Need help troubleshooting my 5200

[turbofly]

Hi all,

 

I've been browsing the Atariage forums for a long time now and with help from a lot of great posts I've modded and resurrected quite a few Atari systems. Thank you, to all of the knowledgeable and friendly people who post here.

 

I finally came up against some issues that I haven't been able to sort out by searching the forums. I'll start out by telling everybody about the system I have. It is a 4 port 5200, (the non VCS adapter compatible one). I bought it untested from ebay with 3 games and 4 controllers, but no power supply or switchbox. I bought an original 4 port power supply and did the power mod to eliminate the need for the all in one switchbox, (thank you Mitch and CPUWIZ). I am using the original soldered on RF cable to connect the 5200 to my TV, (I'll likely do the video mod after I sort out the issues I'm having). I've cleaned up 2 of the controllers and the buttons are working good. The joysticks were too, but that's changed. Anyhow after I completed the power mod and hooked the 5200 to my TV it looked and sounded great, no problems with any of the games I tried. However some of the controller inputs weren't reading, but the system seemed to be working great otherwise. After a while thought a new problem started. The system would not turn off with the power button and I started getting graphical errors with lockups. The power is on as soon as I plug it in, without even pressing the power button.

 

After reading a few posts here it seemed that I my power problem and graphics troubles were likely due to a bad 4013 flip flop. I bought a pack of 25 and have tried about half of them - no change, the problems remain. As a matter of fact even without a 4013 socketed the 5200 powers on as soon as I plug it in and will on turn off. I also still get graphical corruption and lockups after a short time too. Aside from changing out the 4013 I've also tested the power switch for continuity. It has continuity when the button is pushed and none when it isn't, so the button isn't the problem. This has me really stumped, any suggestions would be greatly appreciated.

 

On to problem 2 - the controller input issue. As mentioned I've cleaned and got working 2 of the controllers. So after some searching of the posts it seemed likely that my problem was due to faulty CD4052 ICs. I bought a pack of 25 and replaced the 5 that are on the board. Success! All of the buttons now work as they should. However I am having trouble with the joysticks. The joysticks worked fine initially, before I changed the CD4052s, but started to give me trouble before I did the swap and it didn't go away with the swap. As soon as I start any game it acts like the joystick is being pushed all the way down and to the right. This happens even if the joystick is centered and does not change even if I move the stick all the way up and to the left, or any other direction for that matter. It's the same with both controllers and the same with both ports 1 and 2, I have not tried ports 3 or 4. More reading on the forums is leading me to suspect I've got a fault Pokey chip. The Pokey looks slightly discolored around the edges. I don't have another to swap out, yet, but that is what I plan on trying next. I've also cleaned the pins in all 4 or the ports. Again any suggestions would be greatly appreciated.

 

I do own a multi-meter and a decent soldering iron, (which I am pretty good at using). I also got really lucky with the board revision in my 5200, every IC on the board is socketed, so no de-soldering required to swap out chips.

 

This is the first 5200 I've owned and I've got a lot of great games that I want to start playing. I know some of you guys have seen it all with these systems, so I'm hopeful that I'll be able to sort it out with your help. Thanks in advance.

[turbofly]

Ok time for an update. I hope this helps someone else who is having this trouble.

 

This afternoon I called Best Electronics, I spoke with Brad and told him what was happening. At first he thought the same thing that I did that I might have a bad Pokey, but as we talked he suggested that I check the 5 transistors located at the bottom right corner of the board. He told me that sometimes when people are taking the board out or putting it into the case they might inadvertently hit those and knock a leg off one. He told me that these are part of the joystick circuit and that it could cause the issue I am having and seem like a Pokey going bad. After I got off the phone I had a look at my board and sure enough I had broken one of the legs of one of the transistors from the board. A dab of solder later and my joystick issue was solved.

 

Brad was phenomenal to talk with he really knows his stuff and is very patient. Big thank you to him.

 

He also gave me some advise on where to look next for my power issue. He suggested I test the 2 transistors just below the RF box, (these are labelled MJE210). That's my next step, but just in case I would love to hear any ideas anyone hear on the forum might have, any suggestions will be gratefully received. Of course if I get it fixed I will post a follow up in hopes that it helps someone else.

[turbofly]

Update #2.

 

I tested the 2 MJE210 transistors and one was bad. I've replaced it, and no change. I then went back to the 4013 flip flop and noticed that the socket the chip sits in looked pretty badly oxidized, so I cleaned it up, still no change. In all honesty I didn't expect it would help as the 5200 powers on as soon as I plug it in even without a 4013 being seated in the socket. It seems that another component is bad and is sending power to the system past the 4013 somehow. I could live with a 5200 that is stuck on, if not for the graphical corruption and freeze ups that happen shortly after starting to play a game.

 

Anyway, I see that almost 100 of you have read my posts, and I know from reading other posts that the people here really know the 5200. I'm out of ideas, if anybody has any suggestions I'm all ears and would appreciate the help. Thanks.

