I am helping a friend with his IBM 3279 terminal. I got the PSU working after reverse engineering it. Concluded that the SMPS control chip was a standard TDA 2640 in IBM disguise.
But of course the thing didn't work with the PSU fixed. A fuse which was in series with the vertical deflection coil on the analogue board went toast together with a 200 ohm resistor in the vertical deflection circuit. https://i.imgur.com/O4YEsdL.jpg At first I thought this had to do with some kind of fault in the vertical deflection coil itself. It measured only 1.2 ohms. But checking the IBM MAP manual revealed that both deflection yokes are supposed to be less than 2 ohms. So there has to be something else. I powered up the analogue board with the 5V and 8.5V that is normally supplied by the PSU (left the 103V unconnected since it used for the HV mostly) just to see what happened. The power consumption was reasonable, 0.3 amps on the 5V and 0.16A on the 8.5V. I then probed around among all those annoying IBM marked chips and square aluminium boxes. I found one chip that had 250 kHz on a pin. The chip appeared to be a divider since there were also 125 kHz, 62 Khz 31 kHz and 15 kHz. It seems to be originating from a square aluminium box. And then I found a signal which was around 20 Hz on another chip. As far as I understand the 15 kHz would fit well with the 64us sweep in the manual. But what is the purpose of the 20Hz signal? Vertical deflection? http://bitsavers.trailing-edge.com/pdf/ibm/3279/SY33-0069-3_3279_Color_Display_Terminal_Maintenance_Information_Part_1_Nov81.pdf According to the above manual the frame sync is 10 ms for model 2 and 13.8 ms for model 3 (this is a model 3). And moreover if 20Hz is the frequency used for generating the vertical it could explain why the fuse blew since the current in the deflection yoke should normally increase linearly with time. Three times longer deflection period would yield a three time higher current, thus very likely to exceed the 1.5A rating of the fuse. So I am looking for either another analogue board for an IBM 3279 or a schematic for the thing or maybe both would be great. IBM stuff is annoyingly hard to repair... Has anyone seen schematics for the IBM 3279? Does anyone have a spare analogue board? /Mattis