Panasonic FS-A1GT Troubleshooting Guide
.png)
Panasonic's Technical Guide for the Panasonic FS-A1GT includes a set of pass/fail checks with test points and expected readings, and a table of the parts to suspect for each test on its hard-check floppy. Both are summarised here, with the faults owners have reported since. The guide is hosted as the Panasonic FS-A1GT Technical Guide; page numbers in the references are its printed page numbers. For opening the case and routine work see the Panasonic FS-A1GT General Maintenance guide.
Things that are not faults
[edit | edit source]- With the built-in software switch on, the machine starts in the VSHELL menu; with it off, it starts in BASIC. A game cartridge in a slot is started instead of VSHELL.[1]
- A clock that resets, and settings that are lost, at every power-on usually mean the two AA backup cells are flat or missing. Panasonic shipped the machine without them.[1] Holding ESC during start-up also erases the SRAM settings.[2]
- Vertical bars ("jailbars") on an RGB monitor come from the board design. Jumpers J1 and J2 select whether the DIN RGB socket carries composite video (the factory setting) or composite sync, and the colour carrier in the composite signal shows up as bars. Cutting J2 and bridging J1 sends composite sync instead; composite video is then taken from the S-video socket with an S-video to composite cable.[3]
- FM music that sounds dull and quiet against the PSG and cartridge sound is the factory calibration of the audio circuit. A published recalibration replaces seven surface-mount parts on the FS-A1GT board.[3]
- A crash after
CALL MUSIC(0,0)followed by aPLAYtempo command, or after CTRL-STOP, is a bug in the FS-A1GT's MSX-MUSIC ROM.CALL MUSIC(0,0,1)avoids it.[2]
Dead machine
[edit | edit source]Panasonic describes the supply as a DC/DC converter for +5 V and regulators for +12 V and −12 V, all fed from the secondary of the mains transformer T1.[4] Work through it in this order.
Transformer
[edit | edit source]The FS-A1GT runs on 100 V only.[5] If the power lamp stays dark, open the machine, unplug CN1 and, with the mains connected again, measure the AC voltage between the transformer's centre-tap wire and each of the other two. Warning: with the case open, the power switch, the mains lead and the transformer's primary connections are at mains voltage. A forum member gives the turbo R transformer's secondary as nominally 12 V AC, measuring 14 to 15 V. An FS-A1GT that died after its transformer overheated under a heavy cartridge load had no output from the transformer at all, and worked again with the transformer from an FS-A1ST. Another member pointed out that transformers like this usually have a thermal fuse buried in the windings, which opens on overheating.[6] The original transformer is DDT5L6J01Z; the circuit diagram draws a protective element in series with its primary.[7][8]
The +5 V switching regulator
[edit | edit source].jpg)
Measure +5 V at the collector of Q1. Panasonic gives 5 V ±0.25 V as correct.[4] The slots carry +5 V on pins 45 and 47, with pins 41 and 43 as ground.[9]
The +5 V rail comes from IC1, an MC34063 converter that switches the PNP transistor Q1 (2SA1643) into the inductor L2. A crowbar sits on the output: if +5 V rises past the 5.6 V zener D9, the thyristor D1 fires and shorts the rail, which stops the converter.[8] Alexandre Souza of Tabajara Labs repaired an FS-A1GT that would not start because +5 V was missing; it worked from an external 5 V supply. His notes, based on the FS-A1GT, FS-A1ST and FS-A1WX manuals, say the supply is the same design on all the Panasonic machines in this case style apart from the switching transistor, and that the ±12 V regulators rarely fail.[10]
Souza's procedure, with the FS-A1GT's own designators:[10][8]
- Unplug the machine and lift one end of L3, the small 20 µH inductor between the converter's output and the crowbar. This disconnects the load and the crowbar while the converter keeps its feedback, which is taken from the C10 side of L3. Do not lift L2, the large 220 µH inductor: that opens the feedback loop and the output rises to about 22 V. (Souza's post calls these parts L2 and L6, from another model's manual.)
- Power up and measure across C10. Up to 5.2 to 5.3 V with no load is acceptable.
- If the output is missing or wrong, check for about 22 V at pin 6 of IC1 and at the emitter of Q1. If it is missing there, look for a burnt track between the rectifiers and the converter, and bridge it with wire.
- If the input is present, replace Q1 (2SA1643) and IC1 (MC34063), then check the resistors round IC1 on both sides of the board, including R11 (0.2 Ω).
