Gottlieb System 1 Won't Boot: No Display, No Start, and No Diagnostic LED

The short version: before you suspect the CPU board, check the two switches that must be closed for a Gottlieb System 1 to boot — the slam switch and the ball roll tilt — then measure the -12 volt rail. Those three checks account for a great many "dead" System 1 CPU boards. And read nothing into the five-second pause at power-up: it is a delay loop, not a self-test.

Original Gottlieb System 1 A1 CPU board, the Rockwell PPS/4-2 spider-chip MPU fitted to Gottlieb games from 1977 to 1980
The original Gottlieb System 1 A1 CPU board — the board most owners blame first, and often wrongly. Photo: Pinballs Store.

System 1 ran from November 1977 to December 1980 across sixteen games, from Cleopatra to Asteroid Annie. If you own a Sinbad, Joker Poker, Dragon, Genie or Pinball Pool, this is your platform.

Contents

Why there is no flashing LED to read

On a Bally or Williams of the same era the machine tells you something at power-up. A Bally MPU blinks its LED a set number of times and the count narrows the fault — our guide to what each Bally MPU LED flash means decodes them. Williams put numeric codes on the displays, covered in Williams System 7 diagnostic codes.

Gottlieb did neither. As Clay Harrell's System 1 repair guide puts it, "the Gottlieb CPU had no way to tell the user of any errors (there was no CPU board LED flash sequence)." Owners routinely take the five-second pause before the game wakes up for a self-check. The same guide is blunt: "that 5 second boot up delay is just a farce. It's a delay, and nothing more." So the method is a fixed order of checks — and the first two cost nothing.

What the symptom already tells you

Watch the machine for the first ten seconds after switch-on. What the displays do, and when, narrows things considerably.

What you see Where to look first
Nothing at all: no displays, no lamps, no sound Mains, main fuse, transformer, the A2 power supply
Displays light immediately, slightly scrolling, no five-second pause Slam switch or ball roll tilt switch stuck open
Coils buzz or lock on the instant you power up, no boot The -12 volt rail is missing — power off now
Illumination on, displays dark, nothing else happens +5 volts to the CPU, or the CPU board itself
Boots, but the score displays stay blank Display supplies: +60 V, +42 V and their offsets
Boots differently each time, or drifts worse as it warms Connectors and grounds, not chips
Green or grey crust near the battery Battery corrosion — go straight to step 4

If there is no sign of life at all, the fault is upstream of anything System 1 specific, and our guide to a pinball machine that will not turn on is the faster route.

Step 1: the two switches that must be closed

This is the most useful thing to know about System 1, and it catches experienced technicians as often as beginners. Two switches are wired normally closed, and the game reads either one open as a reason not to run. The slam switch — the sprung contact that detects the machine being struck — "MUST be closed or a game won't start", and the ball roll tilt, the pendulum switch that catches a machine being lifted, "must be CLOSED for the game to even 'boot'."

A bent slam switch blade, a pendulum out of position, a broken wire, or forty years of oxide will therefore present as a dead CPU board. The give-away: with the slam switch open, the blue displays "come on immediately with slight scrolling, no five second delay." If your machine skips its pause and sits there glowing, look here first.

Both test with a meter on continuity, game off. Clean the contacts and set the gaps properly rather than bending blades to force them shut. PinWiki notes that replacing capacitor C2 with a jumper wire "permanently holds the slam switch closed on a Gottlieb System 1 CPU board" — fair for proving where the fault lies, but it leaves the game with no slam protection, so undo it afterwards.

Step 2: the power supply, in order

The A2 power supply feeds everything else and is a common failure point in its own right. Work from its input to its output with a meter:

Test point Expected What it feeds
A2-P1 pin 1 11.5 volts AC Low-voltage supplies
A2-P1 pin 4 14 volts AC Low-voltage supplies
A2-P1 pin 6 69 volts AC Display high voltage
A2-P2, +5 V 5.0 to 5.15 volts DC CPU board logic
A2-P2, -12 V -11.9 to -12.1 volts DC CPU board, spider chips

Two quirks matter. The small transformer, Gottlieb part B-17921, has no fuses on its 11.5 volt and 14 volt outputs, so a shorted rectifier diode downstream can destroy the transformer instead of blowing a fuse; adding those fuses is worthwhile. And the main filter capacitor C1 was specified at 2900 µF, undersized from the factory — 6800 to 10,000 µF at 16 volts or higher is standard practice on a restoration. A tired C1 produces exactly the marginal booting that sends people chasing chips.

Step 3: the -12 volt rail

This deserves its own step because the symptom is alarming and the cause is cheap. If the -12 volt supply is missing, the machine does not fail quietly: all the CPU-controlled coils lock on, most CPU-driven lamps light, and the game will not boot. Owners meeting this reasonably conclude the driver board has died. Usually it has not — the -12 volt regulator has, and it is a 7912, marked LIC1 on the schematic.

Power the machine off while you work on this. Coils are rated for milliseconds, not minutes, and a locked-on coil cooks its winding and can scorch the playfield — see a pinball coil stuck on. The -12 volt line also appears on test connector TC3, handy for measuring but another thing that can interrupt it.

Step 4: battery corrosion, connectors and grounding

System 1 CPU boards carry an on-board battery — originally a rechargeable "DataSentry" pack or an AA nickel-cadmium cell — to preserve high scores and audits in the 5101 RAM. After four decades many have leaked, and what leaks is alkaline potassium hydroxide. It creeps along traces into connectors, concentrating in one predictable place: the edge connector fingers at A1J6 and A1J7, immediately downhill of the battery. PinWiki's tell is worth memorising — look for "solder joints which become green or gray and crusty as opposed to a shiny silver."

