Skip to content

Sega System 18

The successor to the System 16B (1989): same family, same 68000, but one more graphics layer — a Mega Drive VDP grafted onto the board. That is what gives Moonwalker, Alien Storm and Shadow Dancer their look. For preservation it carries two protections where the 16B has one.

System 18 main board.
System 18 main board. Photo uploaded by Mikel to Sega Retro, a site under CC BY 4.0.

Identification

Field Value
Manufacturer Sega
Years 1989-1992
Main CPU Motorola 68000 @ 10 MHz — or FD1094 (same clock) on encrypted games
Sound CPU Zilog Z80 @ 8 MHz
Sound 2 × YM3438 @ 8 MHz + Ricoh RF5C68 @ 10 MHz (PCM)
Graphics System 16B tilemaps/sprites + Sega 315-5313 VDP (the Mega Drive VDP)
Memory mapper Sega 315-5195 @ 10 MHz
Protection FD1094 (battery-backed key) and/or Intel 8751 MCU, game depending
PCB format JAMMA
Suicide? Yes on FD1094 games

Spotting one at a glance

A 16B board with a Mega Drive VDP added. If the board carries a 315-5313 — the same video chip as in a Mega Drive — it is not a System 16B.

Architecture

The System 18 is an augmented System 16B. The 68000, the Z80, the 315-5195 mapper and the tilemap/sprite logic all come from the 16B; two things change.

  • One more VDP. The 315-5313 adds its own planes and sprites to the 16B's. The hardware therefore displays two graphics systems on top of each other — which is why Moonwalker has more layers on screen than a Golden Axe.
  • A richer sound stage. Two YM3438s instead of one YM2151, plus the RF5C68 for PCM. Partial silence therefore does not point at the same chip as it would on a 16B.

Expected clocks — measurement table ✅

Values taken from MAME's sources (sega/segas18.cpp). On a scope or a frequency counter, a clock that is missing or off points straight at the crystal, the divider or the chip itself.

Component Expected clock Role
M68000 10 MHz Main CPU
FD1094 10 MHz Encrypted main CPU (protected games)
Z80 8 MHz Sound CPU
YM3438 8 MHz FM (sound) — two of them
RF5C68 10 MHz PCM (sound)
315-5195 10 MHz memory mapper

The VDP has no publishable figure here

MAME instantiates the 315-5313 at 15 MHz, but the code comment says in so many words that this frequency is a guess (“Frequency is a complete guess”). It is therefore not given as an expected value: a crystal condemned on the strength of a guess is a board broken for nothing. Compare against a working board instead.

Scan rate

~15.0 kHz / 57.23 Hz. The vertical figure is marked “verified on pcb” in MAME; the horizontal one follows from it (57.23 Hz × 262 lines ≈ 14.99 kHz). A standard 15 kHz chassis suits it.

57 Hz, not 60

The System 18 refreshes at 57.23 Hz. On a fussy upscaler or modern display that is outside the 60 Hz window and the picture may refuse to lock — with nothing whatsoever wrong with the board.

The double protection

This is what sets the System 18 apart from the 16B in preservation: two locks, not one.

  1. FD1094 — the program is encrypted, the key lives in the module's battery-backed SRAM. Dead battery = key lost = silent board. The encrypted code sits in sockets A5 and A6 of the ROM board.
  2. Intel 8751 MCU — an extra protection microcontroller on some games, which MAME instantiates separately (system18_i8751).

→ System 18 de-suicide & conversion: decrypted ROMs in A5/A6, remove the 8751, FD1094 replaced by a bare 68000.

Notable games

Michael Jackson's Moonwalker · Alien Storm · Shadow Dancer · D. D. Crew · Laser Ghost · Clutch Hitter · Desert Breaker · Bloxeed · Wally wo Sagase! · Pontoon.

Clockwork Aquario, released in 2020

Developed in 1992 by Westone and never commercially released, it was finished and published nearly thirty years later. The period ROM is in MAME: a textbook preservation case, a game only the archive saved.

Common faults

Symptom Probable cause Next step
FD1094 game dead at boot key battery gone (suicide) Decrypted ROMs + bare 68000 → de-suicide
Black screen, board powered 315-5195 mapper, 68000, reset PCB diagnosis
Partial graphics layers 315-5313 VDP or video RAM Compare with a good board — the 16B and the VDP overlay each other
Partial FM sound one of the two YM3438s The two chips are independent: a single dead channel tells them apart
No PCM (voices/samples) RF5C68 or its RAM The FM stage can be fine while the PCM is silent
Picture will not lock 57.23 Hz, outside the 60 Hz window A tolerant 15 kHz chassis, or see modernising

Repair (technician)

  • FD1094 battery: on an encrypted game a dead battery is not a hardware fault — the board is fine, the key is gone. Replace it without cutting the key (piggyback) while it still lives; otherwise de-suicide. → Backup batteries
  • Two sound stages to tell apart: FM (2 × YM3438) and PCM (RF5C68) fail separately. Partial silence therefore narrows things immediately.
  • Two graphics systems: corruption affecting only some layers separates the VDP from the 16B logic.

→ Methods: PCB diagnosis · Recap · EPROM

Procedure — a dead board, step by step ✅

What this adds

The general method lives in PCB diagnosis. This is its application to this platform: the same order, but with the values actually expected here, so you do not have to go back and forth.