- With the converter working, check the crowbar. A shorted D1 holds the rail at zero. Panasonic lists D1 as a 5P4M-type thyristor (DED5P4M) and D9 as a 5.6 V chip zener (DEDDTZ5R6ATT).[11] Souza found the thyristor on his machine burnt and fitted a TIC112.[10]
- Refit L3 and check that the +5 V rail holds at 4.75 to 5.25 V with the machine running.[4]
The parts list gives Q1 as a 2SA1598, while the circuit diagram labels it 2SA1643.[11][8] Read the part on the board.
±12 V rails
[edit | edit source]Panasonic's readings: IC2 about +21 V in (pin 1) and 12 ±1.2 V out (pin 3); IC3 about −22 V in and −12 ±1.2 V out (pin 3).[4] The guide's text names pin 1 as IC3's input, but on the circuit diagram and the silkscreen pin 1 of IC3 is ground and pin 2 is the input.[8] On the slots, +12 V is pin 48 and −12 V is pin 50.[9] Souza notes that the 12 V rails of an MSX feed mainly the audio and video circuits, so a missing one shows up there before it stops the machine.[10] Check the regulators' input capacitors (C4 for IC3) and the rectifiers D4 and D7.[8]
Clock test points
[edit | edit source]Panasonic's clock checks, all on an oscilloscope:[4]
| Signal | Test point | Scope setting in the guide's figure |
|---|---|---|
| CPU clock | IC12 (VDP) pin 8 | 2 V, 0.1 µs/div |
| VDP clock | IC12 pin 64 | 2 V, 50 ns/div |
| Real-time clock | IC36 (MSX-Engine) pin 21 | 2 V, 20 µs/div |
| Floppy controller clock | IC26 pin 45 | 2 V, 50 ns/div |
| FM (OPLL) clock | IC10 pin 8 | 1 V, 0.2 µs/div |
| R800 clock | IC37 pin 66 | 2 V, 20 ns/div |
The crystals are X1 (16 MHz), X2 (32.7 kHz), X3 (28.6 MHz) and X4 (21.47 MHz).[11] The hard-check floppy pins the 28.636 MHz oscillator to the R800 (IC37), the 21.4772 MHz oscillator to the VDP (IC12), the real-time clock to X2 and IC36, and the MIDI timer's 16 MHz clock to the floppy controller's oscillator.[12]
Picture faults
[edit | edit source]Panasonic's list of picture faults and where to look (soldering faults or failed parts in each case):[12]
| Symptom | Where to look |
|---|---|
| Wrong colours | One of the R, G or B signals between the VDP (IC12) and the colour encoder (IC11) |
| No colour (on S-video) | Transistors and resistors in the chrominance path to the S-video socket CN8 |
| Picture rolls | The CSYNC# signal path between IC11 and IC12 |
| Odd colours and a rolling picture | Transistors and resistors in the luminance path to CN8 |
| Corrupted characters or graphics | The VDP (IC12) or the video RAM (IC13 to IC16) |
| Blurred picture on RGB: YS voltage wrong | R86 and R84 |
| Blurred picture on RGB: AVCONT voltage wrong | R85 and R87 |
For video checks with a scope, type COLOR 15,15,15 in BASIC and look at the VDP's R, G and B outputs on IC12 pins 22, 23 and 24 and its sync on pin 6, then the encoder's R, G and B outputs on IC11 pins 21, 22 and 23, sync on pin 11 and composite video on pin 20. On the S-video output, Q3 to Q6 amplify the luminance and Q7 to Q9 the chrominance.[4]
Hard-check floppy results
[edit | edit source]Panasonic's diagnostic disk (management number DF-4) is not in the hosted scan. Its error tables still name the parts behind each kind of fault, which is useful without the disk.[12]
| Test | Parts named for a failure |
|---|---|
| Main RAM read/write | Main RAM IC38 to IC41 and their decoupling capacitors |
| Main RAM size | Upper address lines MA17 to MA19; R352, R354, D301 |
| Memory mapper register | IC36 |
| Video RAM | IC13 to IC16 (CAS0 for IC13 and IC15, CAS1 for IC14 and IC16); IC12; decoupling capacitors |
| SRAM | SRAM IC19, IC18 (HC02), Q201, decoupling capacitors; address lines MA13 and MA14 for a size error |
| ROM checksum | ROMs IC20 and IC23, system IC35, IC25, D203, D204, R229, broken or shorted tracks |
| Keyboard input | Flexible pattern not fully in its connector, poor contact, broken pattern; IC36 pins 121 to 141 |
| Keyboard layout (JIS or kana) | IC36 pin 114 (KLS) |
| SRAM and RTC | IC19, IC18, IC35, IC36 |
| Printer port | IC46, Q206, IC36; connector CN4 |
