (8580-311 / A31) with two ESDI drives use:
IBM's stated most likely cause of no POST, a blank screen and no beep is '''a device shorting out the power supply'''. '''The system board must present Power Good to the supply within 150 ms or the supply shuts itself down internally''' โ€” which is the mechanism behind the "click and then nothing" symptom. IBM's usual suspects are a SIMM in backwards, an adapter not fully seated, the power cord not plugged in, an unsupported or defective adapter, or too many drives spinning up at once on a heavily loaded machine (leave the motor-start jumper open so the drives start after the controller interrogates them).<ref name="ard_err" />


* '''10480''' โ€” Drive 0 fatal error.
# '''Check the line cord at both ends.''' IBM singles this out: intermittent faults, including unexpected power-on resets during operation, are caused by line cords that are not fully seated or too loose to make a tight connection. Remove the cord from the outlet before working on its contacts.
* '''10481''' โ€” Drive 1 fatal error.
# Remove all adapters, options and extra memory.
# Remove all external connections.
# Power on and listen for a beep.
# '''No beep:''' check speaker continuity; if good, replace the system board memory and retry; if still silent, measure the rails โ€” correct rails mean the system board, wrong rails mean the supply.
# '''A beep:''' refit adapters one at a time, retrying after each. When the beep stops, the last item fitted is the suspect.


This is the only place the second drive is called out separately in POST.
On machines fitted with '''LogicLock''', the access-detect switches can be left in the wrong position during reassembly, producing a machine that appears dead.<ref name="ard_err" />


=== 24xx โ€” Onboard VGA ===
== Diagnostics procedure ==


Same as Model 70:
* '''Cold-boot the Reference Diskette.''' Soft-booting it can report false errors and falsely indicate that a power-on password is set.<ref name="ard_err" />
* '''If several errors appear, diagnose the first one first''' โ€” the first can cause the others, especially with configuration errors.
* '''Run Advanced Diagnostics (Ctrl-A from System Programs) before replacing anything.''' If the diagnostics do not fail, replacing parts probably will not help.
* On a multi-FRU adapter, remove the optional FRUs one at a time before replacing the adapter.
* '''External surge suppressors''' are a documented cause of hard-to-diagnose faults.
== Battery ==


* '''2401''' โ€” Onboard VGA POST failure.
'''This machine takes a 6 V lithium battery pack, IBM FRU P/N 72X8498''' โ€” a two-cell pack, usually yellow with black ends, held in a dedicated Battery/Speaker Assembly. It is a camera battery: equivalents are the '''CR-P2''' (Varta, Panasonic, GP, Sony, Sanyo), '''DL223A''' (Duracell), '''EL223AP''' (Ever Ready), '''K223LA''' (Kodak) and '''PC223''' (Ray-O-Vac).<ref name="ard_batt" />
* '''2402''' โ€” VGA video memory failure.
* '''2410''' โ€” Planar VGA card failure.


A 24xx on a Model 80 is one of the more common SMD electrolyte leak symptoms.
'''There is no Dallas RTC module in this machine.''' Earlier versions of this page told the reader to replace a DS12887 module on a "late planar". The Dallas self-contained module (IBM FRU 8509237, Dallas DS1287) is fitted to the Models 25-286, 30-286, 35, 40, 55 SX, 55 LS and 65 SX โ€” not to the 50, 50 Z, 60, 70, 80, P70 or P75, all of which use the 6 V pack.<ref name="ard_batt" />


== SMD Electrolyte Leak Diagnostic Workflow ==
'''Defective or weak batteries cause the loss of all setup information. If only part of the setup is lost, the battery is probably NOT the cause.''' Inaccurate time is usually software, though the older models with 6 V packs are prone to the clock running slow.<ref name="ard_batt" />
== The real Model 80 ECA: 051, memory parity errors ==


The Model 80 planar's SMD electrolytic capacitors are the leading cause of "the Model 80 won't POST" complaints. The same symptoms apply as on the Model 70:
'''This is the one Engineering Change Announcement that can be checked, and it is a good one to know.'''<ref name="ard_t3" />


* '''104''' (protected mode) โ€” SMD leak near the CPU.
'''M/T 8580 Models A21, A31 and A16 โ€” intermittent OS/2 trap errors.''' 25 MHz 8580 systems with 8 MB of system board memory fitted, experiencing intermittent planar memory parity errors such as OS/2 Trap 0002 or NMI errors under other complex operating systems.
* '''201''' (memory) โ€” SMD leak near the SIMM controller.
* '''2401''' (VGA) โ€” SMD leak near the VGA chip.
* Random reboots โ€” SMD leak on a bus signal.
* No POST at all โ€” SMD leak shorting a power rail.


If a Model 80 is exhibiting any persistent or intermittent POST fault, '''inspect the planar for SMD electrolyte leakage before any other diagnosis'''. Read [[IBM PS/2 Model 80 Capacitor Replacement Guide]] for the recap procedure.
'''Physical check:''' examine the modules in positions '''U24''' and '''U34''', near the system board memory sockets J8 and J9. '''If the part number on the module is FCT841, the system board qualifies for replacement under this ECA.'''


The Model 80 has 40โ€“50 SMD electrolytics (more than the Model 70's ~36โ€“40) because of the larger MCA bus controller and server-class memory subsystem. The likelihood of at least one SMD cap leaking is correspondingly higher.
IBM's own caution is important: '''ignore the information printed on the system board near the module โ€” use only the information printed on the module itself.''' Modules with any other identification (F841, 841D, 72F841, 72F841D and so on) are good and the board does not qualify.


== MCA-Specific Faults ==
The underlying fault is a timing condition on the planar. Only the three 25 MHz models listed are affected; other 8580 models are not eligible.


=== 165 Card-ID Mismatch ===
'''What has been removed:''' this page previously named '''ECA 087''', '''ECA 092''' and '''ECA 117''' with Model 80 meanings. ECA 092 is documented as the '''8570''' Data Modification Patch โ€” a hardfile data-transfer patch for 25 MHz Model 70 machines โ€” not a Model 80 planar fix.<ref name="ard_8570" /> ECA 087 and ECA 117 could not be corroborated anywhere.


