This is a short story of the Roland MT32 Midi repair.

The MT32 MIDI interface had an annoying audio problem (you can listen to the audio files attached to this post).

It seemed like the classic problem of exhausted electrolytic capacitors, but instead it was due to the ROM ROLAND R15449121 (TC534000P) position IC37.

To replace the original ROM it would have been enough to use an OTP Eprom 27C040 which however i didn’t have in the laboratory so i used a FLASHROM AM29F040 by modifying some connections.

The filename of the file used to program the FLASHROM is “r15449121.ic37.bin” which you can find in the archive attached to this post.

Below is the documentation how to adapt the AM29F040 instead of a 27C040.

• Atmel AT27C040 – 4Mb (512K x 8) OTP, EPROM (32-pin Package)
• Amd AM29F040 – 4Mb (512K x 8) FLASHROM (32-pin Package)

Tie 29F040 WE (PIN 31) and connect to VCC (PIN 32)
Tie 29F040 A18 (PIN 1) and connect to PIN 31 on Socket or PCB

Monkey Island Audio difference before and after repair:

• Before the Repair (322)
• After the Repair (322)

Gallery:

Download: Roland MT32 (various OS and extra ROMs) (409)

source: Synth Eprom Dumps

I wanted to remake the RGB connections of my CBS ColecoVision console a bit and design a new diagram for connecting the RGB video output.

This console was sold with an external modulator (CBS Electronics Adaptateur Antenne 7466 – 6F) that converted the signal from RGB to RF and it seems a Scart cable was not provided for direct connection to the television.

The original ColecoVision D-SUB DB15 to Scart cable that i have doesn’t have the correct pinout but it works perfectly with another ColecoVision always with RGB output but with an internal encoder that converts the Y (B-Y) (R-Y) to RGB signals (PAL Secam Converter A).

This fact is really curious, apparently there are more versions with RGB output as well as RF.

I decided to remove the old RGB cable that i soldered internally about 15 years ago and replace it with an RGB cable connected to the Coleco native connector.

The problem with doing this operation has always been the D-SUB DB15 connector which must be of a certain size, otherwise it won’t fit.

Brilliantly solved using a SNK NeoGeo Joypad extension cable, where we have a DB15 D-SUB connector pre-cast in the plastic where all 15 pins are used (in theory only 14 are needed for the NeoGeo Joypad, but all 15 are connected)

I cut the cable from the neogeo joypad male connector side and soldered the wires to a Scart connector and for the Fast Switching signal before connecting it to pin 16 of the scart i used a 100ohm resistor in series, i did not use the Function Switching signal.

One more thing, the Audio signal from the console native RGB connector is very low, so i made a jumper (see photo) to bypass a resistor and a ceramic capacitor, therefore i took the audio directly from PIN 7 of the SN76489 (Sound Generator).

Of course everything works perfectly.

Gallery of work done:

The return of the Hybrid Sid Player by xAD/Nightfall

Based on the Project and Hardware Schematic by Markus Gritsch released in the years 2011-2015 on the forum dangerousprototypes. Thanks Markus.

Changelog:

v2.00 – 11/5/2024

Added:

• Added a new parameter in the configuration file (config.ini).
  • Now you can also choose the FPGASid as well as the ARMSID for SID auto switching.

Below is an excerpt from the “config.ini” file.

# FPGASid/Armsid Auto Switching (0:Disabled – 1:Socket #1 – 2:Socket #2) – default: 0
# When you use this mode the other socket it does not activate automatically.
FPGASID_SOCKET=0

• Implemented auto switching for the FPGASid.

When choosing this mode the FPGASid must be installed ONLY in one of the two sockets,  you must also choose which socket to use for the FPGASid by editing the configuration file.

When configuring auto switching for the FPGASid during boot the SID Player is displayed the message “FPGASid Socket #” followed by the socket number where the FPGASid was configured in the configuration file.  This message is intended only to inform you that the configuration has been correctly activated.

*** Please read if you want to use the FPGASid in Auto Switching mode ***

The status LED indicating the SID that is used to play the song will remain fixed on the socket where the FPGASid was installed, therefore it will not change between Socket #1 and Socket #2.

The display of the SID model from the music track being played will consequently be displayed Only on the display.

The ‘sidSet’ mode will work exactly as before, the only difference is that if no SID is installed in the other Socket no sound will be played :-D while if a SID model 6581 or 8580 is installed you will be able to compare the fidelity of the FPGASid by switching between the two SIDs (obviously better if the same model).