| Joystick port 1 / port 2 | CN10 and LC301 to LC309 / CN9 and LC310 to LC318; IC35, IC36 |
| Floppy write protect | Floppy controller IC26, connector CN13, the WP track between them |
| PCM recording | Microphone amplifier round IC4, filter round IC8, A/D converter round IC5 to IC7, IC35, RA2 |
| Sprites | VDP IC12 |
| Natural image (YUV) display | VDP IC12, colour encoder IC11 and its circuit |
| FM sound | IC10, IC8 and the filter round them |
| Z80 mode | R800 oscillator (28.636 MHz) or VDP oscillator (21.4772 MHz); IC35 |
| Turbo and super turbo | R800 IC37, the WAIT# and RBUSRQ# signals; R352, R354, D301 |
| RTC | IC36 and X2; IC12's oscillator |
| MIDI timer | Oscillators of the floppy controller IC26 and the VDP IC12 |
| MIDI interface | CN5 and CN6, photocoupler IC34, IC31 (8251), IC32 (8254), IC27 to IC30, decoders IC33 and IC42 to IC44 |
| MIDI and timer interrupts | IC29, IC30, IC32, IC33; the EXINT# track |
Sound faults
[edit | edit source]Panasonic's sound checks use two short BASIC programs:[4]
- For the PSG, run
10 PLAY"V15C","V15C","V15C"/20 GOTO 10and look at IC36 pins 97, 98 and 99 (the PSG outputs), then at IC8 pin 7 (the output op-amp). - For FM, run
10 CALL MUSIC/20 PLAY #2,"V15C"/30 GOTO 20and look at IC10 pin 14.
The PSG, the PPI click and the sound from both slots are mixed and pass through IC8 to the audio output.[4] If the PSG signal is present at IC36 but missing at IC8 pin 7, the fault is in the mixer and IC8. If FM alone is missing, check IC10, IC8 and the filter round them.[12] The audio output is on the RGB DIN socket; there is no separate audio socket.[2]
Floppy drive faults
[edit | edit source]An FS-A1GT owner on the MSX Resource Center forum described the usual failure: the drive gave a high-pitched sound, the disk no longer turned and every access ended in "Disk not ready". The replies pointed to the belt, and a second owner reported the same symptoms on his machine after fifteen years.[13] Other owners found the original belts so far gone that the rubber had melted onto the pulleys. The residue had to be cleaned off with isopropyl alcohol before a new belt would stay on, and some replacement belts were too narrow or too loose and slipped or came off.[14] The MSX Resource Center gives the belt as 0.3 mm thick, 3 mm wide and 58 mm in diameter. A forum member gives the original's circumference as about 180 mm (about 90 mm when laid flat without stretching) and says the same belt fits the FS-A1F, FS-A1FX, FS-A1WX, FS-A1WSX, FS-A1ST and FS-A1GT.[2][14]
If a drive with a new belt turns and steps but BASIC reports "Disk offline", the forum's suspects are a belt of the wrong size, dirt on the heads (one drive recovered after a cleaning disk), worn heads and head misalignment. An unstable rotation speed, seen on an oscilloscope at the index sensor, points to the belt or the disk.[14] Replacement belts lasted five to ten years for two owners.[14]
A drive that writes to protected disks, or refuses to write to unprotected ones, fails the hard-check floppy's write-protect test. Panasonic points to the controller IC26, connector CN13 and the WP track between them.[12] Fitting the belt is covered in the Panasonic FS-A1GT General Maintenance guide.
Related pages
[edit | edit source]- Panasonic FS-A1GT
- Panasonic FS-A1GT General Maintenance
- Panasonic FS-A1GT Capacitor Replacement Guide
- Panasonic FS-A1GT Technical Guide
- Capacitor Failure Symptoms
- Floppy Drive Repair
References
[edit | edit source]- ↑ 1.0 1.1 Matsushita Electric Industrial Co., Technical Guide No. 107, FS-A1GT, February 1992, printed page 5, in-service notes: with the built-in software switch on the front panel set to on, the VSHELL screen appears at power-on, unless a ROM cartridge such as a game is fitted; with the switch off, BASIC starts. Section 3: the backup RAM keeps the word processor's learning dictionary and BASIC SET settings, powered with the clock by two AA cells, which are not fitted at the factory. See Panasonic FS-A1GT Technical Guide (File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF page 5).