With 7 ร— 32-bit MCA slots typically populated to 3โ€“5 cards in a Model 80 (vs. 1โ€“2 on a Model 70), the 165 fault is more common. Causes and fix are the same as on the Model 70:
== Surface-mount electrolyte leakage ==


# Boot the Reference Diskette.
The 8580 planar's surface-mount aluminium electrolytics are the leading cause of "the Model 80 will not POST" reports. Leaking electrolyte can produce almost any POST code and can short a rail outright.
# Select '''Set Configuration โ†’ Run Auto Configuration'''.
# Insert option diskettes as prompted.
# Save and reboot.


If 165 persists, reseat every card, clean edge fingers, and bring the card complement up one at a time to isolate the faulty card.
'''Inspect the planar for electrolyte residue before any other diagnosis, and before powering the machine on.''' [[IBM PS/2 Model 80 Capacitor Replacement Guide]] has the recap procedure.


=== 8-bit Slot Limitations ===
Two things removed from this page, both unsourced: a count of the SMD electrolytics on the planar and a comparison of that count with the Model 70's, and a mapping of specific POST codes to specific areas of the board. A POST code does not tell you where a leak is โ€” inspect the whole board.


The Model 80 has '''one 8-bit MCA slot''' (in addition to seven 32-bit slots). The 8-bit slot accepts 8-bit MCA cards (rare) but is also wired to accept 16-bit MCA cards in 8-bit mode. Do '''not''' plug a 32-bit MCA card into the 8-bit slot โ€” it will not POST and may damage the card.
== Micro Channel faults ==


== Type 1 vs Type 2 Planar Differences ==
=== 165 card ID mismatch ===


Diagnosing a fault requires knowing which planar is in front of you:
The card IDs stored in the configuration do not match the cards fitted, because a card was added or removed without running Auto Configuration, because edge-contact oxidation made a card disappear during POST, or because a flat battery lost the configuration record โ€” which is why 165 usually arrives clustered with 161.


{| class="wikitable styled-table" style="width:100%; text-align:center;"
# Boot the Reference Diskette.
|+'''Type 1 vs Type 2 planar โ€” identification'''
# '''Set Configuration โ†’ Run Auto Configuration'''.
! Marker !! Type 1 (16 / 20 MHz) !! Type 2 (25 MHz)
# Insert each card's option diskette when prompted so its ADF is copied to the working diskette.
|-
# Save and reboot.
| Submodels || 041, 071, 111, 311 || A21, A31
|-
| CPU speed || 16 or 20 MHz || 25 MHz
|-
| SIMM speed required || 80 ns or faster || 70 ns or faster
|-
| Reference Diskette || Type 1 || Type 2
|}
ย 
A Type 1 Reference Diskette booted on a Type 2 planar will report incorrect memory configuration and may refuse to complete SETUP. A Type 2 Reference Diskette on a Type 1 planar will also misconfigure.
ย 
== ECA Recalls and Service Bulletins ==
ย 
The Model 80 was affected by several Engineering Change Authorisation (ECA) bulletins:
ย 
* '''ECA 087''' โ€” Reference Diskette compatibility update for newer MCA cards.
* '''ECA 092''' โ€” Planar fix for one revision of the Model 80.
* '''ECA 117''' โ€” ESDI controller firmware update (particularly relevant on the 314 MB ESDI drive configurations).
ย 
These ECAs apply only to specific FRU part numbers; check the planar / card FRU against the bulletin before applying.
ย 
== PSU Faults ==
ย 
Symptoms and diagnosis:
ย 
* '''Dead โ€” no fans, no power''': Bulk capacitor or mains rectifier; PSU recap required. See [[IBM PS/2 Model 80 Capacitor Replacement Guide]].
* '''Fans spin briefly, then click-retry''': Power Good not asserted in 150 ms. Could be PSU fold-back or shorted planar tantalum / leaked SMD electrolyte.
* '''Boots cold, fails when warm''': Secondary electrolytics aged.
* '''Audible whine, smell of fish''': RIFA X2 cap is venting.
* '''Rails low/high''': PSU feedback path issue.
ย 
The 225 W PSU on the Model 80 has comfortable headroom for the 80386DX, the planar VGA and up to 7 MCA cards. A rail sag on a properly-functioning Model 80 PSU usually indicates an aged secondary electrolytic.
ย 
== Drive Stiction ==
ย 
ESDI drives in the Model 80, particularly the 314 MB drives, are notoriously prone to spindle stiction after long storage. Same field fix as on the Model 60 / 70:
ย 
* Power off.
* Open the chassis. Locate the drive.
* Gently rotate the drive case 45โ€“90ยฐ around its spindle axis in both directions to free the heads.
* Re-install. Power on.
ย 
The 314 MB drive is full-height and heavier than the smaller ESDI drives; handle with care during the stiction-freeing procedure.
ย 
After the drive boots, '''immediately''' image its contents to a modern disk image file. The drive cannot be trusted to spin up reliably again.


== Server Configuration-Specific Faults ==
If it persists, reseat every card and clean the edge fingers, then bring the card complement up one at a time.


The 8580-311 and 8580-A31 ship with the 314 MB ESDI drive in the top bay and a smaller (1.44 MB) floppy in Bay 2. Things to know:
ADF files are named '''@xxxx.ADF''' where xxxx is the four-hex-digit, 16-bit card ID.


* The 314 MB ESDI drive draws more startup current than the 70 / 115 MB drives. PSU rail sag at boot is more pronounced; if the 225 W PSU has aged caps, the 314 MB submodels will exhibit boot failures before the 111 / 071 / 041 submodels do.
=== Slot fitting ===
* The dual-drive 8580-311 / A31 configurations require the ESDI controller card to be configured for two drives (via the Reference Diskette SETUP).
* Novell NetWare on the 8580-311 / A31 is sensitive to drive timing โ€” aged ESDI controller caps produce NetWare-side errors that look like volume corruption.


