Jump to content

IBM PS/ValuePoint Troubleshooting Guide

From RetroTechCollection

This guide documents fault diagnosis for the IBM PS/ValuePoint family — machine types 6381 / 6382 / 6384 / 6387 (1992–1995). PS/ValuePoint inherits the classic IBM PC/PS/2 POST numeric error code series — 161 / 162 / 163 CMOS battery cluster, 2xx memory codes, 6xx floppy codes, 17xx hard disk codes — plus PS/ValuePoint-specific behaviour around the on-planar SVGA, VLB / PCI riser, and ICS-vs-Chrontel video clock generator.

Reference Documents

[edit | edit source]
  • IBM Hardware Maintenance Manual S61G-1423-01 (September 1993) — authoritative HMM covering 6381 / 6382 /S / 6384 / 6384 /D / 6384 P60/D / 6387 /T.[1]
  • IBM HMM S61G-1423-02 (August 1996) — covers 638x and the later 64xx transitional machine types.[2]
  • IBM PS/ValuePoint and Model 64xx Beep-Code Index, kev009 mirror.[3]

Entering Setup

[edit | edit source]

PS/ValuePoint is NOT Reference-Diskette-based. Setup is in ROM. Entry keys:

  • F1 during memory count — enter Setup. The system counts memory at boot; F1 must be pressed during that count.
  • ESC during memory count — skip the memory count and boot directly.[4]

The optional Advanced Diagnostics Diskette is a separate floppy (VPADV100.DSK on the original 6384, vp2diags.exe / vpsidia2.exe / p60dgs.exe on later boards) for hardware self-test only — not for Setup.

Initial Diagnosis

[edit | edit source]

Power on the system and observe in order:

  1. Power LED lights, fan spins.
  2. Memory count appears on screen (you can interrupt with ESC).
  3. Single short beep at end of POST.
  4. Boot device located; OS loads.

If any of these does not happen, stop and diagnose at that stage.

Stage 1 — No Power, No Fans

[edit | edit source]
  • Voltage selector switch on rear set correctly for your mains (115 V or 230 V)? IBM ships in 230 V position; a 230 V switch on 120 V mains is the most common "DOA" symptom.
  • Mains power good and on? Test the lead.
  • Cover-interlock switch depressed? Some PS/ValuePoint chassis have a cover interlock — the system will not power on if the cover is not seated.[5]

Stage 2 — Fan Spins But No POST Beep / No Video

[edit | edit source]
  • Listen for beep codes (see table below).
  • Reseat RAM modules (in matched banks where required).
  • Reseat the L2 cache module if present.
  • Reseat the on-planar CPU if socketed.
  • If the L2 cache module is the wrong type for the board (Type 1 vs Type 2), POST will halt with 12902 / 12904 — see "Cache Codes" below.

Stage 3 — Beeps But Video Absent

[edit | edit source]
  • 2-2-2 or 1 long + 2 short — video adapter fault. Reseat planar video subsystem (no removable cards on PS/ValuePoint — video is on-planar).
  • Try connecting a known-good ISA VGA card and disable the planar video (see Setup).

Stage 4 — Video Present, Halts at BIOS with Error Code

[edit | edit source]

Match the code in the table below.

Stage 5 — POST Passes, OS Won't Load

[edit | edit source]
  • Drive type set correctly in Setup?
  • Boot order correct in Setup?
  • If I 99903 05 appears — no startable device found. Verify boot sequence.[6]

POST Numeric Error Codes

[edit | edit source]

PS/ValuePoint inherits the IBM PC/PS/2 numeric POST series.[7][8]