- ↑ 2.0 2.1 2.2 2.3 Panasonic FS-A1GT, MSX Resource Center wiki, 14 August 2026—link(accessed 2026-10-21) Community page: holding ESC during boot erases the SRAM settings; MSX-MUSIC bug: CALL MUSIC without any OPLL channel (for example CALL MUSIC(0,0)) followed by a tempo command such as PLAY"T120", or a program stopped with CTRL-STOP, crashes the machine, avoided with CALL MUSIC(0,0,1); composite video comes from the RGB connector through the supplied cable; floppy drive belt 0.3 mm thick, 58 mm diameter, 3 mm wide.
- ↑ 3.0 3.1 MSX hardware fixes and enhancements– FRS (sd_snatcher)—link(accessed 2026-10-21) Section 2.3, FS-A1GT jailbar fix (credited to Leonard Oliveira): J1 and J2 select CVBS or CSYNC on the DIN RGB output, CVBS by default; cut J2 and close J1. Section 2.2, FS-A1ST and FS-A1GT audio: slot sound too loud, OPLL sound too quiet and over-filtered, global output weak; FS-A1GT values C17 and C18 1.8 nF, R50 22 kΩ, R47 15 kΩ, R55 47 kΩ, C51 180 pF, C45 150 pF, all 0805 SMD.
- ↑ 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 Technical Guide No. 107, printed pages 14 to 16, pass/fail checks. Page 14, power circuit: +5 V correct if the output at the collector of Q1 is +5 V ±0.25 V; +12 V correct if pin 1 of IC2 is about +21 V and pin 3 is +12 ±1.2 V; −12 V correct if pin 1 of IC3 is about −22 V and pin 3 is −12 ±1.2 V. Clock waveforms: CPU clock at IC12 pin 8 (2 V, 0.1 µs/div), VDP clock at IC12 pin 64 (2 V, 50 ns/div), RTC clock at IC36 pin 21 (2 V, 20 µs/div), FDC clock at IC26 pin 45 (2 V, 50 ns/div), OPLL clock at IC10 pin 8 (1 V, 0.2 µs/div), 16-bit CPU clock at IC37 pin 66 (2 V, 20 ns/div). Page 15, video: after COLOR 15,15,15, VDP R, G and B at IC12 pins 22, 23 and 24, CSYNC at IC12 pin 6, encoder R, G and B outputs at IC11 pins 21, 22 and 23, SYNC OUT at IC11 pin 11, composite at IC11 pin 20; the S-video luminance is amplified by Q3 to Q6 and the chrominance by Q7 to Q9. Page 16, audio: PSG output at IC36 pins 97, 98 and 99 and op-amp output at IC8 pin 7 with test program 1 (PLAY"V15C","V15C","V15C"); FM output at IC10 pin 14 with test program 2 (CALL MUSIC, PLAY #2,"V15C"); backup circuit test with test program 3. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF pages 14 to 16.
- ↑ Technical Guide No. 107, cover (printed page 1), specification: AC 100 V ±10 % 50/60 Hz, 22 W. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF page 1.
- ↑ Turbo R GT Power Supply gone?, July 2016 to February 2020—link(accessed 2026-10-21) The FS-A1GT owner found no 12 V output from the transformer and fitted the transformer from an FS-A1ST, after which the GT worked; Wierzbowsky: the board takes about 12 V AC from the transformer, measuring 14 to 15 V; RetroTechie: most transformers have a thermal fuse packed inside the windings that blows on overheating; replacement with a 230 V 12-0-12 V 36 VA Elma TT transformer (Conrad 516007) confirmed on GTs.
- ↑ Technical Guide No. 107, printed page 65, mechanical parts list: E1 power transformer DDT5L6J01Z, E3 power switch ESB70707V, E4 AC cord DFJA20ZAKH. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF page 58.