== Keyboard / Mouse Faults ==
Micro Channel connectors are physically keyed, so a 32-bit card will not seat in a 16-bit slot. Check which slots on your planar are 32-bit from the table above rather than by counting from one end of the row.


Same as Model 70:
== Power supply ==


* '''301''': Keyboard or mouse in the wrong port.
* '''Dead, no fans''' โ€” check the mains, then the bulk capacitor and mains rectifier. See [[IBM PS/2 Model 80 Capacitor Replacement Guide]].
* '''305''': +5 V fuse on planar blown.
* '''Fans start and the supply retries''' โ€” Power Good not asserted within 150 ms, from the supply's own fold-back or a short on the planar or an adapter. Go to [[#Dead system|Dead system]].<ref name="ard_err" />
* '''365''': PS/2 mouse fault.
* '''Boots cold, fails warm''' โ€” secondary electrolytics.
* '''Acrid smell with a whine''' โ€” a mains-filter capacitor venting. Stop using the machine.
* '''Rails high or low''' โ€” the feedback path.


== Memory Faults ==
'''No wattage figure is given here.''' The figure previously on this page, and the comparison of headroom between submodels, had no source. IBM's own 17xx entry is worth noting in this context: the FRU list for the general ST-506 fixed disk group ends with '''power supply''', so a supply that sags is a documented cause of disk errors rather than only of boot failures.<ref name="ard_err" />


* '''201''' with planar-range address: Planar SIMM failure. Identify the failing bank from the address.
== Drive stiction ==
* '''201''' with MCA Memory Adapter address: Memory adapter card failure.
* '''215 / 216''': MCA Memory Adapter configuration error.
* '''225''' on a Type 2 planar: SIMMs too slow.


== When to Suspect the Planar ==
ESDI drives seize after long storage: the drive does not spin up and POST reports a 104xx code. The field fix works '''once''', for data recovery only โ€” power off, rotate the drive gently 45โ€“90ยฐ around its spindle axis both ways, refit, power on. '''Image the drive immediately.''' The larger full-height drives are heavy; support them while doing it.


* 1xx errors that persist after CMOS battery replacement and Auto Configuration.
== Keyboard, mouse and memory ==
* No video and no beeps after PSU verified known-good.
* Repeated 165 errors after every card removed and reseated.
* '''Any''' visible SMD electrolyte leak.


The Model 80's larger SMD cap count makes planar leakage statistically more likely than on the Model 70.
* '''301''' โ€” IBM's FRU order is the keyboard cable, then the system board.<ref name="ard_err" />
* '''305''' โ€” keyboard voltage error. IBM's order is the keyboard port fuse, the cable, then the system board, '''and if there is no fuse in the system, replace the system board.''' The Type 3 planar does have fuses; see above.<ref name="ard_err" /><ref name="ard_t3" />
* '''2xx''' โ€” IBM's FRU order is the system board memory, then the system board. Its "wrong type or wrong speed" group is 214/215/216/221/226/235/240/241. Both the Type 1 and Type 2 planars are specified with '''80 ns''' memory kits, so the "70 ns or faster" requirement previously stated here for the 25 MHz board has no source.<ref name="ard_err" /><ref name="ard_t1" /><ref name="ard_t2" />
* On a 25 MHz machine with 8 MB fitted and intermittent parity errors, check ECA 051 above before replacing memory.<ref name="ard_t3" />


== Diagnostic Workflow ==
The minor-code-to-socket decode for 2xx errors '''applies only to the Models 56/57, 76/77, 85 and 90/95''', not to the Model 80.<ref name="ard_err" />


# Visually inspect planar for SMD electrolyte leak first. If found, do not power on.
== When to suspect the planar ==
# Confirm Type 1 vs Type 2 planar and use correct Reference Diskette.
# Power on. Listen for beep.
# Note POST screen โ€” any leading numeric code.
# If 161/162/163 cluster โ€” replace DS12887 module.
# If 165 โ€” run Auto Configuration; cards reseated.
# If 104xx โ€” verify drive type in CMOS; reseat drive cable; consider stiction; consider controller cap failure.
# If 24xx โ€” planar VGA fault, often SMD electrolyte leak.
# If 104 โ€” 80386 protected mode fault, often SMD electrolyte leak near CPU.
# If no beep and no video โ€” PSU first (rail check), then planar SMD inspection.
# Run Advanced Diagnostics (Ctrl-A) once basic POST passes.


== โš ๏ธ Power-supply RIFA capacitor and tantalum shorts ==
* Any visible electrolyte residue.
* 1xx errors that persist after a new battery and a full Auto Configuration.
* No video and no beeps with a known-good supply.
* Repeated 165 with every card removed, reseated and reconfigured.
* On a 25 MHz machine, intermittent parity errors with FCT841 modules at U24 and U34 โ€” that is ECA 051, and the answer is the board.


Two age-related failures are near-universal on this era of IBM hardware:
== Diagnostic workflow ==


