My new oscope arrived today : 72-3055

My scope usually stays in AC coupling mode. :)

(EWWW....AC coupling!) :ROFLMAO:

andrewb said:
I also mention it for pedagogical reasons, for folks who don't have a scope. All you really need is a DMM for checking the DC and AC voltages.

That last thing is a very good reason! Every repair person should have a decent scope. But most in this hobby don't have one.
The rest of that is just lazy scopesmanship and "precision delirium". Especially for someone with a 2465. :p

Try this with the scope:
Put both channels on DC coupling and set up the scope to display (CH1 - CH2) in YT mode.
Measure the same signal and get the difference zeroed out on the scope.
Change only CH2 to AC coupling, which removes the DC component.
Now your difference signal should be the DC component only!

You can use the PCB adjustment to minimize the difference signal on the scope.
And if you really need a number, measure the difference with the cursors.
(But you don't really need a number.)

Scopes are cool. :D
 
My scope usually stays in AC coupling mode. :)

I also just like to see the DC number, as that's what really matters. It's just easier to adjust with the DMM, as for boards that have a XCTR/YCTR pot, you can tweak it and get the DC number as close to zero as possible before you connect the monitor. My 2465 doesn't have a DMM function.

I also mention it for pedagogical reasons, for folks who don't have a scope. All you really need is a DMM for checking the DC and AC voltages.
When you have an "asymmetrical" waveform like typical X or Y vector signals of most games, the "DC" value your meter will give you is "average" value of the waveform. This is NOT the true DC offset of the waveform (from the standpoint of setting the offset of a BIPolar DAC output). You might want to use the scope on DC, where it should normally be (IMHO), to verify where you are setting those pots.
 
Try this with the scope:
Put both channels on DC coupling and set up the scope to display (CH1 - CH2) in YT mode.
Measure the same signal and get the difference zeroed out on the scope.
Change only CH2 to AC coupling, which removes the DC component.
Now your difference signal should be the DC component only!

You can use the PCB adjustment to minimize the difference signal on the scope.
And if you really need a number, measure the difference with the cursors.
(But you don't really need a number.)


Why would I do all of that when I can just clip a DMM to the XY test points, and adjust the value to zero?

AC coupling is safer for the scope. Especially when you're working with things that can output large amounts of DC without you knowing. :)

Things should be made as simple as possible, but not simpler.


When you have an "asymmetrical" waveform like typical X or Y vector signals of most games, the "DC" value your meter will give you is "average" value of the waveform. This is NOT the true DC offset of the waveform (from the standpoint of setting the offset of a BIPolar DAC output). You might want to use the scope on DC, where it should normally be (IMHO), to verify where you are setting those pots.


I understand what you're saying about the offset of the BIP. But we're just looking for a ballpark check of the DC component to make sure it's in the range that is safe for the monitor before connecting it. So the average value is what we want here. Any DC component could be coming from whatever the board is drawing at the moment (correct OR incorrect), plus any offset. But what the monitor cares about (in terms of not blowing it up) is the average value. We're not trying to get it perfect, but rather just make sure we don't nuke the monitor, and that we have it in a place to get a decent starting point image on the screen.

The ideal way to optimize the pots is by adjusting the image on the monitor screen (i.e., what they are meant for), which you still need to do once you get everything hooked up.

Looking at a number also gives you a feel for what's 'right', and when a number isn't right, it's easy to spot. For example, when adjusting a Tempest, I want to see the DC value rise up to about 3.5V briefly when the attract mode is drawing the scrolling TEMPEST graphic, then come back down. And I know that when it's in test mode, the DC should be around -1V, because the test mode result screen draws more things on the bottom of the screen than the top. The DMM gives you plenty of information. (And again, I'm trying to provide info here for people that doesn't require them to have a scope.)

Any procedures I describe here were refined over years of working on these boards, in practical conditions. There's a method to my madness. I also worked in high-speed T&M for 20 years, and I currently work at Keysight, so I know scopes. But I'm not going to go full-blown measurement geek here, when my main goal is trying to help people fix their games with the simplest equipment possible.
 
Why would I do all of that when I can just clip a DMM to the XY test points, and adjust the value to zero?

AC coupling is safer for the scope. Especially when you're working with things that can output large amounts of DC without you knowing. :)

Things should be made as simple as possible, but not simpler.
And why would you use 2 tools, when you can make it simpler and use only one? (either a scope or a DMM)
I get your point about people not having a scope, I already said it's a good point.
But you're seriously contradicting yourself here.

As far as protecting the scope...
How the hell are you going to get enough voltage from an Atari vector PCB to blow up the scope?
As you're so fond of saying (and correctly so)....you should check the system components one at a time to make sure everything is working correctly and safely. First block to check is the power supply starting at the mains input. Unless something is crazy wrong with the transformer block or your public utility the highest peak voltage you should see anywhere in the system is about 170V (for a 120V utility) up to about 200V during a 15% surge. But on the PCB you have *much* lower voltages. Even with a 1X probe the scope inputs should handle that no problem.
 
Back
Top Bottom