PS/ValuePoint POST numeric codes
Code Meaning First action FRU
110 Memory parity / general system error Memory Module, System Board
161 Battery dead / disconnected Run Configuration Utility, Clock Battery (CR2032), System Board
162 System options not set (CMOS checksum bad) Run Configuration Utility, Clock Battery, System Board
163 Time and date not set Clock Battery, System Board
164 Memory configuration changed (memory size mis-match with CMOS) Run Configuration Utility, Memory, System Board
199 Installed-devices list mismatch See "Installed Devices List" in HMM supplement
1XX General planar fault System Board
225 Unsupported memory Memory Module
2XX Memory error (address shown is failing block) Memory Module, System Board
303 + 8603 Pointing-device error Mouse, Keyboard, cable, System Board
303 (no 8603) Keyboard error Keyboard, cable, System Board
305 Keyboard / keyboard cable System Board, Keyboard
3XX Keyboard Keyboard, cable, System Board
604 Diskette drive Drive A or B, cable, System Board
662 Wrong diskette drive type Configure in Setup
663 Wrong media type
6XX Diskette subsystem Drive, System Board, cable, PSU
7XX Math co-processor FPU, System Board
9XX System board System Board
1047 16-bit AT Fast SCSI Adapter SCSI adapter
10XX Alternate parallel adapter Riser, adapter
11XX System board (built-in serial) System Board
12XX Alternate serial adapter Adapter, Riser
13XX Game / joystick adapter Adapter
14XX Printer See HMM Printer chapter
15XX SDLC comms adapter Adapter, Riser
1701 / 1711 / 1730 / 1780 / 1782 Hard disk 1 PSU, HDD, cable
1702 Hard disk adapter System Board, controller
17X0 / 17X1 / 17X2 / 17X3 Hard disk 1 / 2 / 3 / 4 PSU, HDD, cable
209X Diskette / 16-bit SCSI Drive, cable, SCSI
21XX SCSI subsystem SCSI device, adapter, cable
2401 On-planar video / display System Board, Display
2402 On-planar video System Board
2409 / 2410 Display / planar video
30XX / 31XX PC Network adapter
86XX Mouse Mouse, System Board
12902 L2 cache adapter L2 cache module
12904 L2 cache adapter / system board L2 cache module, System Board
I 99903 01 Hard disk reset failure
I 99903 05 No startable device Re-check boot sequence in Setup

The 161 / 162 / 163 cluster on a PS/ValuePoint always points to the same fault: the CR2032 lithium coin cell at B1 / BT1 is dead. Replace the cell and re-enter Setup (F1 during memory count). See IBM PS/ValuePoint Maintenance Guide.

POST Beep Codes

[edit | edit source]

PS/ValuePoint uses a piezo beeper, not a magnetic speaker.[9]

PS/ValuePoint beep codes
Beep pattern Meaning First action FRU
1 short POST passed; normal boot
1-3-1, 1-3-2 Memory failure Memory Module / System Board
1-4-4 Keyboard / system board
2-1-1, 2-1-2 Setup not run / configuration error Run Setup / System Board
2-2-2 Video card / on-planar video failure Video chip / System Board
2-3-2 Memory module failure Memory / System Board
2-4-3, 2-4-4 Run Setup / Memory / System Board
1 long, 2 short Display / planar video
3 short Memory failure
Continuous System board fault
Repeating short Stuck key on keyboard / cable / System Board
Other patterns System Board

Memory-Specific Faults

[edit | edit source]
  • 201 with address — failing memory address shown on screen. Identify the failing SIMM bank from the address.
  • 225 — unsupported memory. PS/ValuePoint requires parity SIMMs; a non-parity SIMM will halt POST.
  • 2-3-2 beep + 201 — memory module fault.
  • 110 — memory parity error during POST.

Memory installation rules:

  • 325T — 2 sockets, 72-pin parity, 70–85 ns.
  • Original 6384 (1992)30-pin parity SIMMs in banks of 4. All 4 slots in a bank must be populated; partial banks halt at 225.
  • 6381 /Si, 6382 /S, 6384 /D, 6387 /T — 4 × 72-pin parity sockets, may be installed individually or in pairs depending on submodel.
  • P60/D — 4 × 72-pin parity, must be installed in pairs (dual-channel).

Cache Codes

[edit | edit source]

The 12902 and 12904 codes specifically reference the L2 cache module:

  • 12902 — L2 cache module fault. Reseat or replace the module.
  • 12904 — L2 cache adapter or system board. Check that the module is the correct type for the planar (Type 1 boards use a different cache module than Type 2).