* '''RIFA mains-filter capacitors''' in the power supply are metallised-paper parts that crack and fail '''short''' with age, producing acrid smoke shortly after power-on. Replace them pre-emptively with modern X2-class parts.<ref name="ibm_rifa">[https://www.minuszerodegrees.net/failure.htm minuszerodegrees.net โ€” IBM failure symptoms]; [https://www.classic-computers.org.nz/blog/2010-11-04-restoring-an-IBM-xt.htm Repairing and Restoring an IBM XT]; and [https://retrorepairsandrefurbs.com/2025/05/15/1983-ibm-pc-5160-xt-power-supply-rebuild-modifications/ Adam's Vintage Computer Restorations]. Source for the RIFA mains-filter capacitor failing short (smoke) and the tantalum capacitors failing short and preventing the PSU from firing.</ref>
# '''Inspect the planar for electrolyte residue before powering on.'''
* '''Tantalum capacitors''' on the planar (system board) and on ISA cards fail short with age. A shorted tantalum will '''prevent the power supply from starting''' (dead machine, PSU protection latched) &mdash; look for a cracked or discoloured tantalum and lift suspect ones to find the short.<ref name="ibm_rifa">[https://www.minuszerodegrees.net/failure.htm minuszerodegrees.net โ€” IBM failure symptoms]; [https://www.classic-computers.org.nz/blog/2010-11-04-restoring-an-IBM-xt.htm Repairing and Restoring an IBM XT]; and [https://retrorepairsandrefurbs.com/2025/05/15/1983-ibm-pc-5160-xt-power-supply-rebuild-modifications/ Adam's Vintage Computer Restorations]. Source for the RIFA mains-filter capacitor failing short (smoke) and the tantalum capacitors failing short and preventing the PSU from firing.</ref>
# Identify the planar type and use the matching Reference Diskette.
# Power on and listen for the beep.
# Note any numeric code.
# '''161 / 163''' โ€” the battery. See [[#Battery|Battery]].
# '''165''' โ€” Auto Configuration, then clean the MCA contacts.
# '''104xx''' โ€” drive type in configuration, reseat the cable, then stiction, then the controller.
# '''24xx''' โ€” IBM's discriminator is whether the screen colours change: if they do, suspect the display; if not, the system board.<ref name="ard_err" />
# '''191''' on a 25 MHz machine โ€” the 82385 cache subsystem.<ref name="ard_err" /><ref name="ard_t3" />
# No beep and no video โ€” go to [[#Dead system|Dead system]].
# Once POST passes, run Advanced Diagnostics with Ctrl-A.


IBM PC/XT switching supplies also need a '''minimum load''' to start, so a bare supply on the bench may not run without a dummy load.<ref name="ibm_rifa">[https://www.minuszerodegrees.net/failure.htm minuszerodegrees.net โ€” IBM failure symptoms]; [https://www.classic-computers.org.nz/blog/2010-11-04-restoring-an-IBM-xt.htm Repairing and Restoring an IBM XT]; and [https://retrorepairsandrefurbs.com/2025/05/15/1983-ibm-pc-5160-xt-power-supply-rebuild-modifications/ Adam's Vintage Computer Restorations]. Source for the RIFA mains-filter capacitor failing short (smoke) and the tantalum capacitors failing short and preventing the PSU from firing.</ref>
== What has been removed from this page ==
ย 
== โš ๏ธ CMOS / RTC battery ==
ย 
This machine keeps its configuration in battery-backed CMOS, and the battery is a common failure. On AT-class boards the clock/CMOS is often a '''Dallas DS1287/DS12887''' module with the cell sealed inside; it lasts about ten years and then dies, giving '''161 / 163''' CMOS and clock errors at POST (and sometimes spurious floppy-drive errors). PS/2 planars use a rechargeable barrel or pack battery that '''leaks''' and corrodes the board. Replace a dead Dallas module (or rework it with an external coin cell), and on a leaking planar battery remove it and clean the corrosion before it eats the traces.<ref name="ibm_batt">[https://www.classic-computers.org.nz/blog/2009-10-10-renovating-a-dallas-battery-chip.htm Fixing a Flat Dallas DS1287 RTC], Classic Computers; and [https://www.ardent-tool.com/misc/Dallas_Rework.html Reworking Dallas RTC Modules], Ardent Tool. Source for the Dallas DS1287/DS12887 internal-battery death (161/163 CMOS errors) and the leaking planar battery.</ref>
ย 
== References ==


<references />
* '''The hand-transcribed POST summaries''', replaced by the single reference on the [[IBM PS/2 Model 50 Troubleshooting Guide#Numeric POST error codes|Model 50 guide]]. They gave 104 as a protected-mode failure (IBM's is 115), 114 as an 80386 paging test (IBM's 14xx is a ROM error), 365 as a mouse fault (not in IBM's list), 2401/2402/2410 as the 24xx family (IBM's four entries are different), and headed the coprocessor family 8xx when IBM's is 7xx.
* '''The instruction to replace a DS12887 module''' โ€” see [[#Battery|Battery]].
* '''The two-type planar table''', which put the -111 and -311 models on the Type 1 planar and described the 25 MHz board as Type 2, and gave a 70 ns memory requirement that is not in IBM's data.
* '''"Seven 32-bit MCA slots" and "one 8-bit MCA slot"''', neither of which exists, and the claim that fitting a 32-bit card to an 8-bit slot may damage it.
* '''The PSU wattage''' and the submodel headroom comparison.
* '''The NetWare volume-corruption claim''' and the startup-current comparison between drive sizes. Both were speculation.


== Related Pages ==
== Related pages ==


* [[IBM PS/2 Model 80]]
* [[IBM PS/2 Model 80]]
* [[IBM PS/2 Model 80 Maintenance Guide]]
* [[IBM PS/2 Model 80 Maintenance Guide]]
* [[IBM PS/2 Model 80 Capacitor Replacement Guide]]
* [[IBM PS/2 Model 80 Capacitor Replacement Guide]]
* [[IBM PS/2 Model 70 Troubleshooting Guide]] โ€” desktop sibling, same planar family
* [[IBM PS/2 Model 50 Troubleshooting Guide]] โ€” the numeric error code reference
* [[IBM PS/2 Model 60 Troubleshooting Guide]] โ€” same tower chassis
* [[IBM PS/2 Model 70 Troubleshooting Guide]]
* [[Capacitor Failure Symptoms]]
* [[Recommended Tools]]