[+Mitch]

Have you checked the 5200 service manual?

 

Mitch

[turbofly]

Hello Mitch and thanks for the reply.

 

I have a manual called the CX 5200 Field Service Manual, if that's the same service manual you mean, then yes I've had a look through it. Nothing there seems to cover my power issue. The closest thing I can find is in the section for the 4 port Diagnostic Flowchart that is for Warped Ragged Picture Troubleshooting. It kind of fits as I do get graphical corruption after a short while of game play. I read the flowchart but haven't yet done the testing there. Looks like this will be my next step.

 

If you are talking about another manual please let me know where to find it.

 

Thanks.

[Henry Lee]

Did you try this:

 

[turbofly]

Hello Henry Lee,

 

Thanks for the reply. I've already managed to fix my controller issue with some help from Brad at Best Electronics. He had correctly guessed that I had a damaged transistor on the lower right hand corner of the mainboard. Just below the calibration potentiometer discussed in the video you linked. All I have left to diagnose and repair is what ever is causing my power issues, so if you have any ideas about that keep 'em coming.

[zylon]

Try flexing the board slightly while powered on. Some of these had an issue with a solder pad being near the trace for the upper rightmost pin for the cart connector. The solder on the pad can be sloppy and if the board warped when you removed it, it can short over to the address line.

[turbofly]

Hi zylon,

 

Thanks for the reply. I had a look at that pad, on both sides of the board and it looks pretty clean on my board, and doesn't appear to be contacting the trace. I powered the 5200 up and tried flexing the board, it didn't help with the always on power issue, and it seemed to shorten the game play time before graphical corruption. Worth a try though, keep the ideas coming please.

 

The thing about this system is that I bought it from Ebay not working and without the adaptor or automatic switch box. I soon found an adaptor and then decided to do the power mod and eliminate the need for the switch box. After that I cleaned the case, the board and the cartridge slot, I plugged in a game and it fired up no problem. The power button would turn the system off and on like it should. Then I was having controller issues so I started plugging and unplugging the controller, (I thought maybe the port pins or the plug might be oxidized and this might scrap them clean). At this point I noticed that the power light was flickering and shortly thereafter the power was on continuously. I thought that the 4013 must have went bad and replaced it and it made no difference. That's when I reached out to Brad at Best Electronics and you guys on the forum. So I don't know if the system had this issue previously, or if I inadvertently did something to it when I did the power mod or when I was plugging and unplugging the controller. What puzzles me the most is how the system can power on even without a 4013 slotted. I think what ever is causing that is the root of the problem.

[zylon]

Somehow, you must have a short letting power from before the 4013 to the circuits after it. Take a good look at the bottom side as you may have bent something just right to do this when you were working on the mod or RF box. I've seen this only once before. I'm thinking that after the system runs a while, the "bridged" point heats up and fails a little. This could be your scrambled graphics. Since you said flexing the board made that occur faster. Another possibility is a dent in the bottom side of the metal shielding.

[turbofly]

Hey xylon,

 

I've been super busy lately and haven't had a chance to get to the 5200. I had a really good look at both sides of the board last night with the shielding off and my illuminated magnifier on. I didn't find anything that is touching that shouldn't be. I also tried using the 5200 without the shielding to see if it was touching that somewhere. There was no improvement. So far I've only done the power mod and haven't touched the RF box on the board. I was thinking that maybe the momentary switch, (used for the power button), may have failed in a way that left it in an always connected state. However I've since checked that with my multi-meter and that's not the case. I'm wondering if one the capacitors I used to do the power mod might be the problem. When I went to the electronics store they didn't have the ceramic capacitor that I needed in stock to buy new. But they did have old one there that they gave me. (It's labelled 104 and below that Y5T, I don't know what the Y5T means). I know that capacitor is there as a filter, if it is junk could that be causing the problem? Thanks.

[+-^CrossBow^-]

My understanding is that the 4013 IC acts as a flip flop in regards to the power switch either being on or off correct? I know that a 7800 I worked on about a year ago had the same issue in that it would power on immediately when I plugged it in. I removed the 4013 off it and it still powered on. It ended up being two things, one was a small capacitor next to the IC and the other was a bad MJE201 transistor. I yanked one of the two present on a donor 5200 board I had laying around and that fixed up the on/off issue on the 7800 and it worked normally after that. So..might need to look around the IC and where the power goes to see if anything that looks like a diode might be cooked. I said looks like a diode because Atari had a habit of making tiny capacitors that looked like diodes on their boards. If not, you might see about replacing one or both of the MJE201s that sit side by side under the RF modulator.