Video Faults

[edit | edit source]
  • 2401 / 2402 / 2409 / 2410 — on-planar video fault.
  • 2-2-2 beep — video chip not responding.
  • Blank screen, no beep — possible video clock generator failure. Look for the ICS / Chrontel chip just left of the Tseng video chip. If the system worked before a driver install and now fails, the OS/2 2.00.1 SVGA driver does not work with Chrontel-equipped boards — install the revised driver.[10]

Storage Faults

[edit | edit source]
  • 1701 / 1702 / 1711 — IDE hard disk drive 1 fault. Verify drive type in Setup; clean / reseat cable; suspect drive failure (PS/ValuePoint era IDE drives are 25+ years old).
  • 1782 — disk controller failure on planar.
  • 604 / 611 / 613 — floppy drive fault. Common cause: aged SMD electrolytic on Alps drive logic board has leaked.[11]
  • 662 — wrong floppy drive type set in Setup. Re-enter Setup; verify A: and B: types match physical drives.
  • 663 — wrong media type (e.g. 1.44 MB drive but disk formatted 720 KB).

2.88 MB Floppy Note

[edit | edit source]

The 486 boards do not support 2.88 MB drives. Only the P60/D supports 2.88 MB. A 2.88 MB drive installed on a 486 board will halt with 662 / 663 errors.

Keyboard Faults

[edit | edit source]
  • 301 / 303 (no 8603) — keyboard fault.
  • 303 + 8603 — pointing device (mouse) fault.
  • 305 — keyboard cable / fuse.
  • 86XX — mouse fault.
  • Repeating short beep — stuck key on keyboard.

The widely-documented PS/ValuePoint Windows 3.1 / WfWg 3.11 keyboard lockup requires the IBMKBFIX.EXE / 6381WG31.ZIP SYSTEM.INI patch — not a hardware fault.[12]

SCSI Faults (Optional Adapter)

[edit | edit source]
  • 1047 — 16-bit AT Fast SCSI adapter (Future Domain).
  • 21XX — SCSI subsystem. Check termination, cable, device addresses.

Common Field Symptoms

[edit | edit source]
  • 161 + 163 at every cold boot — Battery dead. Replace CR2032.
  • Powers on intermittently when cover is replaced — cover-interlock switch fault. Clean / replace.
  • Will not power on at all — Voltage selector switch on rear set to wrong position; or PSU dead.
  • Random reboots, especially under heavy disk activity — PSU caps aged on 200 W supply. See IBM PS/ValuePoint Capacitor Replacement Guide.
  • Floppy drive reads disks but writes corrupt them — Alps drive SMD electrolytic leaked. Recap drive logic board.
  • DOS reports "Bad CMOS" on boot — 162 / 163 cluster. Battery, then Setup.
  • Windows blue-screens with hardware error on cold start — IBMKBFIX issue; or PSU rails sagging.
  • OS/2 SVGA video fails on mode change — ICS / Chrontel chip mismatch with old OS/2 SVGA driver.

ECAs and Defects

[edit | edit source]
  • Defect 57012 — F-Disk error during boot — required fix for OS/2 1.30.1 / 1.30.2 on the 325T.[13]
  • Defect 57146 — Date/Time 162 / 163 error — required fix for OS/2 on the 325T.[14]
  • 6384-F20 hard-drive upgrade — 80 MB → 120 MB; dealer return path through 1-800-426-9733 before 26 February 1993.[15]

No formal ECA numbers (in the IBM "ECA-nnn" sense used on PS/2 line) were located for the PS/ValuePoint family. Updates and bug fixes were handled by Technical Information Letters and BIOS / driver updates rather than ECAs.

Diagnostic Workflow

[edit | edit source]
  1. Confirm the voltage selector switch is correct for your mains.
  2. Power on; observe power LED, fans, memory count, beep.
  3. If no power: voltage selector, mains, cover interlock, PSU.
  4. If power but no POST beep: reseat memory, cache module, CPU.
  5. If beep code: match to table above.
  6. If numeric POST code: match to table above. 161/162/163 → CR2032 replacement.
  7. Press F1 during memory count to enter Setup; verify drive types, boot order, video config.
  8. If POST passes but OS won't load: run Setup, verify boot device, then suspect drive.
  9. Run the Advanced Diagnostics Diskette for sustained hardware self-test.
[edit | edit source]

References

[edit | edit source]