== References ==
== References ==


* [https://www.ardent-tool.com/trouble/ps2error.html PS/2 Error Codes โ€” Ardent Tool of Capitalism]. Source for the 1xxโ€“104xx error code tables.
<references>
* [https://www.ardent-tool.com/qtechinfo/GJAN-43VS3S.html IBM PS/2 Model 80 โ€” Ardent Tool Quick Reference]. FRU data, planar Type 1 vs Type 2, ECA history.
<ref name="ard_err">''PS/2 Error Codes'' and the linked error-code pages, [https://www.ardent-tool.com/trouble/ps2error.html Ardent Tool of Capitalism]. A community reference reproducing IBM's own symptom-and-FRU lists; content credited on the site to Louis F. Ohland, Peter H. Wendt, David L. Beem, William R. Walsh, Tatsuo Sunagawa, Tomรกลก Slavotรญnek, Jim Shorney, Tim N. Clarke, Kevin Bowling and others, maintained by Tomรกลก Slavotรญnek. Source for the five beep-code patterns, the numeric error-code list and its FRU/action column (including 0001 91XX "82385 cache test failed", the keyboard voltage error's "if no fuse in system replace system board" note, the 17xx FRU list ending with the power supply, and the 24xx colours-change discriminator), the diagnostics procedure and the dead-system procedure. This wiki's transcription lives on the [[IBM PS/2 Model 50 Troubleshooting Guide]].</ref>
* IBM, ''IBM Personal System/2 Hardware Maintenance Manual'' (S52G-9971-02, October 1994). POST error code reference.
<ref name="ard_batt">''PS/2 RTC/CMOS Batteries'', [https://www.ardent-tool.com/misc/batteries.html Ardent Tool of Capitalism], based on content by Bob Eager and Peter H. Wendt. Source for the per-machine battery cross-reference โ€” the Models 50, 50 Z, 60, 70, 80 and P70 all take the 6 V lithium pack, FRU P/N 72X8498, with CR-P2 / DL223A / EL223AP / K223LA / PC223 equivalents, held in a dedicated Battery/Speaker Assembly, while the Dallas RTC module (FRU 8509237, Dallas DS1287) belongs to the Models 25-286, 30-286, 35, 40, 55 SX, 55 LS and 65 SX โ€” and for the statement that a weak battery loses all of the setup information while a partial loss is probably not the battery.</ref>
<ref name="ard_t1">''8580 Type 1 Planar'', [https://www.ardent-tool.com/8580/Planar_T1.html Ardent Tool of Capitalism]. Source for the 16 MHz Type 1 planar, models -041 and -071, FRU P/N 33F8416 and 72X7492, the 1 MB 80 ns memory kit 72X6670 / 6450375, the slot map (J1/J2/J4 32-bit with MME, J3/J5/J7/J8 16-bit, J6 16-bit with AVE), the J18 Battery/Speaker connector, and the device complement including the CR8 LM340T5 7805 regulator, U13 386DX-16, U72 72X8287ESD VGA, U187 MC146818A RTC and U188 8042 keyboard controller.</ref>
<ref name="ard_t2">''8580 Type 2 Planar'', [https://www.ardent-tool.com/8580/Planar_T2.html Ardent Tool of Capitalism]. Source for the 20 MHz Type 2 planar, models -081, -111, -121, -161, -311 and -321, FRU P/N 33F8506 / 92F0756 / 92F0767, the 2 MB 80 ns memory kit 92F0669 / 6450379, and the same slot map as the Type 1.</ref>
<ref name="ard_t3">''8580 Type 3 Planar'', [https://www.ardent-tool.com/8580/Planar_T3.html Ardent Tool of Capitalism]. Source for the 25 MHz Type 3 planar, models -A16, -A21 and -A31, its Japanese design shared with the PS/55 Model 5570-Vxx, FRU P/N 34F0022 / 33F5571 / 85F0046 / 91F7513, the 386DX-25 at U96 with a 387DX-25 socket at U165 and 64 KB of L2 cache behind an 82385-25 cache controller at U54 with M5M5178P-33 SRAMs at U99โ€“U106, DMA support on the A21 and A31 parallel ports, up to 8 MB on the planar using 4 MB System Memory Boards, the four 32-bit MME slots (J10โ€“J13) and four 16-bit slots (J6/J7 with AVE, J14/J15) with one normally occupied by the 16-bit SCSI controller, the planar fuses F2 1 A, F3 1.5 A and F4 2.0 A, the MC146818AFN RTC at U132, the socketed 8042 at U59, the 90X8941 VGA at U16, and the full text of '''ECA 051''' โ€” 8580 memory parity errors on Models A16, A21 and A31 with 8 MB fitted, the FCT841 physical check at positions U24 and U34 near J8 and J9, IBM's instruction to read the module rather than the board silkscreen, and the note that other identifications (F841, 841D, 72F841, 72F841D) are good.</ref>
<ref name="ard_8570">''8570 Type 3 &amp; 4 Planar'', [https://www.ardent-tool.com/8570/Planar_T3-4.html Ardent Tool of Capitalism]. Source for ECA 092 being the 8570 Data Modification Patch, applying to 25 MHz 8570 systems with an 80 MB or 160 MB hardfile running DOS or OS/2.</ref>
</references>


{{Navbox-IBMComputers|state=collapsed}}
{{Navbox-IBMComputers|state=collapsed}}

Latest revision as of 10:36, 27 September 2026

IBM PS/2 Model 80. Source: Wikimedia Commons.

Fault diagnosis for the IBM PS/2 Model 80 (machine type 8580). The Model 80 shares its POST architecture with the IBM PS/2 Model 50.

Which planar do you have

[edit | edit source]

There are three documented 8580 planars, not two, and an earlier version of this page assigned the models to the wrong ones:[1][2][3]

8580 planars
Type Speed Models Memory kit Slots
Type 1 16 MHz -041, -071 1 MB 80 ns kit 72X6670 or 6450375 J1, J2, J4 32-bit with MME; J3, J5, J7, J8 16-bit; J6 16-bit with AVE
Type 2 20 MHz -081, -111, -121, -161, -311, -321 2 MB 80 ns kit 92F0669 or 6450379 Same as Type 1
Type 3 25 MHz -A16, -A21, -A31 Up to 8 MB on the planar using 4 MB System Memory Boards J10โ€“J13 32-bit with MME; J6, J7 16-bit with AVE; J14, J15 16-bit

Planar FRUs: Type 1 33F8416 / 72X7492; Type 2 33F8506 / 92F0756 / 92F0767; Type 3 34F0022 / 33F5571 / 85F0046 / 91F7513.[1][2][3]

