Ardent Tool of Capitalism

8530 Planar (8086)

30start.exe Model 30 Starter Disk v1.02 (zipped image)
adv30dig.exe Model 30 Advanced Diagnostics v1.05 (zipped image)

187-048 IBM PERSONAL SYSTEM/2 (TM) MODEL 30 (8530-002,-021)

SHS64F3987 IBM PS/2 Model 30 HMR
SHS15F2191 IBM PS/2 Model 25/30 HMS

PS/2 Model 30 Planar Registers (by "mkarcher" @ vogons)

8530 Power
ISA Interrupt Sharing (PS/2 Model 25 & 30)
MCGA - Multi-Color Graphics Array

8530 Planar
8530 Greenock Planar
8530 Greenock Planar (alt)
System Firmware
   ROM Images
Riser Card
Known Problems
   8530 Fixed Disk, System Board, Adv. Diags (H007157)
   8530 Cursor Keys and 501 POST Errors (H043664)
   8530 Bus Adapter Battery Voltage (H067282)
   8515/8516 Exhibits Distorted Video When Attached to 8530 (H094152)
Model 30 Production Tales

8530 Planar "P-PLANAR", P/N 61X8823 (schematics)

U48,55 OKI MSM3764A-15, TI TMS4164-15, or compatible 64Kx1 DRAM

Note: Despite the drastically different PCB layouts, the reference designators are largely consistent across all 8086-based Model 30 and Model 25 planars.

This planar is also used in the IBM 7496 Executive Workstation.

8530 Greenock Planar P/N 33F4888

Different PCB layout, but the component references and types are the same as on the P-PLANAR, except for the following items:

U11 SLA6430J GA 11F8028 MCGA Mem. Ctrl. (new)

8530 Greenock Planar (alternative) FRU P/N 33F8474, P/N 33F4871(F)

Same as the other Greenock planar, except for the following items:

CR7-10 RAMDAC clamping diodes were added (in place of J8)
F2 fuse is missing
J8 header artifact is missing

System Firmware (POST & BIOS)

Firmware stored in mask ROM or EPROM.

ROM Images

68X1687 / 68X1627 - 02 Sep 1986, rev. 0, 2x 27256 (U17 / U20)
61X8938 / 61X8937 - 12 Dec 1986, rev. 1, 2x 27256 (U17 / U20)
61X8940 / 61X8939 - 05 Feb 1987, rev. 2, 2x 27256 (U17 / U20)
33F4498 / 33F4499 - 31 Jan 1989, rev. 4, 2x 27256 (U17 / U20)

Riser Card P/N 61X8864

B1 3V lithium RTC battery (Panasonic BR-⅔A)

Three 8-bit ISA slots. 3V lithium RTC battery soldered to the board.

Known Problems

Erroneous/False Display Characters (ECA 023)

8530 Fixed Disk, System Board, Adv. Diags (H007157)

Successful fixed disk operation on the 8530 model 001 and 021 systems is dependent upon using the correct combination of system board, hardfile, and Advanced Diagnostics. Mismatch of the required FRU's or formatting the fixed disk with the incorrect level of Advanced Diagnostics will cause fixed disk errors.

Required FRU P/N and advanced diagnostics:

* Fixed disk P/N 6128285 and P/N 6128279 must not be installed in any 8530 model 021 below serial number 2500000.

* 8530 Model 021's above serial number 2500000 and all 8530 model 001's must be formatted only with 8530 Advanced Diagnostics version 1.05.

8530 Advanced Diagnostics version 1.05 is available by ordering publication number S15F-2153.

8530 Cursor Keys and 501 POST Errors (H043664)

Symptom

On a limited number of PS/2 8530 systems, use of the cursor movement keys may cause unexpected characters to appear on the display. This problem is limited to 8530 models 001 and 021 only. This problem may be intermittent, and may not be detected by Advanced Diagnostics.

The 501 POST error appears when two or more option cards are installed in the 8530. These options include the 3270 and Token Ring adapters (half size cards). This problem is also limited to 8530 models 001 and 021. The 501 will show as a display error.

Problem Isolation Aids

If symptoms appear as described, visually inspect the 8530 system board modules in positions U9 and U15. Module u15 is next to the speaker and U9 next to the math co-processor chip holder.

U9 P/N 72X8203 with IBM printed on chip is suspect
U15 P/N 61X8965 with IBM printed on chip is suspect

Good chips will have the word "Japan" printed on them; affected chips will not.

System units with serial above 55-00HV000 have the new board factory installed.

