PCXT
Very Active Member
- Joined
- Sep 14, 2016
- Messages
- 277
- Age
- 36
Recently I bought a Soviet microcomputer. It has been probably made in Moscow plant, designed in 1980s, but made until 1990s, my unit is from 1991, when it was sold either as a TV game, or as some kind of hobbyists tool. The computer is 8080-based, has some amount of 8255 for I/O and a 8253 timer. 2kB ROM consists of a monitor with rudimentary tape input routines to bootstrap the proper loader from tape. There is also a DMA controller, display generator (also some Intel clone) and 32kB of RAM in sixteen 4116-like chips. I'm not sure yet are these negative-voltage versions or not, but if they are and PSU was bad or someone tried to run it with 5V only, I'm deeply in the bottom.
I have visually inspected the mainboard. The laminate has a poor quality, much worse than Polish or even Ukrainian computers made these times. If something will be bad, I'll just borrow hot-air and try with it training on some old boards before, as using a typical soldering iron will destroy tracks quickly.
The keyboard looks and feels nice, but there should be definitely one more spring in space key. They installed one instead of two in the factory.
I have also looked into power supply unit and there was a thing which surprised me: There is a linear regulator in TO220 package fastened to the large, aluminum heatsink (mounted on holes for TO3 case which was probably not accessible). This heatsink is a typical extrusion, it has a flat part to mount components and a few fins protruding about 1.5cm, there's about 1.5cm between them. Fins end with smaller fins parallel to the flat part, leaving a few mm between.
According to all rules of power supply design the space between these fins should be kept clear of components to allow convection cooling. But in the factory they designed a PCB a bit different and decided to go a different way.
Let's put electrolytic capacitors in there! What can go wrong?
Yes, they put electrolytic caps inside the heatsink windows. I desoldered two and tested them. And after an hour of pumping their maximum voltage, measuring loss current and periodically checking ESR, another surprise: These things are NOT broken.
I have visually inspected the mainboard. The laminate has a poor quality, much worse than Polish or even Ukrainian computers made these times. If something will be bad, I'll just borrow hot-air and try with it training on some old boards before, as using a typical soldering iron will destroy tracks quickly.
The keyboard looks and feels nice, but there should be definitely one more spring in space key. They installed one instead of two in the factory.
I have also looked into power supply unit and there was a thing which surprised me: There is a linear regulator in TO220 package fastened to the large, aluminum heatsink (mounted on holes for TO3 case which was probably not accessible). This heatsink is a typical extrusion, it has a flat part to mount components and a few fins protruding about 1.5cm, there's about 1.5cm between them. Fins end with smaller fins parallel to the flat part, leaving a few mm between.
According to all rules of power supply design the space between these fins should be kept clear of components to allow convection cooling. But in the factory they designed a PCB a bit different and decided to go a different way.
Let's put electrolytic capacitors in there! What can go wrong?
Yes, they put electrolytic caps inside the heatsink windows. I desoldered two and tested them. And after an hour of pumping their maximum voltage, measuring loss current and periodically checking ESR, another surprise: These things are NOT broken.