There is no 8-bit Micro Channel slot on any Model 80, and there are never seven 32-bit slots. Types 1 and 2 have three 32-bit and five 16-bit; Type 3 has four 32-bit and four 16-bit, two of the 16-bit ones carrying the AVE extension and one normally occupied by the 16-bit SCSI controller.[1][3]

What is distinctive about the Type 3

[edit | edit source]

The 25 MHz planar is a Japanese design, shared with the PS/55 Model 5570-Vxx, and is the only one with:[3]

  • A 386DX-25 at U96, a socket for a 387DX-25 at U165, and 64 KB of L2 cache behind an 82385-25 cache controller at U54 with eight M5M5178P-33 8K ร— 8 SRAMs at U99โ€“U106. This is the planar on which IBM's POST code 0001 91XX, "82385 cache test failed", can actually apply.[4]
  • DMA support on the parallel port โ€” the A21 and A31 parallel ports support DMA parallel communications.
  • Real fuses on the planar: F2 1 A, F3 1.5 A, F4 2.0 A, with solder pads at F1. On the Type 1 and Type 2 planars IBM's own note for the keyboard voltage error still applies โ€” if there is no fuse, replace the system board.[4]
  • A socketed 8042 keyboard controller at U59, an MC146818AFN RTC at U132, and a 90X8941 VGA at U16.

Reference Diskette and diagnostics

[edit | edit source]

Use the Reference Diskette matching the planar type. A mismatched diskette will misreport the memory configuration and may refuse to complete configuration.

  • F1 boots the Reference Diskette; Ctrl-A from its menu enters Advanced Diagnostics.
  • A four-digit numeric code at the top left of the screen is a fault code from the test block that was running, and is the FRU lookup key.

POST beep codes

[edit | edit source]

IBM's PS/2 audio codes are five patterns, and three of the five mean the same thing:[4]

PS/2 beep codes (* = short, _ = long)
Pattern Meaning
* Normal operation
* * A problem exists โ€” check the configuration. Run diagnostics if no error code is displayed
_ *, _ * *, _ * * * Display or display adapter problem
Continuous short beeps Check the keyboard
Continuous solid beep System board

An earlier version of this page gave a different table in which the long-plus-short patterns were split between "planar fault" and "display adapter fault", and added rows for "continuous โ€” power supply or planar" and "none, no display". IBM's list does not make those distinctions; the no-beep case is dealt with under Dead system.

Numeric POST error codes

[edit | edit source]

This wiki keeps one copy of IBM's PS/2 numeric error code list, on the Model 50 guide. It is transcribed from the Ardent Tool of Capitalism's reproduction of IBM's own symptom-and-FRU lists and keeps IBM's FRU / action column.[4]

Do not work from a per-machine summary of these codes. The tables previously on this page were hand-transcribed, mixed PC/AT-era meanings into PS/2 codes, and โ€” in the ESDI family in particular โ€” contained a consecutive run of codes of which ten do not exist in IBM's list and several more had the wrong meaning. They have been replaced by the link above rather than corrected in place, so that there is one copy to keep right.

Three specific corrections worth carrying in your head:

  • The math coprocessor family is 7xx, not 8xx. This page previously headed it "8xx โ€” Math Coprocessor" with a single 801 code. IBM's coprocessor family is 0007 XXXX; 0008 does not exist.
  • 104 is not a protected-mode failure. IBM's protected-mode code is 115. 104 appears only grouped with 102, 103 and 107 against a system board or processor board FRU.
  • 162 is not a checksum error. It tells you to check that devices are powered on and whether the configuration has changed. The bad-checksum code is 173, and IBM's note there is "possibly a weak battery".

Dead system

[edit | edit source]

IBM's stated most likely cause of no POST, a blank screen and no beep is a device shorting out the power supply. The system board must present Power Good to the supply within 150 ms or the supply shuts itself down internally โ€” which is the mechanism behind the "click and then nothing" symptom. IBM's usual suspects are a SIMM in backwards, an adapter not fully seated, the power cord not plugged in, an unsupported or defective adapter, or too many drives spinning up at once on a heavily loaded machine (leave the motor-start jumper open so the drives start after the controller interrogates them).[4]

  1. Check the line cord at both ends. IBM singles this out: intermittent faults, including unexpected power-on resets during operation, are caused by line cords that are not fully seated or too loose to make a tight connection. Remove the cord from the outlet before working on its contacts.
  2. Remove all adapters, options and extra memory.
  3. Remove all external connections.
  4. Power on and listen for a beep.
  5. No beep: check speaker continuity; if good, replace the system board memory and retry; if still silent, measure the rails โ€” correct rails mean the system board, wrong rails mean the supply.
  6. A beep: refit adapters one at a time, retrying after each. When the beep stops, the last item fitted is the suspect.

On machines fitted with LogicLock, the access-detect switches can be left in the wrong position during reassembly, producing a machine that appears dead.[4]

Diagnostics procedure

[edit | edit source]
  • Cold-boot the Reference Diskette. Soft-booting it can report false errors and falsely indicate that a power-on password is set.[4]
  • If several errors appear, diagnose the first one first โ€” the first can cause the others, especially with configuration errors.
  • Run Advanced Diagnostics (Ctrl-A from System Programs) before replacing anything. If the diagnostics do not fail, replacing parts probably will not help.
  • On a multi-FRU adapter, remove the optional FRUs one at a time before replacing the adapter.
  • External surge suppressors are a documented cause of hard-to-diagnose faults.