Fix

If the modules in position U9 and U15 do not have "Japan" printed on them, the system board should be replaced by new system board FRU P/N 64F3453. The removed system board should be returned to be updated to new EC level via normal repair channels (ERC) (if the math coprocessor is installed it should be removed and reinstalled in the new FRU).

8530 Bus Adapter Battery Voltage (H067282)

Early releases of the 8530 HMS (3-ring binder) maps, system board, page 0100-4, show incorrect pins for checking voltage on the bus adapter card.

In figure one, the pins labeled "0" and "39" should actually be labeled pins "1" and "40". Correctly measuring the voltage is then done at pins 38 and 39 (the second and third pins from the right edge).

Replace the bus adapter card if voltage is less than 2.5 volts between the correct pins.

8515/8516 Exhibits Distorted Video When Attached to 8530 (H094152)

Symptom

PS/2 8530 (8086 version) exhibits compressed and distorted video when connected to a 8515 or 8516 display.

Problem Isolation Aids

• 8530 (8086 version) models 001, 002, and 021 do not support attached PS/2 8515 and 8516 displays.
• 8530 (8086 version) models 001, 002, and 021 are supported with the following attached PS/2 displays: 8503, 8504, 8507, 8512, 8513, 8514.
• 8530 (286 version) models E01, E21, E31, and E41 are supported with the following attached PS/2 displays: 8503, 8504, 8512, 8513, 8514, 8515, 8516.

Fix

Advise customer that 8515 and 8516 displays are not supported on 8086 version of the 8530, and they should consult their point of sale if required. Normal problem isolation procedures should be used on supported displays exhibiting the above symptoms.

Model 30 Production Tales (by VK3DRB @ EEVBlog Forum; slightly edited)

I was involved in the initial introduction of the PS/2 Model 30 in late 1986. Code name was Palace. Considering you are in Australia, it most likely originated from our IBM plant in Wangaratta. It was the first of the PS/2 models, followed shortly by the Model 50 and the Model 80. That machine cost a fortune in its day, but it was well designed. I seem to recall there were two BIOS chips... I think they were odd/even addresses. I used to have the entire source code listing to the BIOS and Advanced Diagnostics on my PC at work, which I used to hack to make tests from. It would not surprise me your machine still functions. Prime spec parts were used throughout and the SMPS's were extremely robust. You will find the RAM is ECC parity.

The amount of testing per machine leaves what is done today for dead. You motherboard underwent a 12 hour dynamic burn-in test from about 5 °C to 65 °C, then it was functionally tested on a FACT tester imported from Boca Raton in Florida. (I now own that FACT test machine. The machine originally cost $350,000. I bought it for $100 and stripped it down make a nice ESD bench with built in power supplies.) As an assembled PC, your machine underwent further extensive testing. Less than 1% valid warranty claims were received.

The Palace motherboard was most likely the first surface mount PCB to be made in Australia, because we installed Australia's first SMT plant initially for the PS/2 family. If it were made in Wang, I could tell you the names of assembly and test people who put your machine together, and the names of the engineers involved. When the Palace was made, the Wangaratta plant had only about 90 full time personnel... a great place to work where we were like one big family. At its peak 10 years later there were the equivalent of 600 full time hires and we had the biggest electronics plant in the country by far. We made a few MILLION PCs there for all of Australasia and some of EU and the USA. The reason I moved to Wangaratta for the great opportunity to be at the forefront of technology and get to travel around the world. We had a ball, and in those days money was no object in getting things done. But in 1998, IBM cashed in by selling us off to a shonky startup which ceased business about 2 years later, leaving in its wake a path of social, electronic and economic desolation in Wangaratta.

A little known fact is that we also manufactured a very large volume of Apple Power MAC processors. My job was developing the hardware and software for the testing them. Because of the Power MAC and the IBM RISC systems, in 1995 we were the biggest exporter of non-primary produce in Australia.

...

There were two 27C512's 27[C]256's in the machine, containing BIOS. The chips shared the 16 address lines. One supplied the lowest 8 bits of data, the other the highest 8 bits of data (ie: odd/even byte arrangement). Many of the chips on that board and others from the era were from VLSI Logic. I have no idea what happened to VLSI Logic, but they were a key vendor to IBM for a few years.

IBM never supplied BASIC in PROM on any of their machines, except on the old 5100, 5110 and 5120 minicomputers (which I also worked on). For PC's, BASIC was always supplied on diskettes or cassette. (Ed. This is of course incorrect...)

Another fact about IBM boards is they never had NiCad batteries in them. It was against company policy to design anything with them in due to the cadmium environmental pollution. Most clones however had NiCad cells in them.