[turbofly]

Hi CrossBow,

 

I had already replaced one of the MJE201 transistors, and after reading your post I decided to replace the other. I had an extra as I bought 2 of them when I replaced the first one. There's no change. I then started looking around for these diode looking capacitors you mentioned. I found a number of black tube shaped components on the board, there is no line like what you'd find on a diode but one end shows what looks like a white epoxy where the lead comes out. L12 is silkscreened on the board beside one of them, and since most of the capacitors have a C and a number, I'm not sure if those are what you are talking about. None of those appear to be damaged. Like to 7800 you described, my 5200 has power whether the 4013 is in the socket or not. The power light comes on and kinda flickers but the screen stays black. With the 4013 socketed if I flex the board near it I get graphical corruption. If I move it just right I can actually play a game for a little while. I wonder if there anything under the socket that is shorting, that shouldn't be. I'm trying to decide if it is worth de-soldering the socket and looking underneath it.

 

Do you have a picture of what the diode looking capacitors look like? Also how did you determine that it was bad, was it puffed out or blackened, or did you check it with a multi-meter? If you checked it with a multi-meter, where you able to do so in the circuit or did you have to de-solder it to check it? Thanks.

Edited February 21, 2016 by turbofly

[zylon]

I'm thinking you might need to replace the socket as you said earlier it was corroded inside. Another possibility is a crack on a trace somewhere near that location. Some people get a little over-zealous when prying the board up off the locating pins.

[+-^CrossBow^-]

At this point, since you mention that you get different results while flexing the board, this proves there are traces that are weak or possibly broken. Since you have the service manual, you will need to look at the schematics and just try checking for continuity with your meter between various points. As for the capacitor that looked like a diode issue. That was on my 7800 and I found it because it was just above where the MJE201 went in. It was black, cracked, and literally broke apart when I touched it. I looked up what it was and found out it was actually a capacitor even though it was clear like a diode. I found an exact match once again on my 5200 donor board and replaced it.

 

I would also suggest just pulling the chips, especially where the GTIA and Pokey are, then desolder the sockets and replace them. Before you put the new ones on, take a look at the traces that were hidden by the sockets for any possible cracks or damage. Broken traces can be repaired pretty easily just using small wire, staples, paper clips, or the extra from snipped off leads from other components to bridge the breaks.

 

I'm not in the same camp as most, but at this point with all the work you have already done, I would have saved it as a parts unit and found another...LOL. But that is mainly due to me not being nearly as knowledgeable or perhaps as persistent as others.

[turbofly]

Hey guys,

 

Just before the last 2 replies I got another 5200. I was actually trying to get my hands on an XEGS. I found and won an auction on ebay that included an XEGS, a 7800, a 4 port 5200, and a 2600 jr. All untested and with out cords. I found another auction for a new XEGS power supply, and when I got the packages the XEGS worked perfectly. The 7800 did not. I new tact power switch and cart slot cleaning and I now have a second working 7800. The 2600 jr had a bad power switch too, I took the switch apart and cleaned the contacts and was back in business. As for the 5200 I cleaned it up and prepped it for the power mod. I was lucky and enough to get a second Sunny Vale unit with all of the ICs socketed. I got the packages late Friday and so I have to wait until Monday for the electronics part store to open to get the stuff I need for the power mod.

 

After I had the other 3 systems up and running and the new 5200 ready for the power mod, I looked and saw that zylon and CrossBow had posted some new replies to this post. At this point I figured that I would be taking CrossBow's advice and moving on to another 5200, and maybe continuing to work on the other system as I had time. (I hate to leave things once I've started to repair them, I need to see them through. A little OCD perhaps).

 

Waiting for Monday and the OCD got the better of me and I decided to pull the flip flop socket, (on the problem 5200), and look for broken traces of some other issue. when I got the socket off I couldn't see any glaring problems. I drew a diagram of the traces under the socket, thinking I would solder on new wire connections on the bottom of the board, (in case the traces were bad and I just couldn't see it). I then re-installed the socket, and slotted a flip flop. I starred at it for a minute wishing I knew what the F was wrong with it, and then for the hell of it I plugged it in. The power light did not come on. I pushed the power button and the system started up and when I pushed the button again it actually powered off. A few more tries and the 5200 continued to function like it is supposed to. I played some River Raid and Moon Patrol, no issues. I put the system back in the RF shielding and case, and it's still working great.

 

All I can conclude is that I must have had a cold joint between the socket and the board, (I probably broke it taking the board out of the case, like zylon said "over-zealous when prying the board up off the locating pins"). The funny thing is that I had previously reflowed all of the pins on the socket while troubleshooting, and it didn't help.

 

Anyway, thanks for the help to all those who replied. It looks like I have some video modding to do now, (gotta ditch the RF).

[+-^CrossBow^-]

Excellent! Glad to hear you got it working. Always best to have two working spares vs one and extra parts. The video mod is pretty easy as long as you order the kit from Electronicsentimentalities. Has everything you need pretty much and of all the AV mods I have done, the 5200 s-video is by FAR the best looking of the bunch.

[jmbrissie]

Did you try this:

 

 

Thanks for the post. That's exactly the problem I am having. I will give it a try tonight and hopefully I can start playing and enjoying breakout soon. It will probably help the other games that were acting strangely as well.