Battery

[edit | edit source]

This machine takes a 6 V lithium battery pack, IBM FRU P/N 72X8498 โ€” a two-cell pack, usually yellow with black ends, held in a dedicated Battery/Speaker Assembly. It is a camera battery: equivalents are the CR-P2 (Varta, Panasonic, GP, Sony, Sanyo), DL223A (Duracell), EL223AP (Ever Ready), K223LA (Kodak) and PC223 (Ray-O-Vac).[5]

There is no Dallas RTC module in this machine. Earlier versions of this page told the reader to replace a DS12887 module on a "late planar". The Dallas self-contained module (IBM FRU 8509237, Dallas DS1287) is fitted to the Models 25-286, 30-286, 35, 40, 55 SX, 55 LS and 65 SX โ€” not to the 50, 50 Z, 60, 70, 80, P70 or P75, all of which use the 6 V pack.[5]

Defective or weak batteries cause the loss of all setup information. If only part of the setup is lost, the battery is probably NOT the cause. Inaccurate time is usually software, though the older models with 6 V packs are prone to the clock running slow.[5]

The real Model 80 ECA: 051, memory parity errors

[edit | edit source]

This is the one Engineering Change Announcement that can be checked, and it is a good one to know.[3]

M/T 8580 Models A21, A31 and A16 โ€” intermittent OS/2 trap errors. 25 MHz 8580 systems with 8 MB of system board memory fitted, experiencing intermittent planar memory parity errors such as OS/2 Trap 0002 or NMI errors under other complex operating systems.

Physical check: examine the modules in positions U24 and U34, near the system board memory sockets J8 and J9. If the part number on the module is FCT841, the system board qualifies for replacement under this ECA.

IBM's own caution is important: ignore the information printed on the system board near the module โ€” use only the information printed on the module itself. Modules with any other identification (F841, 841D, 72F841, 72F841D and so on) are good and the board does not qualify.

The underlying fault is a timing condition on the planar. Only the three 25 MHz models listed are affected; other 8580 models are not eligible.

What has been removed: this page previously named ECA 087, ECA 092 and ECA 117 with Model 80 meanings. ECA 092 is documented as the 8570 Data Modification Patch โ€” a hardfile data-transfer patch for 25 MHz Model 70 machines โ€” not a Model 80 planar fix.[6] ECA 087 and ECA 117 could not be corroborated anywhere.

Surface-mount electrolyte leakage

[edit | edit source]

The 8580 planar's surface-mount aluminium electrolytics are the leading cause of "the Model 80 will not POST" reports. Leaking electrolyte can produce almost any POST code and can short a rail outright.

Inspect the planar for electrolyte residue before any other diagnosis, and before powering the machine on. IBM PS/2 Model 80 Capacitor Replacement Guide has the recap procedure.

Two things removed from this page, both unsourced: a count of the SMD electrolytics on the planar and a comparison of that count with the Model 70's, and a mapping of specific POST codes to specific areas of the board. A POST code does not tell you where a leak is โ€” inspect the whole board.

Micro Channel faults

[edit | edit source]

165 card ID mismatch

[edit | edit source]

The card IDs stored in the configuration do not match the cards fitted, because a card was added or removed without running Auto Configuration, because edge-contact oxidation made a card disappear during POST, or because a flat battery lost the configuration record โ€” which is why 165 usually arrives clustered with 161.

  1. Boot the Reference Diskette.
  2. Set Configuration โ†’ Run Auto Configuration.
  3. Insert each card's option diskette when prompted so its ADF is copied to the working diskette.
  4. Save and reboot.

If it persists, reseat every card and clean the edge fingers, then bring the card complement up one at a time.

ADF files are named @xxxx.ADF where xxxx is the four-hex-digit, 16-bit card ID.

Slot fitting

[edit | edit source]

Micro Channel connectors are physically keyed, so a 32-bit card will not seat in a 16-bit slot. Check which slots on your planar are 32-bit from the table above rather than by counting from one end of the row.

Power supply

[edit | edit source]
  • Dead, no fans โ€” check the mains, then the bulk capacitor and mains rectifier. See IBM PS/2 Model 80 Capacitor Replacement Guide.
  • Fans start and the supply retries โ€” Power Good not asserted within 150 ms, from the supply's own fold-back or a short on the planar or an adapter. Go to Dead system.[4]
  • Boots cold, fails warm โ€” secondary electrolytics.
  • Acrid smell with a whine โ€” a mains-filter capacitor venting. Stop using the machine.
  • Rails high or low โ€” the feedback path.

No wattage figure is given here. The figure previously on this page, and the comparison of headroom between submodels, had no source. IBM's own 17xx entry is worth noting in this context: the FRU list for the general ST-506 fixed disk group ends with power supply, so a supply that sags is a documented cause of disk errors rather than only of boot failures.[4]

Drive stiction

[edit | edit source]

ESDI drives seize after long storage: the drive does not spin up and POST reports a 104xx code. The field fix works once, for data recovery only โ€” power off, rotate the drive gently 45โ€“90ยฐ around its spindle axis both ways, refit, power on. Image the drive immediately. The larger full-height drives are heavy; support them while doing it.

Keyboard, mouse and memory

[edit | edit source]
  • 301 โ€” IBM's FRU order is the keyboard cable, then the system board.[4]
  • 305 โ€” keyboard voltage error. IBM's order is the keyboard port fuse, the cable, then the system board, and if there is no fuse in the system, replace the system board. The Type 3 planar does have fuses; see above.[4][3]
  • 2xx โ€” IBM's FRU order is the system board memory, then the system board. Its "wrong type or wrong speed" group is 214/215/216/221/226/235/240/241. Both the Type 1 and Type 2 planars are specified with 80 ns memory kits, so the "70 ns or faster" requirement previously stated here for the 25 MHz board has no source.[4][1][2]
  • On a 25 MHz machine with 8 MB fitted and intermittent parity errors, check ECA 051 above before replacing memory.[3]

The minor-code-to-socket decode for 2xx errors applies only to the Models 56/57, 76/77, 85 and 90/95, not to the Model 80.[4]

When to suspect the planar

[edit | edit source]
  • Any visible electrolyte residue.
  • 1xx errors that persist after a new battery and a full Auto Configuration.
  • No video and no beeps with a known-good supply.
  • Repeated 165 with every card removed, reseated and reconfigured.
  • On a 25 MHz machine, intermittent parity errors with FCT841 modules at U24 and U34 โ€” that is ECA 051, and the answer is the board.

