Expected Effects of a Bad Transistor

Thread Starter

nelno

Joined Dec 3, 2020
4
Background (you can skip this bit and go to the actual problem / question sections, but I'm not sure how much of this will be useful in answering it):

Sorry if this is the wrong group (maybe consumer electronics would be better?), but basically I'm hoping that a test result I'm seeing can be explained.

I'm trying to fix an old (c.a. 1983) SpectraVideo 328 computer. The machine is mostly working but has no video output. I can tell the computer is mostly still working because I can type programs in blind and get the expected effects, hear key clicks, etc. It's literally just the video.

Using a multimeter I have verified all of the voltages are correct where I expect them to be. On my oscilloscope I'm not seeing anything odd on any of the DRAM, etc. Everything looks pretty good overall so I don't think there's any systemic failure due to the new power supply which is only very slightly different in amperage: original is 16VAC 0.8A and 9VAC 1.5A and the new is 16VAC 1A, 9VAC 1.5A. Tell me if I'm wrong but my understanding is the amperage difference doesn't really matter as long as it's = to the original.

The problem:
In testing with an oscilloscope and multimeter, I noticed that the component video out (pin 36) on the TMS9918A video chip is basically flat-lined. I have placed this same chip in another computer that uses it (a TI/99-4A) and it works perfectly and I can clearly see compositve video signal on pin 36 there. I've also put that computer's chip in the non-working computer and it gets the same flat line on the component video out pin.

The schematic for the component video out can be seen in the attached pic (source here: http://www.samdal.com/SVIDOCS/STM-A_SVI328.pdf), the COMVID input comes directly from pin 36 on the TMS9918A. I have tested the T2 transistor after that (removed from the board), which is directly in line to pin 4 on the DIN plug that carries the composite video signal and it looks bad to me (infinity both directions between emitter and base, and about 5 ohms both ways between collector and base. The 470 ohm resistor is also showing about 350 ohm which is pretty far off of its 5% tolerance rating.

The question:
Since all of the other pins on the chip seem to have appropriate signals on them (more or less... I cannot say for absolute certainty) and the chip works in the testbed computer, is it possible that the failed transistor (or I guess resistor) is causing the flatline of the COMVID pin, even though the COMVID pin is outputting to the transistor? I just don't see any bad inputs to the chip and from what I know (which, admittedly, ain't much) the transistor and resistor seem to be the only issues with this circuit. If so, what exactly causes that effect?

comvid_schematic.PNG

Some info on the TMS9918A VDP chip and its pinouts can be found here: http://www.mainbyte.com/ti99/hardware/chips/tms9918a.html

Thanks,
-Jonathan
 

Thread Starter

nelno

Joined Dec 3, 2020
4
Sorry, I completely missed the Technical Repair forum and I suspect this would have been better-suited to that forum. Is there a way for me to move the post?
 

AlbertHall

Joined Jun 4, 2014
11,313
You have measured 'infinity' between base and emitter so there is no way that the input signal (on the base) can get to the output (on the mitter).
 

Thread Starter

nelno

Joined Dec 3, 2020
4
You have measured 'infinity' between base and emitter so there is no way that the input signal (on the base) can get to the output (on the mitter).
Right, that much I understand. But at that point what I don't quite understand is what happens to the signal. The chip is still generating it, so either it just doesn't travel from the pin because the electrons essentially have nowhere to go, or maybe it's there but effectively canceled out by many signals overlapping.

I appreciate the help, though! I have a replacement transistor (and resistor, though that's working, just out of spec) on the way.
 
Top