- ↑ 8.0 8.1 8.2 8.3 8.4 8.5 Technical Guide No. 107, printed pages 35, 36, 41 and 42: main board layout (component side) and circuit diagram sheet 3. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF pages 34 and 37. Mains through the power switch to the transformer, whose primary is drawn with a fuse element inside the transformer outline; CN1 (B3P-VH, pins CT, T1, T2); L1; rectifiers D8 and D3 (30D1FC) into C9; IC1 (MC34063 on the silkscreen) with Q1 (2SA1643), R11 (0.2 Ω, 1/2 W), D2 (Schottky), L2 (220 µH) and C10 (10 V, 4700 µF); feedback divider R2 (3.09 kΩ) and R4 (1 kΩ, 1 %) from the C10 node to IC1 pin 5; then L3 (20 µH) to C3 (6.3 V, 470 µF), with zener D9 (5.6 V), R1 (1 kΩ) and C12 triggering thyristor D1 across the output; D4 and D7 (ERA15-01) to IC2 (µPC24M12HF) and IC3 (AN7912T), C4 (35 V, 470 µF) at the input of IC3, C5 and C7 (16 V, 100 µF) at the outputs. The silkscreen marks IC2's pins I, G, +12 and IC3's pins G, I, −12.
- ↑ 9.0 9.1 MSX Cartridge slot, MSX Resource Center wiki, 6 July 2026—link(accessed 2026-10-21) Pins 41 and 43 GND, 45 and 47 +5 V, 48 +12 V, 50 −12 V.
- ↑ 10.0 10.1 10.2 10.3 A fonte do MSX Turbo-R (e dos FX, WX e WSX tambem!)– Alexandre Souza, 9 July 2018—link(accessed 2026-10-21) In Portuguese. An FS-A1GT that would not start had no +5 V and ran from an external 5 V supply; the power supplies of the FS-A1FX, WX, WSX, ST and GT are the same apart from the switching transistor (2SA1643 suits all); the +12 V and −12 V linear regulators rarely fail; when the 5 V supply fails, suspect the MC34063 and the transistor first, then an open resistor on the underside of the board or a shorted crowbar; remove the small exposed-copper inductor of about 1 cm diameter to isolate the supply and crowbar (not the large inductor, which breaks the feedback loop and sends the output to about 22 V); up to 5.2 to 5.3 V unloaded is acceptable; on this GT the crowbar thyristor (5P6J) had burnt and was replaced with a TIC112, and the SMD 5.6 V zener under the board can be replaced with a through-hole one; check for about 22 V at pin 6 of the MC34063 and at the transistor's emitter, and if it is missing look for a burnt-out track from the rectifiers and bridge it with wire. The designators in the post (L2 small, L6 large, R96 and R182 in the feedback) are from another model's manual.
- ↑ 11.0 11.1 11.2 Technical Guide No. 107, printed pages 59 to 62, electrical parts list. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF pages 53 to 56. D1 thyristor DED5P4M; D2 DEDSB350500L; D3 and D8 DED30D1FCXN; D9 and D206 chip diode DEDDTZ5R6ATT; Q1 2SA1598 in the transistor list; R11 ERX12SJR20P; IC1 DA34063AML (DC/DC converter), IC2 DAC24M12HF and IC3 AN7912T (regulators).
- ↑ 12.0 12.1 12.2 12.3 12.4 Technical Guide No. 107, printed pages 17 and 27 to 32, hard-check floppy (management number DF-4): test list, equipment (the floppy, a television or monitor, a printer such as the FS-PC1, a joypad and a MIDI cable), procedure, and for each test the parts to suspect. File:Panasonic FS-A1GT Technical Guide No. 107.pdf, PDF pages 17 and 27 to 32.
- ↑ FS A1 GT disk drive - some belts in stock?, October 2008 to January 2009—link(accessed 2026-10-21) FS-A1GT drive: high-pitched sound, disk not rotating, "disk not ready" every time; replies recommend a new belt from Repair-Bas; hamlet reports the same symptoms on his turbo R after 15 years.
- ↑ 14.0 14.1 14.2 14.3 Turbo-R Panasonic floppy drives, December 2017 to February 2021—link(accessed 2026-10-21) FS-A1ST and FS-A1GT drives: original belts melted onto the pulleys; residue cleaned off with isopropyl alcohol; several belts tried before one fitted (Grauw); "disc offline" after a new belt cured by a head-cleaning disk (gdx); head misalignment found on one GT drive (Wolverine_nl); Josevil gives the original belt as about 180 mm in circumference (about 90 mm folded flat without stretching), the same for the FS-A1F, FS-A1FX, WX, WSX, ST and GT; heads wear with age (Josevil); an unstable rotation speed usually means a bad belt or a bad disk (sd_snatcher); replacement belts lasted about 5 to 10 years for two owners.