falter
Veteran Member
Think I found my problem. One of the wires for the @-G keys was, I think, attached to the wrong switch pole. When I removed it temporarily, the bits started changing for the first time when I pressed other keys. Sheesh.
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
TV Typewriter replica project back on track
- Thread starter falter
- Start date
Chuck(G)
25k Member
So, if I understand correctly, for example. the digits 0-7 are connected between the "0-7" line and the "000" through "111" lines, right? You should be able to omit the keyboard temporarily and check this using a piece of wire, connecting between the appropriate pins. You should see a1-a3 outputs change state as you move from 0 to 7.
Don't worry about the "key pressed" output yet.
Don't worry about the "key pressed" output yet.
falter
Veteran Member
I think I am within an inch of solving this. Keypressed is working -- I actually connected it to the TVT (leaving A1-A7 disconnected) and it prints ? marks each time I type. Very exciting!
The only problem I am having appears to be with bit 1.. when I hit A for example, my code should be 1000-001... but bit 1 is shifting the wrong way. It goes low when I press A, instead of high. I think this is likely a wrong wire somewhere but will take a while to think it through and figure out where it's going wrong.
The only problem I am having appears to be with bit 1.. when I hit A for example, my code should be 1000-001... but bit 1 is shifting the wrong way. It goes low when I press A, instead of high. I think this is likely a wrong wire somewhere but will take a while to think it through and figure out where it's going wrong.
falter
Veteran Member
Okay so I noticed something odd. On all the 'letter/number' pins, I had a nice, clean, low signal on my logic probe. But when I hit the @-G pin, it was all squelchy and noisy. Anyway, it seemed to come down to the diodes along the trace that connected to that pin. I removed one end of each diode one by one until the noise went away and I had a nice clean low signal. The diode I disconnected was D14. To disconnect I just desoldered the anode side. Soon as I did that, the signal was clean again.
So maybe bad diode? I can't see anything else that would be acting on it. The cathode side of each of the three diodes connected to @-G are low by my logic probe.. there's nothing else connected to that trace other than those..
Which makes me wonder.. bits 1-3 are also making a squelchy noise when nothing is pressed.
So maybe bad diode? I can't see anything else that would be acting on it. The cathode side of each of the three diodes connected to @-G are low by my logic probe.. there's nothing else connected to that trace other than those..
Which makes me wonder.. bits 1-3 are also making a squelchy noise when nothing is pressed.
Chuck(G)
25k Member
If you unsolder D14, does H-O work okay? If not, then it's something a little more involved.
falter
Veteran Member
I didn't try that but can tomorrow. Definitely the tone on my logic probe sounds better.. and all I did was change diodes.
However I have the same ringing on A1-A3 but not A4-A7. When I disconnect A1-A3 from the TVT key operation is somewhat normal (minus the missing bits). I'm trying to follow it from Q2 upwards through all the transistors (Q1 has no ringing on one of its legs (I think it's the emitter? It's not the one connected to +5v or ground).
However I have the same ringing on A1-A3 but not A4-A7. When I disconnect A1-A3 from the TVT key operation is somewhat normal (minus the missing bits). I'm trying to follow it from Q2 upwards through all the transistors (Q1 has no ringing on one of its legs (I think it's the emitter? It's not the one connected to +5v or ground).
Chuck(G)
25k Member
Q1 shouldn't have any ringing, as the base-to-emitter circuit is a big (100 uF) electrolytic cap.
Before you go off replacing parts willy-nilly (small signal diodes rarely fail), consider this: Much of this keyboard encoder is operated as high-impedance stuff (note, for example, the 100K resistors at Q2-Q4). It isn't like TTL at all.
So the way to go about things is to eliminate any extraneous electrical noise.
Start by removing your keyboard and shorting the pads manually with a jumper wire--you could have noisy contacts in your keyboard, particularly if the switches are old.
If the noise persists, try using a known-good 5V power supply, not the TVT one.
Check your signal grounding--a high impedance circuit can magnify any stray noise.
Before you go off replacing parts willy-nilly (small signal diodes rarely fail), consider this: Much of this keyboard encoder is operated as high-impedance stuff (note, for example, the 100K resistors at Q2-Q4). It isn't like TTL at all.
So the way to go about things is to eliminate any extraneous electrical noise.
Start by removing your keyboard and shorting the pads manually with a jumper wire--you could have noisy contacts in your keyboard, particularly if the switches are old.
If the noise persists, try using a known-good 5V power supply, not the TVT one.
Check your signal grounding--a high impedance circuit can magnify any stray noise.
falter
Veteran Member
Don't know what I did but the ringing is gone now... now I just have wrong bits... for example pressing A should put bits 1 and 7 high (if I've read my ascii code right), but it only puts 7 high. A1 is high by default when you power up and switches low when A is pressed.
Gotta re-learn the schematic.
Gotta re-learn the schematic.
falter
Veteran Member
So as far as my testing can confirm, A1-A3 are the problem right now. A4-A7 appear to do exactly what they're supposed to when keys are pressed. But A1-A3 always go low when any key is pressed. Seems to be this way with the keyboard connected or disconnected. I really wish I understood the schematic better, but it looks to me like the alphanumeric pins are working properly.. and my problem is happening somewhere over on the binary pin side. Since A1-A3 start out high and shift to low only when a key is pressed, I'm convinced the hex inverters are working properly. It's like something is shorting when a key is pressed, when it shouldn't be.