Tools: multimeter, oscilloscope (or frequency counter), current-limited bench supply, desoldering braid. → Tooling

If the board is still in a cabinet

A CRT holds its charge with the machine unplugged. → CRT safety

1. Inspection, power off. Corrosion, cut tracks, oxidised EPROM legs, leaking SMD capacitors, dubious earlier repairs. On this platform the battery is the first suspect: its leakage eats tracks under the holder, and the damage is often hidden by the battery itself. Lift it to look.

2. The connector, before anything else. A large share of "dead boards" are a contact problem, not a fault. Clean the contacts with isopropyl alcohol and reseat — not with an abrasive eraser on a gold edge, which takes the plating off and brings the problem back worse. It costs two minutes and settles the question.

3. Power, measured on the board. Aim for +5.0 V measured as close as possible to where power reaches the board — not at the PSU output: the drop is in the loom, and that is exactly what makes a board boot on the bench and not in the cabinet. Where you probe depends on this board's connector — JAMMA edge, JVS loom or a proprietary harness — plenty of makers kept one long after JAMMA arrived — see Identification above. → Wiring Check −5 V and +12 V too if the sound stage uses them. Suspected short: bring it up current-limited and find the hot spot.

4. /RESET. It must release after power-up. Held low = the CPU never starts. A 1–2 Hz loop is not the reset circuit, it is the watchdog: the CPU is crashing at boot, so look at the bus, not at the reset.

5. Clocks — the values expected on this platform.

  • M68000 → 10 MHz (Main CPU)
  • FD1094 → 10 MHz (CPU principal chiffré (jeux protégés))
  • Z80 → 8 MHz (CPU son)
  • YM3438 → 8 MHz (FM (son) — deux exemplaires)
  • RF5C68 → 10 MHz (PCM (sound))
  • 315-5195 → 10 MHz (mapper mémoire)

No clock on the M68000 (10 MHz) and there is no point measuring anything else: the fault is the crystal, the divider, or the chip itself. That single measurement splits a dead board in two.

6. Bus. Activity on the main CPU's address lines tells you whether it is running or frozen. Frozen with good clocks and good reset → work RAM, bus buffers, ROMs.

7. It boots but the picture or the sound is wrong — that is another job: see Common faults on this page, then signal tracing.

Procedure — replacing the battery without losing the key ✅

Not every board in this family is concerned

Check on this page whether your game is one of the encrypted ones before touching anything. On the others the battery only holds settings, and losing it costs nothing but a re-setup.

The order is the whole procedure

The decryption key of the FD1094 lives in SRAM held up by that battery. Cutting the old cell before the new one is soldered erases it, instantly and for good: the board goes silent and only a de-suicide brings the game back. So the new cell goes in first, in parallel.

  1. Board powered off, old battery left in place.
  2. Measure the old cell in circuit and note it (a CR2032 is 3 V nominal). This is what tells you whether you are ahead of the failure or already behind it.
  3. Solder the new cell in parallel with the old one — piggyback — watching polarity. A reversed cell discharges the old one through itself and erases the key just as surely.
  4. Re-measure at the holder: the voltage must have come back up. Do not go on before it has.
  5. Only now, remove the old cell.
  6. Power up and check the game boots.

→ Backup batteries

Already dead? This procedure will not bring it back — the key is gone. → de-suicide

Modernisation & mods

What every JAMMA board shares

Video, power, control panel and upscalers depend on the cabinet, not the board: it is all gathered in Modernising a cabinet. To run the board outside a cabinet: supergun.

  • Anti-suicide 🟠: FD1094 games do suicide. The lasting route is de-suicide → conversions, not simply replacing the battery.
  • Scan rate 📺: ~15.0 kHz / 57.23 Hz — a standard 15 kHz chassis suits it, but the 57 Hz may upset an upscaler.
  • ⚠️ Do not host ROMs — method only.

Emulation & FPGA

MAME emulates the System 18 (sega/segas18.cpp), FD1094 and 8751 included: the platform's FD1094 keys were extracted and archived by the community, and that is what makes de-suicide possible today.

Photos / assets

(To take: bare board, the 315-5313 VDP that tells the platform apart from the 16B, the FD1094 module and its battery, the 8751 socket, the ROM board with A5/A6.)

Sources & attribution

Text (synthesised here; sources listed for attribution):

  • MAME — src/mame/sega/segas18.cpp (BSD-3-Clause / GPL): clocks for the 68000 (10 MHz), Z80 (8 MHz), both YM3438s (8 MHz), RF5C68 (10 MHz) and the 315-5195 mapper (10 MHz); refresh 57.23 Hz annotated “verified on pcb”; the system18_i8751 variant; ⚠️ the 315-5313 VDP clock is explicitly annotated “Frequency is a complete guess” and is therefore not carried over here — https://github.com/mamedev/mame/blob/master/src/mame/sega/segas18.cpp
  • VGMRips — Sega System 18 (hardware summary and game list) — https://vgmrips.net/wiki/Sega_System_18

Horizontal frequency: calculated, not measured — 57.23 Hz × 262 lines (set_size(342,262) in MAME) ≈ 14.99 kHz. The method is given so it can be checked.

Image:

Review & corrections

How this space is used

Spotted a wrong value, an outdated procedure, a chip reference that does not match your board? Say so here, with what you observed (model, board revision, serial number, measurement). Every report is cross-checked against a source before anything changes — an unverifiable correction is published as “reported by …, not cross-checked” rather than silently applied.

Reading is open to everyone; posting requires signing in with Discord. Reports from the wiki's declared reviewers are handled first; anyone else's are read too, but go through a human before anything is changed.