2600 games work, 7800 games need ~15min warmup OR ditch RF/EMI shield.

[CZroe]

https://youtu.be/XNOQk1nQaEQ

 

Other than a flaky power button which I'm replacing with new parts from Console5, my old Rev B 7800 works fine. I recently got a Rev A 7800 (expansion port also unpopulated but casing has an opening for it). This one only plays 2600 games unless you leave it on for ~15 minutes then cycle the power.

 

After the Atari boot screen 7800 games will just show a blank screen with green lines or scrambled diagonal rows that roll across the screen. During troubleshooting I determined that it will also work perfectly as long as I leave off the lower RF shielding, though I don't see anything it could possibly be shorting out and I don't see how it would stop shorting after a 15min warm-up either. Bizarre.

 

When I connect the bare board everything works. As soon as I place the bottom shield on, 7800 games stop working without the extended warm-up I described earlier. All the capacitors look flawless and this board looks great... far better than my other, working, console. It looked so good inside I felt ashamed to have sullied it with my handprints.

 

Anyone else seen this issue before? What could cause it?

[Jinks]

Very interesting! Does the sheild connect a ground circut for more power draw which somehow is not needed for the 7800 games to run or is it drawing just a tiny bit of power on the hot side and causing a short?

Check with a ohm meter for shorts and voltage drop with and without the shield..

Maybe this is a common fault?

[CZroe]

Very interesting! Does the sheild connect a ground circut for more power draw which somehow is not needed for the 7800 games to run or is it drawing just a tiny bit of power on the hot side and causing a short?

Check with a ohm meter for shorts and voltage drop with and without the shield..

Maybe this is a common fault?

I've definitely heard talk of 7800s that can only play 2600 games, which is how most people would identify this issue since most are not likely to try power cycling after 15+ minutes warming up like I did. That particular issue is sometimes explained by people using an AC adapter without enough amps (the 2600 part isn't as power-hungry as the full 7800).

 

I actually isolated it from the bottom traces with paper all the way around and it still does this. I looked all around and I can't find any place where it would contact anything it shouldn't for creating a short. Everywhere it passes over traces that aren't ground it seems to have strategic cuts that prevent it from even touching the solder mask. I even tried swapping the RF/EMI shield from my Rev B 7800 and it behaves the same.

 

It seems to be proximity-related more than contact-related because I've "floated" every side and it still exhibits this. If I lift a certain side to a certain height it will boot with the other side still in contact, but if I lower the board closer and power cycle the problem returns before any more tabs actually make contact.

 

Carefully placing the top shield in place (requires the bottom plate to hold its alignment), it doesn't seem to have an effect. The system boots with or without it as long as that bottom shield is removed.

 

Not sure where to test for shorts and voltage drops. Should I be testing ohms between the positive power terminal and ground (unplugged, of course)? Though I've already been probing around as if I knew what I was doing, I'm really not that savvy.

 

I tested the 7805 and it's a rock-solid 5.01v with either adapter. Though each of my official AC adapters are spec'd for 9v, one outputs 10.7v and the other outputs 11.05v (12.8v and 13.45v unloaded). I do have continuity between the negative power terminal, the shielding, 7805 heatsink, ground plane, etc.

 

The reason I tried warming it up was because a 2-port 5200 from the same lot was also showing rolling video after the Atari logo until I left it on for a while. Unlike the 7800, I didn't have to power cycle it and it actually started on its own a few minutes later. The game just suddenly booted to the title screen with music but it was in black and white. A few minutes later color returned and it's been 100% ever since. Even when I fully drain the power and wait 30+ hours, it now fires right up and in perfect color. If only the 7800 were as lucky!

Edited May 7, 2017 by CZroe

[bah]

I haven't had my 7800 apart for some time, so I don't remember exactly what sits on what. You may want to set the board down on the bottom shield in the housing and see if it flexes when force is put down on the screw holes. Anything that flexes the board could cause an existng solder crack to open up and be an intermittent type failure. My astrocade has one screw purposely loose for unusual symptoms similar to yours. If I snug up the screw, the picture goes screwy in about 6-10 minutes but it plays for an hour with no problems just by loosening the screw.

[CZroe]

I haven't had my 7800 apart for some time, so I don't remember exactly what sits on what. You may want to set the board down on the bottom shield in the housing and see if it flexes when force is put down on the screw holes. Anything that flexes the board could cause an existng solder crack to open up and be an intermittent type failure. My astrocade has one screw purposely loose for unusual symptoms similar to yours. If I snug up the screw, the picture goes screwy in about 6-10 minutes but it plays for an hour with no problems just by loosening the screw.

It doesn't seem to be from the contact itself since it does it as soon as it gets within a certain proximity and before there is actual contact. My guess is that it is resonating with something else and causing interference.

 

I recall a Verizon LG G2 that would not boot if I tightened one particular screw.

[bah]

Well, that is strange as I don't remember anything on the solder side that would be affected by a large metal plane.

 

I would probably just leave the rf shield off, but it would be interesting to see if a smaller piece of sheet metal could be placed and moved around under the board to produce the same effects. IE narrow down the problem area.

 

Good luck.

[DrSidneyZweibel]

Interesting problem to tinker with, but the easiest answer here is just leave the entire RF shield off. You don t need it. The FCC mandated them for small electronics because they might interfere with TVs that still used antennas and analog over the air signals. Those are no longer in use. The RF shield serves no other functional purpose. It is completely unnecessary in 2017.

[+-^CrossBow^-]

However, the bottom part of the RF shield is sorta needed to keep the 7800 sitting in the right place in the bottom half of the shell and to provide the right height for it to sit. So might need to get some small thin washers to affix to the board or something if you decide to remove the RF shielding completely.