Ground lead is WAAAY too long, massive loop area/big inductor picking up heaps of junk. This is exactly what the little spring ground probes that come with scope probes are for!
Second this comment!! Especially if you are measuring in close proximity to a magnetic deflection CRT!! The ground loop is a single turn transformer secondary. The time derivative of the magnetic flux through the loop appears directly as a voltage across the input of your oscilloscope (Maxwell's equations). The CRT's deflection yoke creates a magnetic field that is both large and very rapidly changing.
GROUND LOOPS! GROUND LOOPS! GROUND LOOPS! When I was a strapping young man installing commercial A/V systems in Chicago, many moons ago, one system had a particular gooseneck lectern microphone that made a loud hissing noise at times. My boss couldn't find anything wrong, so he sent me in to try my troubleshooting skills. Turns out that this microphone had two paths to ground. One was the regular path, through the mic cable, to the mixer, and to the building ground through the mixer's power cable. The other was through the gooseneck, to the (metal) lectern, through a power strip mounted on the lectern, to a power outlet (on a different circuit from the mixer!) and to the building ground. This created a huge loop in space the encompassed much of the building. And, in the middle of this loop, was a 3-CRT video projector (yes, I know I just revealed my age). Naturally, whenever the projector was running, the microphone was swamped by flyback noise. That's what you have going on with your oscilloscope setup, except worse because you are trying to measure a single-ended signal. You have one loop involving the very long ground lead on your scope probe, and part of the computer chassis. You may have another loop involving the scope's line cord, the computer's line cord, and the chassis--unless you have been careful to isolate the computer's chassis from the building ground (which you should definitely do, for exactly this reason, when troubleshooting with a scope)
Oh, if you're curious, I fixed the issue in five minutes using a pair of scissors and the cardboard from the back of a legal pad, to make a spacer to isolate the mic from the lectern. Perhaps my proudest MacGyver moment.
To be fair, the brightness control was turned down preventing anything from being seen, so you did fix one aspect in the CRT section. :) Without having the schematic of the driver board, I'm going to go out on a limb, and suggest checking the IC that drives the video. Reaching back in my memory to these kind of circuits, I'm willing to bet it is a line driver type IC, like a 75 series line driver IC, or a simple transistor buffer. Outputs driving longer interconnect cables often get smoked from surges induced into them---the longer wire acts like an antenna, and picks up induced currents. Worth checking I think.
"dad"?!! It's so lovely to see a family working together on an electronics project! My daughter and I do most things together, from hang gliding to visiting museums, to electronics and physics problems to robotics and even playing computer games together (I royally suck at minecraft... ) and I never have as much fun as when I'm doing those activities with my daughter. Kids are awesome ❤
Yay! As a tech, we had to move pretty fast to earn a living. In the old days, I'd have just used the diode-test on the meter to bias Q101 on to see if the screen flashed in response.
I think you’d be able to make more sense of any signals by running another probe to H or V sync to trigger off. You have a repeating signal and a 2-channel scope after all!
Ah -- that display was actually from a different pin. The Flashwriter II card is weird. There's three pins for HSYNC, VSYNC, and TTL video signal. There's a *separate* pin with a composite output, which is what we tried in part 1. However, that composite output doesn't have the HSYNC and VSYNC mixed in properly, which is why the display was all over the place. Now, to get HSYNC and VSYNC mixed in properly for a proper composite output, you have to change some component values But if you do that, then the original HSYNC/VSYNC/TTL outputs don't work properly. It's kind of an absurd design -- you have to chose one or the other, I don't see why they couldn't just make a design where all the outputs work all the time without component value changes. I know where to look next -- the last thing the TTL output sees before leaving the card is an open collector inverter with a pull-up resistor. So the problem has to be there.
Ground lead is WAAAY too long, massive loop area/big inductor picking up heaps of junk. This is exactly what the little spring ground probes that come with scope probes are for!
Second this comment!! Especially if you are measuring in close proximity to a magnetic deflection CRT!! The ground loop is a single turn transformer secondary. The time derivative of the magnetic flux through the loop appears directly as a voltage across the input of your oscilloscope (Maxwell's equations). The CRT's deflection yoke creates a magnetic field that is both large and very rapidly changing.
GROUND LOOPS! GROUND LOOPS! GROUND LOOPS! When I was a strapping young man installing commercial A/V systems in Chicago, many moons ago, one system had a particular gooseneck lectern microphone that made a loud hissing noise at times. My boss couldn't find anything wrong, so he sent me in to try my troubleshooting skills. Turns out that this microphone had two paths to ground. One was the regular path, through the mic cable, to the mixer, and to the building ground through the mixer's power cable. The other was through the gooseneck, to the (metal) lectern, through a power strip mounted on the lectern, to a power outlet (on a different circuit from the mixer!) and to the building ground. This created a huge loop in space the encompassed much of the building.
And, in the middle of this loop, was a 3-CRT video projector (yes, I know I just revealed my age). Naturally, whenever the projector was running, the microphone was swamped by flyback noise.
That's what you have going on with your oscilloscope setup, except worse because you are trying to measure a single-ended signal. You have one loop involving the very long ground lead on your scope probe, and part of the computer chassis. You may have another loop involving the scope's line cord, the computer's line cord, and the chassis--unless you have been careful to isolate the computer's chassis from the building ground (which you should definitely do, for exactly this reason, when troubleshooting with a scope)
Oh, if you're curious, I fixed the issue in five minutes using a pair of scissors and the cardboard from the back of a legal pad, to make a spacer to isolate the mic from the lectern. Perhaps my proudest MacGyver moment.
This series has been very entertaining.
To be fair, the brightness control was turned down preventing anything from being seen, so you did fix one aspect in the CRT section. :) Without having the schematic of the driver board, I'm going to go out on a limb, and suggest checking the IC that drives the video. Reaching back in my memory to these kind of circuits, I'm willing to bet it is a line driver type IC, like a 75 series line driver IC, or a simple transistor buffer. Outputs driving longer interconnect cables often get smoked from surges induced into them---the longer wire acts like an antenna, and picks up induced currents. Worth checking I think.
"dad"?!! It's so lovely to see a family working together on an electronics project!
My daughter and I do most things together, from hang gliding to visiting museums, to electronics and physics problems to robotics and even playing computer games together (I royally suck at minecraft... ) and I never have as much fun as when I'm doing those activities with my daughter. Kids are awesome
❤
Yay! As a tech, we had to move pretty fast to earn a living. In the old days, I'd have just used the diode-test on the meter to bias Q101 on to see if the screen flashed in response.
I think you’d be able to make more sense of any signals by running another probe to H or V sync to trigger off. You have a repeating signal and a 2-channel scope after all!
Do I see a green clip lead attached to your ground clip?! lol. Nice antenna. 😉
hmm if that video pin is dead, how did you get video to display in part 1?
Ah -- that display was actually from a different pin.
The Flashwriter II card is weird. There's three pins for HSYNC, VSYNC, and TTL video signal. There's a *separate* pin with a composite output, which is what we tried in part 1. However, that composite output doesn't have the HSYNC and VSYNC mixed in properly, which is why the display was all over the place.
Now, to get HSYNC and VSYNC mixed in properly for a proper composite output, you have to change some component values But if you do that, then the original HSYNC/VSYNC/TTL outputs don't work properly. It's kind of an absurd design -- you have to chose one or the other, I don't see why they couldn't just make a design where all the outputs work all the time without component value changes.
I know where to look next -- the last thing the TTL output sees before leaving the card is an open collector inverter with a pull-up resistor. So the problem has to be there.
@@Lantertronics Ah great! I look forward to the next video.