The configuration of the model of the Sid with which to play the piece of music in the configuration file although working in this case it is to be considered Virtual because we would no longer have a change of state between Sockets for each SID model, therefore i decided to assign a static numbering to the SID.

• Write 1 for the SID 8580 model
• Write 2 for the SID 6581 model

For files in DMP format (conversion from RSID to RAW) playback is automatically forced with the 8580 SID of the FPGASid.

Jumpers to configure if the FPGASid is installed:

• Filter capacitors (JP4/JP5 or JP6/JP7): Does not matter.
• SID voltage 9v/12v (JP2 or JP3): Does not matter.
• Resistor 1k (JP10 or JP15): Better to remove.

Updated:

• Slightly fixed timing for SID song playback. Unnoticeable problem :-D

Fixed:

• Config.ini Parser BUG. The “ARMSID_SOCKET” parameter was not read correctly.

NOTE: The SID Player currently only supports the FPGASID for SID auto switching.

The Shugart SA455-3AA is a DSDD Floppy Disk Drive (360K).

The work that has been done:

• Replacement of the Electrolytic Capacitors on the Motor PCB.
• Greasing the head sliding rods with silicone grease.
• Reconstruction of the pcb track corroded by the loss of the 10uf electrolytic capacitor.
• Replacement terminals for selection with jumpers.
• Ceramic capacitor replacement.
• General cleaning.
• Read/write test with GreaseWeazle interface.

List of replaced electrolytic capacitors:

• 1 x 10uf 25v
• 1 x 47uf 16v
• 1 x 0.47uf 50v
• 1 x 0.22uf 50v
• 2 x 22uf 25v (non-polarized)

Gallery:

Download: Shugart SA455-SA465 Service Manual (342)

I decided to replace completely the PCB of the Amiga Power Supply (312503-03) because although after replacing all capacitors 1 year ago (30/4/2023) the voltages are not always stable, especially the 5V which fluctuated between 5.2v and 6.9v.

Using a Gotek HXC drive as a floppy device has burned two USB sticks, i recovered one replacing a shorted smd capacitor.

So let’s retire the old power supply pcb and welcome to the Mean Well RPT-60B fully compatible about voltage and amperage.

I leave you the Aliexpress link.

Gallery:

The usual boring replacement of all exhausted capacitors.

Gallery:

I wanted to try the SIDKick pico also with my SID Player, everything is ok, it works as well as i expected.

The green LED just below the SID Socket indicates that the SIDKick is playing.

Video:

SIDKick pico by Frenetic is a microcontroller-based C64 SID replacement. Analog inputs for paddles and mice are available just as for the original SID.

SIDKick pico uses firmware similar to its “big brother” SIDKick. However, the pico variant is based on the Raspberry Pi Pico board that uses the RP2040 microcontroller which is much cheaper and has better availability than the Teensy that is used by the normal SIDKick.

Technical features:

• Dual 6581 and/or 8580 emulation based on reSID 0.16 (optional: extension for digi-playing techniques).
• 2nd-SID address at $d400, $d420, $d500, $d420 + $d500 simultaneously, $de00, $df00 ($d500 is not available on the C128), or any address when an external chip select signal is used (e.g. on Ultimate 64 boards).
• Paddle/Mouse support.
• Built-in configuration menu (launch with “SYS 54301″/”SYS 54333″, also from C128-mode).
• Sound output via PWM (mono) through the C64/C128 mainboard, or in stereo using a PCM5102A-DAC-board.
• Flashing RGB-LED (WS2812) on GPIO 23 (installed on some Pico-clones)

My personal opinion:

My personal opinion as an inveterate purist is that no emulation or implementation will have the sound performance of the native SID (6581/8580).

I can accept as a good implementation the FPGASid and ARMSid where the developers have done a really good job.

SIDKick defends itself well and is low cost, furthermore it’s Open Source, also has a nice feature that i appreciated a lot, it has integrated configuration software that can be run with a simple SYS in both C64 and 128 modes.

So my final opinion is most people will never notice any difference, but i assure you that the differences are there, so i recommend it to everyone unless you are a SID purist.

Note:

The audio output and emulation improves if you use the PCM5102A-DAC add-on card, in my test i chose not to use it because i wanted to show you a plug’n'play without too many cables, jacks and patches. Maybe it was better to include this additional module directly on the SIDKick PCB? probably yes.

