5v issue powering SNES - help needed please!

[bacteria]

I have an issue with my current project, in the SNES section; only had one reply to a plea for help; as this is a technical issue, posting query here in case anyone can resolve:

Situation:

Using full sized PAL SNES version 1 mobo. System normally takes in 9v AC, converts it to 12v DC then runs into the 7805 to convert to 5v DC.

Rigging a 7.4v battery and wiring directly to the 7805 gives a constant 4.99v, which runs the SNES fine. Problem is that after 4-30 minutes the system shuts down. The 7805 gets quite warm, however attaching it to steel with a fan blowing directly on it to keep it cool has no effect (I did this in case the 7805 cut itself off if it got hot). When the system cuts out, I still got 4.99v/4.98v (fluctuating) so no dramatic loss of voltage.

Doesn't make any difference if I use the original 2200uf capacitor, a newer 2200uf capacitor or 1000uf capacitor - same result.

Interestingly, the on/off system button doesn't work when connecting batteries directly to the 7805; also the capacitor doesn't keep enough juice in it to mean you have to turn the system on without power to discharge it when running on batteries.

Everything else works great. I assume the 12v is generated if RGB output to a Scart; after all the system is able to run entirely off 5v (off the 7805), all be it for 4-30 mins at a time; if the 12v was used elsewhere then surely the system wouldn't work at all. Pin 3 of the video output pins (PAL system, not NTSC) outputs this voltage (or 5v if wiring straight to the 7805).

Nothing wrong with my screen/control/battery section - ran it 6 hours straight with the N64 system in, so nothing wrong there.

Hooked up a TI card with a small resistor, about 32R, which gave 5.04v; I removed the 7805 and wired the TI card to the output (5v line) - nothing happened - perhaps the 7805 needs to be in place even if not working (need to try this). Strange the 5.04v didn't otherwise boot the system, maintained 5.04v; after all, using a TI card on an N64 with expansion pack needs about 1.3amps or so, and the SNES quotes 1.3amps on the back of the system, so I guess this is fine.

Driving me nuts. System works great, everything works, looks nice, just need to solve this cut-out problem.

I can open up other SNES's I have, run them off batteries and see if they work for a couple of hours plus and use that board in my project (assuming one works for a long time), but anyone got the solution to offer me before I do this (after re-attaching the 7805 and the TI card to it's output at 5.04v).

PS - the 7805 on the SNES is a 17805 and another one I tried (same result) was a DE7805; if this helps.

Would be grateful for feedback, comments and suggestions please!

[Skyone]

Interestingly, the on/off system button doesn't work when connecting batteries directly to the 7805; also the capacitor doesn't keep enough juice in it to mean you have to turn the system on without power to discharge it when running on batteries.

Of course. There would be a quiescent current let through if the switch didn't control the voltage going into the regulator.

[timmeh87]

Did you take a heatsink off anything on the SNES?

Is the 7805 in good thermal contact with its heatsink (thermal compound).

I mean, it really sounds like some kind of heat issue.

[bacteria]

On the original SNES there were no heatinks, only two metal plates.

Ignore the arrow in pic, using an old pic; however the 7805 is just to the right of the arrow:



The metal place on the right is a few millimeters over the various chips (quite a few of them), although none of the chips have any heatsinks; the one on the left is over the remaining chips, slightly higher than the other one and has the 7805 screwed to it, to keep it cool.

I put a heatsink (quite a large one) onto the 7805, it got quite warm before the system died each time which is why I thought of thermal cut-out; so I put a fan over it; same result. I then screwed the 7805 to some of the metal plate and had a fan blasting over the plate, same result. Only thing I didn't do was use thermal paste, I see where you are coming from; I will get some and try it; and put heatsinks on the front and back of the 7805 this time. I hope that works, if it doesn't work, I can't see any other option but to start again with a new SNES. Won't take me more than 2-3 hours work this time as I know what to do.

BTW - what is the max amp output of a PTH8000 TI card?? (wondering why it didn't boot up my system when I tried it instead of the 7805, at 5.04v).

[Ben Cebhrem]

2.25A for a pth08000. I presume that's what ya have of course.

[GonzoMPM-1]

Did you cut the traces leading back from the 7805 to the original 9v -> 12v etc., power components on the SNES board?

Is it possible you've got back current flowing because of the way you're introducing power and its creating a resonance in those, now vestigial, circuits that, say on charge of a cap to capacity, or the breakdown of a P/N junction from revers voltage creates an interment forward surge that the triggers protection in the 7805 itself or further on down the line?

A back-current drain from somewhere in those original circuits would explain the fact you're not having to manually discharge that capacitor, for example.

Extra / remnant circuits can cause trouble and I've come to try and isolate them away when they're not going to be used any longer.

I know its a random thought, and I'm not articulating it very well, but just an idea.

[bacteria]

Doesn't look like the issue was with heat; rather a faulty board. I have given a bigger explanation in the SNES section of this forum. Should have faulty board replaced and re-wired etc Monday morning.