PDA

View Full Version : TI 99/4A Video DIsplay problem



OS9dude
March 11th, 2014, 02:16 PM
After so many years I finally got my act together and jumped into the TI 99/4A wagon. Had a beat up machine inherited a couple of years ago which I had assumed was not working per the previous owner comments, the video modulator came badly corroded and with no grommets on the cables into it, power supply in equally bad physical shape; never took to installing the computer until today as I was testing a TI-99/4A I got delivered early morning.

The new unit is very well preserved, modulator in top shape as well as the accessory hardware it came with (voice module, cassette cable and joysticks). Upon installing it though I got a rather messed up video display:

17714

To check if the modulator was OK I tested with the older beat-up computer, got the correct boot up screen on that one (surprisingly running, I did not expect it to as noted above).

The newer computer seems to be working, I can blind type into TI Basic and see the screen reflect some changes when it goes through those motions, once in Basic I can FCTN-Quit to what would be the master screen - looking as shown on the picture. If I let the computer on it will time out to the blank screen properly, and will come alive when pressing the keyboard.

Needless to say I have some snooping to do on the TI board, maybe a voltage in the video circuitry is out... time to hit the knowledge bases to find a schematic and proceed.

Any hints on how to go about this video problem will be welcome!

Thanks in advance.

-- RP

TMA-1
March 13th, 2014, 02:20 PM
Your story sounds exactly what I just went through in the thread next door, here (http://www.vintage-computer.com/vcforum/showthread.php?41600-Bad-Video-on-TI-99-4A). The picture, although not exactly the same, suggests bad video RAM again. The fact that you can type blind and make things happen in the computer that seem correct, is exactly what I just experienced.

The bad news (for me) is that I unsoldered, extracted, socketed, and repopulated all 8 4116 chips, later testing the originals by trial error and finding there was only 1 bad one. What I've read since suggests there may be a shortcut that sometimes works.

First, verify the chip voltages. There should be -5V at pin 1, +12V at pin 8, +5V at pin 9, and ground at pin 16. (My -5V was nominal at -2.3V for some reason.)

If okay, then bend the legs of a replacement 4116 inward as you would to insert it in a circuit board anyway, and then piggyback it on to an existing chip to be tested, making sure the legs all make contact with the soldered-in chip. Sometimes this will appear to clear up the problem immediately, pointing to a bad chip underneath. And of course there may be more than one.

Good luck!

KC9UDX
March 13th, 2014, 02:30 PM
It's important to note that sometimes the problem doesn't go away by piggybacking a chip. Watch for any change, not necessarily an improvement. Any change at all when piggybacking a chip means the chip is bad. (Assuming you only attach the piggyback chip with the power off!)

OS9dude
March 13th, 2014, 03:16 PM
Your story sounds exactly what I just went through in the thread next door, here (http://www.vintage-computer.com/vcforum/showthread.php?41600-Bad-Video-on-TI-99-4A).

Thanks!. I did go into the other discussions looking for a similar report - after reading on your experience I went out and stocked on sockets for the 4116 chips thinking that would be the first thing to go after once the voltages out of the power supply are verified. I don't mind socketing all those chips, just as a precaution for any future developments on those memory ICs.

Also went on building a composite video cable to avoid using the modulator, will set it as a permanent output once the display issue is resolved.


-- RP

jltursan
March 17th, 2014, 12:28 AM
Indeed, socketing is a must. 4116 are really prone to faults and a nightmare to deal with if they're still soldered to the PCB.

TMA-1
March 24th, 2014, 07:56 AM
OS9dude, does your post in another thread mean you got this video fixed up ok? Let us know how it went.

OS9dude
March 25th, 2014, 12:53 AM
OS9dude, does your post in another thread mean you got this video fixed up ok? Let us know how it went.

I've just begun to remove the original memory chips from the board. After trying to clean desolder them I begun to get worried about heat damage to the circuit pads... decided to destructively remove the chips: clip the legs and then have the pin stubs removed with less heat applied to the pads; ran through half of the 8 chips hopefully to finish off later today. Found the 4116 on Jameco and got a 16 qty package last weekend (spares for a just in case).

Will update the forum with the end result once I install the sockets, the RAM and fire up the 99/4A

-- RP

OS9dude
March 26th, 2014, 03:42 PM
Done and repaired!

Posted a few pictures of the process at http://www.vintage-computer.com/vcforum/album.php?albumid=228

Unfortunately I can not tell which of the video RAM chips was bad, as noted before I decided to destructive remove the chips, the replacements from Jameco were not expensive so I saw no harm on ordering twice as much I needed and save the solder pads on the board from heat damage.

Thanks for the suggestions!


-- RP

Ksarul
March 26th, 2014, 06:28 PM
It is to resolve problems like these that I bought a fully-socketed TI motherboard from Jürgen Stelter in Germany many years ago. He socketed everything on one of them as an experiment. . .and sold it to me when he was done with it. It still works fine 20 years later.

KC9UDX
March 26th, 2014, 07:09 PM
Technically, a non-socketed one should be more reliable than a socketed one.

Ed in SoDak
July 20th, 2014, 02:16 PM
Nice work! I socketed the RAM in one of my consoles that died. It originally seemed to be a dead RAM chip, but new RAM chips in my sockets didn't fix it. I'm guessing I created a new problem with my repair that didn't look as tidy as yours. I recently moved it on as a parts console. The new owner couldn't get it going either!

I've seen bad cartridge port connections or rotting port wiper foam cause wierd displays and lockups, so check for the easy stuff too.

-Ed