Diagnostic workflow

[edit | edit source]
  1. Inspect the planar for electrolyte residue before powering on.
  2. Identify the planar type and use the matching Reference Diskette.
  3. Power on and listen for the beep.
  4. Note any numeric code.
  5. 161 / 163 โ€” the battery. See Battery.
  6. 165 โ€” Auto Configuration, then clean the MCA contacts.
  7. 104xx โ€” drive type in configuration, reseat the cable, then stiction, then the controller.
  8. 24xx โ€” IBM's discriminator is whether the screen colours change: if they do, suspect the display; if not, the system board.[4]
  9. 191 on a 25 MHz machine โ€” the 82385 cache subsystem.[4][3]
  10. No beep and no video โ€” go to Dead system.
  11. Once POST passes, run Advanced Diagnostics with Ctrl-A.

What has been removed from this page

[edit | edit source]
  • The hand-transcribed POST summaries, replaced by the single reference on the Model 50 guide. They gave 104 as a protected-mode failure (IBM's is 115), 114 as an 80386 paging test (IBM's 14xx is a ROM error), 365 as a mouse fault (not in IBM's list), 2401/2402/2410 as the 24xx family (IBM's four entries are different), and headed the coprocessor family 8xx when IBM's is 7xx.
  • The instruction to replace a DS12887 module โ€” see Battery.
  • The two-type planar table, which put the -111 and -311 models on the Type 1 planar and described the 25 MHz board as Type 2, and gave a 70 ns memory requirement that is not in IBM's data.
  • "Seven 32-bit MCA slots" and "one 8-bit MCA slot", neither of which exists, and the claim that fitting a 32-bit card to an 8-bit slot may damage it.
  • The PSU wattage and the submodel headroom comparison.
  • The NetWare volume-corruption claim and the startup-current comparison between drive sizes. Both were speculation.
[edit | edit source]

References

[edit | edit source]
  1. โ†‘ 1.0 1.1 1.2 1.3 8580 Type 1 Planar, Ardent Tool of Capitalism. Source for the 16 MHz Type 1 planar, models -041 and -071, FRU P/N 33F8416 and 72X7492, the 1 MB 80 ns memory kit 72X6670 / 6450375, the slot map (J1/J2/J4 32-bit with MME, J3/J5/J7/J8 16-bit, J6 16-bit with AVE), the J18 Battery/Speaker connector, and the device complement including the CR8 LM340T5 7805 regulator, U13 386DX-16, U72 72X8287ESD VGA, U187 MC146818A RTC and U188 8042 keyboard controller.
  2. โ†‘ 2.0 2.1 2.2 8580 Type 2 Planar, Ardent Tool of Capitalism. Source for the 20 MHz Type 2 planar, models -081, -111, -121, -161, -311 and -321, FRU P/N 33F8506 / 92F0756 / 92F0767, the 2 MB 80 ns memory kit 92F0669 / 6450379, and the same slot map as the Type 1.
  3. โ†‘ 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 8580 Type 3 Planar, Ardent Tool of Capitalism. Source for the 25 MHz Type 3 planar, models -A16, -A21 and -A31, its Japanese design shared with the PS/55 Model 5570-Vxx, FRU P/N 34F0022 / 33F5571 / 85F0046 / 91F7513, the 386DX-25 at U96 with a 387DX-25 socket at U165 and 64 KB of L2 cache behind an 82385-25 cache controller at U54 with M5M5178P-33 SRAMs at U99โ€“U106, DMA support on the A21 and A31 parallel ports, up to 8 MB on the planar using 4 MB System Memory Boards, the four 32-bit MME slots (J10โ€“J13) and four 16-bit slots (J6/J7 with AVE, J14/J15) with one normally occupied by the 16-bit SCSI controller, the planar fuses F2 1 A, F3 1.5 A and F4 2.0 A, the MC146818AFN RTC at U132, the socketed 8042 at U59, the 90X8941 VGA at U16, and the full text of ECA 051 โ€” 8580 memory parity errors on Models A16, A21 and A31 with 8 MB fitted, the FCT841 physical check at positions U24 and U34 near J8 and J9, IBM's instruction to read the module rather than the board silkscreen, and the note that other identifications (F841, 841D, 72F841, 72F841D) are good.
  4. โ†‘ 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 4.10 4.11 4.12 4.13 4.14 PS/2 Error Codes and the linked error-code pages, Ardent Tool of Capitalism. A community reference reproducing IBM's own symptom-and-FRU lists; content credited on the site to Louis F. Ohland, Peter H. Wendt, David L. Beem, William R. Walsh, Tatsuo Sunagawa, Tomรกลก Slavotรญnek, Jim Shorney, Tim N. Clarke, Kevin Bowling and others, maintained by Tomรกลก Slavotรญnek. Source for the five beep-code patterns, the numeric error-code list and its FRU/action column (including 0001 91XX "82385 cache test failed", the keyboard voltage error's "if no fuse in system replace system board" note, the 17xx FRU list ending with the power supply, and the 24xx colours-change discriminator), the diagnostics procedure and the dead-system procedure. This wiki's transcription lives on the IBM PS/2 Model 50 Troubleshooting Guide.
  5. โ†‘ 5.0 5.1 5.2 PS/2 RTC/CMOS Batteries, Ardent Tool of Capitalism, based on content by Bob Eager and Peter H. Wendt. Source for the per-machine battery cross-reference โ€” the Models 50, 50 Z, 60, 70, 80 and P70 all take the 6 V lithium pack, FRU P/N 72X8498, with CR-P2 / DL223A / EL223AP / K223LA / PC223 equivalents, held in a dedicated Battery/Speaker Assembly, while the Dallas RTC module (FRU 8509237, Dallas DS1287) belongs to the Models 25-286, 30-286, 35, 40, 55 SX, 55 LS and 65 SX โ€” and for the statement that a weak battery loses all of the setup information while a partial loss is probably not the battery.
  6. โ†‘ 8570 Type 3 & 4 Planar, Ardent Tool of Capitalism. Source for ECA 092 being the 8570 Data Modification Patch, applying to 25 MHz 8570 systems with an 80 MB or 160 MB hardfile running DOS or OS/2.


Retrieved from "https://wiki.retrotechcollection.com/index.php?title=IBM_PS/2_Model_80_Troubleshooting_Guide&oldid=14767"