  • Do not power up a visibly corroded board. Corrosion bridges tracks. Assess it first.
  • Get the battery off the board permanently. The standard fix is an off-board three-AA pack fed through a 1N4004 blocking diode, banded end to the board's positive pad, so a future leak happens in a holder.
  • Expect to replace connectors, not just clean them. The repair literature is unambiguous that the .156-inch single-sided Molex edge connector pins "most certainly will need to be replaced (especially if there was any battery corrosion on the CPU board)." Re-pinning J5, J6 and J7 revives a surprising number of "dead" machines.

Grounding belongs in the same conversation. Every other manufacturer of the period bolted its boards to a metal plate acting as a shared ground plane; Gottlieb mounted them on nylon standoffs and daisy-chained ground from board to board through a wire and a connector. When those links gain resistance — and they do — the boards sit at slightly different notions of ground and the machine behaves in ways no schematic predicts. A machine that boots inconsistently, or drifts as it warms, is describing a connection rather than a failed component. Our System 80 ground mods guide explains why the fix works, the voltage-drop test in burnt pinball connectors finds a bad pin faster than swapping boards, and spotting and stopping pinball battery corrosion covers the cleaning chemistry.

Step 5: the spider chips and the 5101

Only now is it reasonable to suspect the integrated circuits. System 1 uses Rockwell's PPS/4-2 family, nicknamed spider chips for their long, splayed leads. They are no longer manufactured, which is exactly why they should be the last thing you blame.

The processor sits at U1 (an 11660), the display driver interface at U6 (a 10788), with a pair of 10696 I/O devices at U2 and U3. The two that matter for diagnosis are U4 (A1753) and U5 (A1752), which carry the game program as well as driving solenoids and reading the switch matrix, and which the repair literature calls "notorious for easily failing." Being program ROMs they are game-specific, and the PPS/4 archive notes the revision levels of the pair must match each other — so you cannot mix halves from two games.

The 5101 RAM at Z22 is a cheerier story: a common part, easy to replace, and the first suspect if the game boots and plays but loses high scores — especially after a battery leak. For the wider map of which board does what, see pinball machine electronics explained; on System 1 it is A1 for the CPU, A2 for power, A3 for the driver board handling eight solenoids and thirty-six lamps, A4 for the score displays and A5 for the status display.

If it boots but the displays stay dark

That is usually a supply problem, not a dead display. System 1's blue Futaba fluorescent displays each need more than one voltage: the six-digit score displays want +60 V DC, +8 V DC and 5 V AC, the four-digit credit and ball display +42 V DC, +4 V DC and 3 V AC. Missing the small offset voltage alone leaves a good display dark, so measure before you swap; the chip driving them is a UDN6118, and a pinball score display not working covers telling a failed display from a failed supply. The self-test helps here too — steps 11 and 12 exercise the displays, step 13 the lamps — and Gottlieb shipped a test PROM, designated T, for bench diagnosis. Check your manual for the entry sequence, which varies by title.

When a board is genuinely the answer

We would rather you fixed a switch than bought anything, and on System 1 that is more often true than not: three of the steps above cost nothing, and two cost the price of a regulator or a capacitor.

A board becomes the honest answer in narrower cases — battery damage that has eaten through traces and pads rather than merely coated them, a failed U4 or U5 given that the parts are out of production, or a board repaired badly enough that the next fault is unpredictable. Our comparison of repairing versus replacing a pinball circuit board sets out where the line sits.

For System 1 the modern route is a replacement CPU built on the LISY project — an independent developer's free, non-commercial DIY project, which we support but did not create. Our explainer on the Lisy1 CPU board for Gottlieb System 1 covers what it changes, including the web-based diagnostics Gottlieb never fitted. System 1 boards are not currently listed in our shop, so please ask about availability rather than ordering blind. We test boards before they ship and support them afterwards, and our help is free whether or not you buy anything.

FAQ

Why won't my Gottlieb System 1 boot?

Check the slam switch and the ball roll tilt switch first: both are wired normally closed, and if either is stuck open the game will not start. Then measure the +5 volt and -12 volt supplies from the A2 power supply. Those three checks find most non-booting System 1 machines, and none involves the CPU board.

Does a Gottlieb System 1 have a diagnostic LED like a Bally MPU?

No. System 1 has no CPU board LED flash sequence and no boot error reporting of any kind. The five-second pause at power-up is a delay loop in the code, not a self-test, so you cannot read anything into it.

All my coils lock on and the game won't boot. What is it?

That is the classic signature of a missing -12 volt supply. Switch off straight away, because coils are rated for milliseconds and will cook. The -12 volt regulator is a 7912, marked LIC1, and it is cheap to replace.

My System 1 displays light instantly with no five-second delay. Is that good?

No, it is a specific clue. Blue displays coming on immediately with a slight scroll and no pause is the documented symptom of a slam switch or ball roll tilt switch sitting open. Test both for continuity with the game off.

Can I still get the Rockwell spider chips for a System 1 CPU board?

Not new. The U4 (A1753) and U5 (A1752) chips hold the game program as well as driving solenoids and the switch matrix, they are game-specific, and the revision levels of the pair must match. That scarcity is the main reason a replacement CPU board is worth considering once one has failed.

Sources

Stuck on a System 1 that will not wake up? Tell us the game, what the displays do in the first ten seconds, and what you measure on the +5 and -12 volt rails. Our technical support is free, before and after any purchase — and on this platform the answer is very often a switch.

Is your machine down right now?

Tell us your machine and the exact symptom — we'll tell you which board you need, or whether you need one at all. It's free, and Valère answers personally.

Describe my fault →

Not ready to order yet?

Get our troubleshooting guides, restoration tips and new product news — about one email a month, never spam.

Terug naar blog

Onze producten

Alles bekijken