Chuck(G)
25k Member
Take a look at what the base voltages are when you press a key on Q2,3 and 4. If they all change state no matter what key is pressed, then look at the base voltages of Q5, 6, 7. Also, check the polarity of the diodes D3-D11. A shorted one or one reversed could really throw a wrench into the works.
falter
Veteran Member
I regret I've never been good at testing diodes. I put my DMM on continuity.. put black lead on anode and red lead on anode and get:
D3 - 8.5M
D4 - 8.5M
D6 - No continuity
D8 - No continuity
D11 - No continuity
D12 - 35.5M
D14 - No continuity
Reversing the leads I get:
D3 - 3.8M
D4 - 1.0M
D6 - 1.0M
D8 - 1.0M
D11 - 1.0M.
D12 - 1.0M
D14 - 1.0M
Chuck(G)
25k Member
Doesn't your DMM have a diode range? My inexpensive one does. If not, I can help you set up an ersatz tester.
The general idea is this:
You measure the forward voltage drop of a diode directly using a current-limited low-voltage source. In the reverse direction, the voltage read will be nearly that of the source; in the forward direction, the voltage read will be that of the forward drop of the diode (about 0.7V typically).
The general idea is this:
You measure the forward voltage drop of a diode directly using a current-limited low-voltage source. In the reverse direction, the voltage read will be nearly that of the source; in the forward direction, the voltage read will be that of the forward drop of the diode (about 0.7V typically).
Last edited:
Agent Orange
Veteran Member
Lift one side of the diode and test for a 10 to 1 front to back ratio using the resistance (ohms) scale on your multimeter. Works sufficiently in most cases.
Chuck(G)
25k Member
falter
Veteran Member
VICTORY IS MINE
https://www.youtube.com/watch?v=nGPKfMnz8a4
Checked the diodes, they were all good. Turned out I had just two little problems in the encoder - one was solder bridging (or more correctly - flux bridging), and then the other was an errant connection from the 110 connector to 4, D, T, L etc that produced the wrong code when those were typed. They should have been (and were) connected to 100, but somehow in my efforts I had accidentally connected them to 110 as well. The TV Typewriter now works exactly as it should!
Sorry for the poor quality video. My brand new Note 8 phone had the lens covering the camera smashed by a screwdriver. I was so excited I just had to share.
This marks the end of this long project. Thank you to everyone who gave advice, especially Chuck! Cannot believe this thing actually works! When I think of how many people tried and failed to build this back in the day -- it's a really amazing feeling to know somehow I did it. This project was 6 years in the making with many nights hunting for information, parts, you name it. A little sad to be done, although some work remains (rebuilding the case), but so satisfying. Quite impressed. Thanks again - could not have done it without you all!
https://www.youtube.com/watch?v=nGPKfMnz8a4
Checked the diodes, they were all good. Turned out I had just two little problems in the encoder - one was solder bridging (or more correctly - flux bridging), and then the other was an errant connection from the 110 connector to 4, D, T, L etc that produced the wrong code when those were typed. They should have been (and were) connected to 100, but somehow in my efforts I had accidentally connected them to 110 as well. The TV Typewriter now works exactly as it should!
Sorry for the poor quality video. My brand new Note 8 phone had the lens covering the camera smashed by a screwdriver. I was so excited I just had to share.
This marks the end of this long project. Thank you to everyone who gave advice, especially Chuck! Cannot believe this thing actually works! When I think of how many people tried and failed to build this back in the day -- it's a really amazing feeling to know somehow I did it. This project was 6 years in the making with many nights hunting for information, parts, you name it. A little sad to be done, although some work remains (rebuilding the case), but so satisfying. Quite impressed. Thanks again - could not have done it without you all!
Last edited:
daver2
10k Member
It's good when a plan comes together!
Well done that man!
Dave
Well done that man!
Dave
Well done. A project I am interested in doing one day, after all my other projects. Also answers another question of who BradH on youtube is. You probably have the most SWTPC related youtube videos. If only you could talk through some stuff so I could understand them better 
.
.
falter
Veteran Member
Well done. A project I am interested in doing one day, after all my other projects. Also answers another question of who BradH on youtube is. You probably have the most SWTPC related youtube videos. If only you could talk through some stuff so I could understand them better
I am considering doing a more comprehensive video on the TVT and TVT-II. I'm not a natural on-screen personality and hate the sound of my voice as it comes out on camera. But yeah apart from TXDj there aren't a lot of youtubers covering anything older than 1980 or so.
falter
Veteran Member
Something interesting that I discovered today - one of the long held strikes against the TVT as a terminal device was supposedly that (owing to memory constraints) when you hit the bottom of the screen, it blanked the whole screen and then started from the top. Thus you'd lose valuable onscreen info or even what the computer on the other end was immediately sending to you if your cursor happened to be near bottom when you sent your command.
I have not observed this behavior - when I get to the bottom of the screen, the cursor simply goes back up top and starts typing over what's there. I wonder if the 'screen blank' issue only happens with some kind of serial board installed?
I have not observed this behavior - when I get to the bottom of the screen, the cursor simply goes back up top and starts typing over what's there. I wonder if the 'screen blank' issue only happens with some kind of serial board installed?