• Please review our updated Terms and Rules here

Homebrew TVT I picked up

I just wish I could figure out what is going on. Every now and again while prodding around I can get the screen to blank and get slightly more reliable data. I've tried gently flexing the cursor board, pressing on certain spots.. one time I just pressed the card a little harder into its (broken) socket.. and the screen blanked. I don't know if it's something I'm doing or just coincidence. But whatever it is, in the right circumstances, it will clear all and only the nonsense characters onscreen, leaving whatever is being output to the TVT onscreen. It'll be interesting to see what happens when I bypass said eprom.

I'm wondering if this card socket is a problem. It was broken in two by shipping. I managed to realign everything and tighten.. but maybe one of the pins in the area of the break isn't making good contact.
 
I have one. Just haven't quite learned how to use it.

I did make another discovery in between events. When I press down on a particular address pin (I think it's A4), I start getting more complete strings of info, if the 6800 is connected. Like, as many numbers as should be appearing per line of S record.

The socket is the type where the pins make contact on the backside. I don't know if my pressing it there is just helping it, or if there's a problem in the socket. I'm going to investigate that shortly and if need be, replace it.

Unfortunately the 1702's program pin was pretty rotted, and fell off. I'm assuming that isn't anything to worry about since we won't be programming this again.
 
Well, here's your proof it's capable of working:

20190701_180754.jpg

Again, EPROM is still in place.. what seems to do the trick is putting my logic probe, or something small enough to wedge into the pin and socket. I got the 6800 going on a punch, and then just kinda put my logic probe tip in there, applied pressure, and after about 20 seconds it suddenly cleared the screen and began showing normal stuff like above. Soon as I let the pressure off, it loses it again.

I don't know really what's going on. I thought maybe my logic probe might be causing some kind of interference or voltage boost, but I'm able to do with with a screw driver tip too. Further, I can see with the logic probe data coming in to that particular pin, so it's not like it's getting a bad connection. I'm wondering if my pressure on the pin is incidental, maybe I'm flexing the board or that damaged socket just enough to make it work for a time.
 
I just discovered something else. If I'm applying pressure to said spot and hit reset on the 6800, it automatically blanks the screen and brings up the SWTBUG prompt. I did this a few times to check.

It also suddenly changes the horizontal sync. I have to resync the monitor. Previously I could only get the characters to be legible if I synced it in a way that brought the whole screen to the left. When I get it to successfully clear screen, it lets me sync things more to center.
 
I guess it must be. It's odd though.. I've tested continuity from the tops of the pins on the 1702 to the bottom of the socket pins on the other side.. all good.
 
falter said:
I guess it must be. It's odd though.. I've tested continuity from the tops of the pins on the 1702 to the bottom of the socket pins on the other side.. all good.
Click to expand...

I've seen them where your pressing on the pin of the part and it makes contact, you take the probe off and the contact is lost.
Dwight
 
For a test, I pulled the 1702 and wired up a cambion socket with address lines patched over to data out lines. I *think* I did it correctly.. I don't know why but the address lines and data lines are numbered differently on the diagram I have - it has Data 1-8, and address A0-A7. I assumed I needed to do A0 to D1, A1 to D2 and so on. If I got that wrong, let me know. Anyway, I plugged it in, made sure it had continuity, and could see signal hitting each pin. However, what I got was worse than with the EPROM.. just a couple of characters wide of garbage.

I'm wondering if I wired wrong?
 
Is there anything a logic probe does electrically that could change the behaviour of what you're probing? I mean, apart from making an physical connection between two things.

The more I play with this, the less I am convinced this is a bad socket or something loose. I've noticed I can probe the 157s, hold the probe on the output pins and still change the output to better output, although perhaps not as good as probing one of the EPROM pins. I've also noticed I can get this result without applying any downward force. It's just kind of odd. I'm wondering if the probe does something with voltages or clocks.
 
If it is, your screwdriver does it too.

I agree that it's probably the socket. But it could be as strange as a rotted conductor inside the chip. You can rule out the socket by running wires from the bottom round to the top. I'd be careful about assuming that the bad connection is the same pin you are touching though. I wouldn't be shocked to find that it's a different pin on that same socket.

I have a TV that would only work if I touched the side of the flyback transformer. It didn't seem like the kind of problem a flyback would have, more like a cold solder joint. But the only thing that would make it work was just touching the flyback. So I bought a replacement flyback which turned out to be bad, and burnt out an unobtainable component. Upon finding that now bad component, I found a cold solder joint about 10"away from the flyback which was the original problem.
 
Last edited:
Actually, you may be on to something. The 1702s are PMOS. They pull up hard but down, they are weak. This is just the reverse of TTL. In places where they may have troubles driving TTL, they'd put 10K or there about pull down resistors, to ground, on the output pins. This increase the pull down of the output drive.
I would think your D1 to A0 and on should work, unless they'd cross wired someplace else. We know from your A9 fix that the builder didn't always follow the normal wiring. This may explain the need for the EPROMs. It may just have been a patch to make the incorrect wiring work. The EPROMs always looked like an after thought, rather than a design intended.
You really need to start using your scope. It is so much better than your logic probe. The logic analyzer would show much more but without a schematic, it would difficult to use. With the scope, you could trigger on the strobe on the input side and trigger on the vertical on the video side. You can then follow signals from the keyboard all the way from the keyboard to the video with an ASCII table.
Dwight
 
Last edited:
I could not make out what the output was with the wires in place. It basically confined itself to the very first column and produced junk that overwrote itself so quickly I had no time to make it out.
 
I just tried replacing the EPROM with wires again and typing.. basically nothing coming out... just cursor moving as I'm hitting keys. I can occasionally see bits changing on the address lines. But that's it. Cannot get a read on the data lines, they are pulled low by resistors.
 
falter said:
I just tried replacing the EPROM with wires again and typing.. basically nothing coming out... just cursor moving as I'm hitting keys. I can occasionally see bits changing on the address lines. But that's it. Cannot get a read on the data lines, they are pulled low by resistors.
Click to expand...

Ahh! That may be the issue. The EPROM works well with pull down resistors ( remember PMOS ). TTL doesn't like pull downs. It has a weak pull up. That is most likely why it is failing.
Dwight
 
The end not connected to the 1702 data out pins is connected to ground.

Here is a pic of them... I've always had trouble reading these. My DMM says 470ohm but I don't know how accurate that is.

20190702_220628.jpg
 
You must log in or register to reply here.
Back
Top