Amiga 600 Reset Circuit Repair

The power-up reset on Commodore's A600 schematic, sheet 9: U14 (NE555) with its timing parts C611 and C612, the output transistor Q511 driving _KB_RESET, and U36 and Q622 below.

The Amiga 600 makes its power-up reset with a 555 timer, U14. Its timing capacitor C612 is a 10 µF surface-mount electrolytic, and when it leaks the machine can sit at a black screen with the Caps Lock light on. Repairers have also found the 0.1 µF capacitor C611 and the 555 itself at fault in this circuit, so the steps below work through all three in turn.

Symptoms

• Black screen with the Caps Lock light on. Pure Amiga calls this the usual sign of an A600 that needs a recap, and says the cause is often C612; the machine may boot after a few minutes.
• Black screen with sync and no start-up even from a diagnostic ROM. On LinuxJedi's board, replacing C612 got DiagROM running but the boot still took longer than expected; C611 was also failing.
• The keyboard is dead after a cold start but works after a soft reboot. LinuxJedi traced this to the power-on reset: the keyboard controller on the main board needs a reset at power-up.

If the machine passes the start-up colours and then stops on a white screen, look at the audio supply instead; see Testing below.

Safety

Corrosion from leaked electrolyte can short the supply rails. LinuxJedi checks for shorts between the rails before powering a corroded A600 board because of the risk of fire.

How the circuit works

U36 and U14 on a Rev. 2D board. The silkscreen legend beside them lists Q511, R624, C611 and C612.

Commodore's schematic shows these connections round U14:

U14 pin	Connection
2 (trigger)	R611 (1 MΩ) to VCC and C611 (0.1 µF) to ground. Q622's collector and the unfitted position R624 also join this node.
6 and 7 (threshold, discharge)	Joined, with R612 (47 kΩ) to VCC and C612 (10 µF, positive to this node) to ground.
4 (reset)	Tied to VCC.
5 (control)	C613 (0.01 µF) to ground.
3 (output)	Through R515 (470 Ω) to the base of Q511 (2N3904), with R514 (4.7 kΩ) to VCC. Q511's collector drives the _KB_RESET line, pulled up by R511D (1 kΩ).

U36, a 74F27 three-input NOR gate, takes three lines from the keyboard connector CN13 and drives the base of Q622 (2N3904), whose collector pulls the trigger node low. On a working board LinuxJedi saw a pulse of roughly half a second at power-up.

Parts in the reset circuit (Commodore parts list)

Ref	Part	Commodore part no.
U14	NE555, surface mount	391104-01
C612	10 µF 25 V aluminium electrolytic, surface mount (case C)	391097-07
C611	0.1 µF Z5U, 1206	310027-02
C613	0.01 µF X7R, 1206	390853-01
R611	1 MΩ, 1206	310026-20
R612	47 kΩ, 1206	310026-03
Q511, Q622	2N3904, SOT-23	391121-01
U36	74F27, surface mount	391140-01
R624	Not fitted	–

Tools and parts

• Multimeter.
• An oscilloscope or logic probe to watch the trigger pin and the reset pulse.
• Soldering equipment for surface-mount parts.
• A replacement 10 µF 25 V electrolytic for C612, and a 0.1 µF capacitor for C611.
• A spare NE555 in the surface-mount package.
• A diagnostic ROM such as DiagROM, or a Kickstart ROM and Amiga Test Kit, for testing. LinuxJedi used both.

Procedure

Inspect the board

1. Take the main board out of the case; see the Amiga 600 Maintenance Guide.
2. Look at C612, C611 and U14 under magnification. On LinuxJedi's dead board C612 had leaked and U14 was heavily corroded.
3. Measure between the supply rails for shorts before applying power.

Replace C612

1. For a quick test, Pure Amiga removes C612 and checks whether the machine then boots normally.
2. Fit a new 10 µF capacitor at C612 with its positive end to the node shared by R612 and U14 pins 6 and 7, as drawn on the schematic.
3. Clean off the leaked electrolyte as described in the Amiga 600 Capacitor Replacement Guide.
4. Power up and test. If the machine now boots every time, go on to Testing.

If it still fails

1. Check C611. LinuxJedi found it starting to fail and pulling the trigger signal lower than it should be, and replaced it.
2. Measure U14 pin 2 at power-up. On another board LinuxJedi found the trigger stuck at ground. He replaced Q622 and tested U36, which were both good; the 555 itself had failed with its trigger input behaving like an output, and a known-good 555 cured it.
3. Replace U14 if it is corroded, even if it tests good. LinuxJedi replaced a heavily corroded 555 as a precaution.

Reassemble

1. Clean the board; LinuxJedi gives repaired A600 boards an ultrasonic clean.
2. Refit the board and the keyboard ribbon, and close the case.

Testing

1. Power on from cold several times. The machine should start each time without a long delay, and the keyboard should work from a cold start; after his repair LinuxJedi reported the reset as "very predictable".
2. If you have an oscilloscope, look for the pulse of about half a second at power-up.
3. Run DiagROM or Amiga Test Kit. LinuxJedi's repaired board passed all the DiagROM tests.
4. On one board Pure Amiga found that removing C612 changed the symptoms but did not cure them: the Kickstart colours stopped at a white screen. The cause was a corroded via near C306 and R309 that broke the supply to Paula. See Amiga 600 Audio Output Repair.

Board revisions

Commodore's schematic set is schematic 315987 Rev. C. The photograph above is of a Rev. 2D board, and it carries the same designators for U14, U36, Q511, Q622, C611 and C612. The board revisions are listed on the Amiga 600 page.

Related pages

• Amiga 600
• Amiga 600 Troubleshooting Guide
• Amiga 600 Capacitor Replacement Guide
• Amiga 600 Keyboard Repair
• Amiga 600 Audio Output Repair
• Amiga 600 System Schematics (400422-02)