The KIT was purchased from restore-store.de, fast shipping and impeccable service.

Gallery of the shipping box:

Gallery:

Video:

source: c64-wiki.com github.com/frntc/SIDKick-pico restore-store.de

The First Amiga Intro by Lynx/Nightfall.

It only took just under two months to learn from scratch 68k Assembler and Amiga Custom Chip registers.

Well done Lynx!

Download: Retro Love Intro by Lynx/Nightfall (519)

source: pouet.net

First of all, i thank my friend Gerben for giving me the monitor.

This is a Black and White monitor but for some reason a thin transparent film has been applied to the front of the cinescope to simulate the tonality of an amber cathode ray tube, furthermore the APF logo placed on the front of the monitor to the right of the power light is masked.

The TVM-10 APF Monitor is approximately 43 years old and was used with the Computer APF Imagination Machine I/II.

The Monitor had several functioning problems, below is the description of the work that has been done.

• Removing the colored film.
• Replacement of all electrolytic capacitors.
• Replacement of a ceramic capacitor.
• Horizontal synchronism calibration (H-HOLD) by delicately removing the wax that was melted inside the variable inductor.
• Installation of a 1uf polyester capacitor in parallel with the 1.5uf one present for width adjustment horizontal (H-WIDTH) of the image, unfortunately the ferrite core inserted into the variable inductor for manual adjustment is blocked, probably with glue, forcing is not a good idea.
• Check of the Diodes.
• Calibration using the trimmer placed on the front of the PCB of the monitor for the power supply voltage which must be 12v.
• Geometry adjustment.
• General cleaning.

Only one geometry problem remains; the image on the top left side of the CRT tends to bend slightly for about 1cm and the left side of the image is slighty rounded, problem surely attributable to some polyester capacitor or resistance that has lost the capacity/value of the time.

Needless to say i checked almost all the polyester capacitors and resistors in the horizontal section and didn’t find nothing scandalous.

The photo gallery contains all the photos i made during the repair, i take this opportunity to thank a user of the plcforum who gave me tips on how to proceed with the repair.

Gallery of the repair:

123►

Apple ][ EuroPlus Repair.

Defects:

• Black screen.
• Power light off.
• Keyboard does not work.

Repair:

About the black screen it was enough to press all the IC in their seat, i have removed some and cleaned the IC's socket and the IC's pins.

The Power light off, the 5v micro bulb was faulty, replaced.

The keyboard had several problems;

The keyboard connector (socket) located on the computer motherboard side was faulty and consequently are missing the -12v for the keyboard encoder PCB.

The 3 TTL chips had been replaced on the keyboard encoder PCB but 2 x 74LS04 are erroneously soldered instead of 2 x 74LS00, the third TTL was correct but not working. Replaced all.

Now everything works reasonably well, unfortunately the bad keyboard has some problems, the internal slats are warped and tend out of their seat, as a result some keys do not work correctly.

I have tried to fix and clean, but the situation hasn't improved much.

Gallery of the repair:

After the usual boring PSU recap i have decided to add two displays to monitoring the main voltages (+5,+12) and the amperage.

The amperage displayed with the blue LEDs in the left display is the sum of all the voltages (+5,+12,-12).

I have also added three ventilation holes, i didn’t want to use any fans, i already have enough fans in the lab that are quite annoying.

Gallery:

Texas Instruments PHA 2037 Connector FIX

Defect:

• Black Screen but sounds works.

Replaced parts:

• 1 x 6 PIN DIN Male.

Gallery:

Texas Instruments TI-99/4A Repair #1

Defect:

• Black Screen and deafening sound in background.

Replaced parts:

• 1 x TMS9900NL CPU (U600)

Texas Instruments TI-99/4A Repair #2

Defect:

• Wave/Raster Lines.
• Garbage on Boot Screen.

Replaced parts:

• 1 x Inductor 2 -> 4.5 μH (L100)
• 2 x TMS4116-15 RAM

Texas Instruments TI-99/4A Repair #3

Defect:

• Garbage on Boot Screen.

Replaced parts:

• 1 x MOSTEK MK4116N RAM

Video amplifier installation for Composite Video output including installation of the two female RCAs on the case.

I also made a simple fix of the metal PAD of the RESET that no longer stayed in place and consequently fell off inside the shell.

Gallery: