Capacitor c55 issue

[Rastamafugg]

1 minute ago, -^CrossBow^- said:

Well it would only cause an issue if it were in a permanent PAUSE ON type position.

 

Do you have an older 2600 game like Pitfall or combat you can put in and see if it comes up in Black and White?

 

Also there is usually tape just under the cartridge sleeve as I believe it was originally there to hold the cart slot and space in place during the initial soldering. But I've never seen red before as it is always masking tape white or beige.

 

I have Combat, and I'm pretty sure it was in color, but I can test it again.

[+-^CrossBow^-]

1 minute ago, Rastamafugg said:

I have Combat, and I'm pretty sure it was in color, but I can test it again.

If it does come up in color that is fine but it just means that you don't have the ability to pause any 7800 games that use it or some 2600 'brews that were programmed to use that switch for pause.

 

I think the main issue here is that the input voltage is marginal and just enough to power and switch on for 2600 games but not for 7800. 

 

[Rastamafugg]

Here's another symptom... At some point during testing, the 7800 stops powering on.  If I leave it for some time (hours at least), it start powering on again.  Don't understand why that is happening.

[+-^CrossBow^-]

Just now, Rastamafugg said:

Here's another symptom... At some point during testing, the 7800 stops powering on.  If I leave it for some time (hours at least), it start powering on again.  Don't understand why that is happening.

This all points to power to me as being the main issue here.

 

So here is what you do:

 

When it is powering on, measure the input voltage like I stated before. Black probe on ground somewhere on the board and Red on the ferrite bead. What is the value?

When it isn't power on, measure again to see what it reads?

 

 

[Rastamafugg]

6 hours ago, -^CrossBow^- said:

This all points to power to me as being the main issue here.

 

So here is what you do:

 

When it is powering on, measure the input voltage like I stated before. Black probe on ground somewhere on the board and Red on the ferrite bead. What is the value?

When it isn't power on, measure again to see what it reads?

 

 

Measured the end of the ferrite bead closest to the big cap...

Powered off: 2.37v

Powered on (9v dc via a brand new power supply through my Dragonfly cart using the pass-through power connector): ~8.5v

Powered on with the original 7800 power supply: ~10.2v

[+-^CrossBow^-]

Okay, so now you brought up the DF cart. Have you been trying to use it the entire time for 7800 games?

 

The 8.5 might be accurate but I thought it was a straight passthrough so should still have the same volts on the output cable as it does going into it.

 

So you show 10.2v powered on using an original 7800 PSU? What is the reading with it powered off at the ferrite bead? It should be basically the same as all of that is before the flip-flop actually powers on the system. 

 

And with the original PSU what the input voltage on that right most leg of the 7805 as your are facing the rear of the mainboard? That really should be about 9v.

 

[Rastamafugg]

1 hour ago, -^CrossBow^- said:

Okay, so now you brought up the DF cart. Have you been trying to use it the entire time for 7800 games?

 

The 8.5 might be accurate but I thought it was a straight passthrough so should still have the same volts on the output cable as it does going into it.

 

So you show 10.2v powered on using an original 7800 PSU? What is the reading with it powered off at the ferrite bead? It should be basically the same as all of that is before the flip-flop actually powers on the system. 

 

And with the original PSU what the input voltage on that right most leg of the 7805 as your are facing the rear of the mainboard? That really should be about 9v.

 

Ok, didn't understand you wanted me to test with the adapter connected but 7800 powered off.  Here are the readings from that bead:

12.75v original psu off

8.94v dragonfly pass-through off
 

She doesn't want to start again, so any powered tests will have to wait until morning.

[Rastamafugg]

 

On 10/22/2021 at 8:47 PM, -^CrossBow^- said:

Okay, so now you brought up the DF cart. Have you been trying to use it the entire time for 7800 games?

 

The 8.5 might be accurate but I thought it was a straight passthrough so should still have the same volts on the output cable as it does going into it.

 

So you show 10.2v powered on using an original 7800 PSU? What is the reading with it powered off at the ferrite bead? It should be basically the same as all of that is before the flip-flop actually powers on the system. 

 

And with the original PSU what the input voltage on that right most leg of the 7805 as your are facing the rear of the mainboard? That really should be about 9v.

 

So I read with the original PSU:

At the bead powered off: 12.4-12.7v

At the bead powered on: 10.4v

At the right most leg of the 7805: 10v

 

It won't stay powered on long, though, with the original PSU.  It powered off while testing and won't turn on again.

Edited October 24, 2021 by Rastamafugg

[+-^CrossBow^-]

At this point I'm inclined to think the issue is either with the 7805 as it might not be providing enough current for the 7800 side to kick in and work, or the NT185 transistor might be sketchy as I know that wasn't original. If you had a bench power supply and were able to provide exact +5 to the system main logic with at least 1A of current, I'd be curious to see if started working at that point.

 

[+Stephen Moss]

On 10/22/2021 at 4:24 PM, -^CrossBow^- said:

I think the main issue here is that the input voltage is marginal and just enough to power and switch on for 2600 games but not for 7800. 

Interesting hypothesis. I agree that the input voltage seems a little low, but not excessively, I wonder what the spec of the power unit is if not an original Atari 7800 PSU.

As far as I can see everything runs off 5V, I don't know how much extra current is consumed in 7800 mode compared to 2600 but even if the PSU is a little under rated current wise although the ability to regulate the voltage would be diminished even with 5V in I would expect 5V out of the regulator.   

On 10/22/2021 at 5:43 PM, Rastamafugg said:

Here's another symptom... At some point during testing, the 7800 stops powering on.  If I leave it for some time (hours at least), it start powering on again.  Don't understand why that is happening.

That sounds like either a thermal fuse in the PSU has tripped or the voltage regulator has gone into thermal shutdown which would be worrying considering the size of the heatsink, but it could just be a bad solder joint loosing connection with thermal expansion.

If it is a thermal shutdown that could indicate that something may be shorted/drawing too much current, the additional current draw could also explain the previously reported lower than expected input voltage as it drags the PSU down. Although the voltages in post #58 seem more normal. 

If you can get hold of some freezer spay, if anything other than the regulator is drawing an excessive amount of current it would defrost much quicker than anything else. 

10 hours ago, -^CrossBow^- said:

If you had a bench power supply and were able to provide exact +5 to the system main logic with at least 1A of current, I'd be curious to see if started working at that point.

That would be a good test if they can do it, plus the current reading on the PSU might indicate if something is creating an excessive current draw.

I am not sure what the dragon fly passthrough is but if it can be easily removed it may also be worth doing so to eliminate that as a possible cause.

 

[+-^CrossBow^-]

The DF can be removed from the console and left turned 'on' so that it still provides power to the 7800. This is essentially what I do when I want to plug in a different cartridge but still leave the DF cart nearby. The 7800 on its own draws about 720mA of current when playing a pokey game like ball blazer or commando. So my thought is that a 500mA output 7805 might have been installed and it isn't able to provide enough for the 7800 side to operate. Now if you replace the original linear regulator with a DC-DC switching type as I've done in quite a few of my consoles, then that current draw in 7800 mode actually becomes more like the 2600 in that it only need about 320 - 350mA of current to operate.

 

@Rastamafugg to summize:

 

2600 games will power on and play normally?

7800 games will sometimes provide you with the rainbow startup logo but never actually get any further than that?

When it does play 2600 games it does so only for a time and then shuts off and does not power on further until left alone?

How long do you have to leave it powered off before it will power on and play 2600 games again?

You get the same results both with a dedicated PSU to the 7800 and also through the DF cart correct?

 

[Rastamafugg]

6 hours ago, Stephen Moss said:

Interesting hypothesis. I agree that the input voltage seems a little low, but not excessively, I wonder what the spec of the power unit is if not an original Atari 7800 PSU.

As far as I can see everything runs off 5V, I don't know how much extra current is consumed in 7800 mode compared to 2600 but even if the PSU is a little under rated current wise although the ability to regulate the voltage would be diminished even with 5V in I would expect 5V out of the regulator.   

That sounds like either a thermal fuse in the PSU has tripped or the voltage regulator has gone into thermal shutdown which would be worrying considering the size of the heatsink, but it could just be a bad solder joint loosing connection with thermal expansion.

If it is a thermal shutdown that could indicate that something may be shorted/drawing too much current, the additional current draw could also explain the previously reported lower than expected input voltage as it drags the PSU down. Although the voltages in post #58 seem more normal. 

If you can get hold of some freezer spay, if anything other than the regulator is drawing an excessive amount of current it would defrost much quicker than anything else. 

That would be a good test if they can do it, plus the current reading on the PSU might indicate if something is creating an excessive current draw.

I am not sure what the dragon fly passthrough is but if it can be easily removed it may also be worth doing so to eliminate that as a possible cause.

 

The last set of voltages I posted were with the original Atari PSU.  The PSU I've been using with the Dragonfly is a Berls adjustable PSU set to 9v DC.  When the 7800 stops powering on, it doesn't matter which power supply I use, the light doesn't come on.

What is freezer spray?  Is there a particular product I should be looking for?

As for a bench power supply, can anyone recommend a good one for retro computer testing?

[+-^CrossBow^-]

Well I'm not sure if there is a specific spray for the purpose, but when I was doing bench work on older cash registers in the late 90s, we actually used canned air turned upside down to apply a nice frost to IC chips to check for those with internal shorts heating up much to quickly and also it came in handy for stunning mice we would find inside the registers when servicing local grocers so we could remove them.

 

My bench supply is old. It is an old B&K precision I picked up off shop goodwill like 2 years or so ago. Only goes up to about 30v (25v in reality according to my fluke), but can provide up to 3A of courrent which is enough for pretty much anything I plan to work on.

 

The Triad branded PSU I'm using with my DF actually outputs 9.04v from the passthrough end when just used to power the 7800. And it doesn't seem to drop at all when connected to the 7800.

[Rastamafugg]

1 hour ago, -^CrossBow^- said:

also it came in handy for stunning mice we would find inside the registers when servicing local grocers so we could remove them.

?

[+Stephen Moss]

15 hours ago, Rastamafugg said:

What is freezer spray?  Is there a particular product I should be looking for?

Several companies produce freezer spray, you want to be looking for something like this. Basically you spray it all over the PCB (or a particular spot in you know roughly where the problem is) for about 20 second and it cools the components so that they become covered in "frost".

After switch on the components warm up and the frost on them disappears, where it disappears on one component much faster than the others that could indicate a potentially faulty component (depending on it function in the circuit) as the more current it uses the faster it will warm up. You could try just touching components instead but risk a burn.

 

Some people say place it in bag in the freezer to achieve a similar result but I personally would not do so as I feel here is more risk of condensation as it heats up resulting in a short doing it that way then when using freezer spray.

 

@Rastamafugg what is the current rating on the Berls adjustable PSU set to 9v DC, the lower input voltage with that may be because it cannot provide enough.

 

 

[+-^CrossBow^-]

19 hours ago, Stephen Moss said:

 

@Rastamafugg what is the current rating on the Berls adjustable PSU set to 9v DC, the lower input voltage with that may be because it cannot provide enough.

 

 

The only issue with this, is that he is using that with his Dragon fly cartridge I believe and the creator of the DF cart was very clear about not providing more than 9v to it.

 

But @Rastamafugg also stated it does this with his original 7800 PSU and that seems to be providing enough input voltage at over 12v so that again points to something internally in the 7800 bringing things down.

 

Rasta: using the original 7800 PSU plugged in, what is the voltage reading off the leftmost leg (As you are facing the rear of the mainboard) when it is powered on with a 2600 game in it because that should be at 5v. Then power it off quickly and plug in a 7800 game and power it on and get another reading. 

 

 

[Rastamafugg]

3 hours ago, -^CrossBow^- said:

The only issue with this, is that he is using that with his Dragon fly cartridge I believe and the creator of the DF cart was very clear about not providing more than 9v to it.

 

But @Rastamafugg also stated it does this with his original 7800 PSU and that seems to be providing enough input voltage at over 12v so that again points to something internally in the 7800 bringing things down.

 

Rasta: using the original 7800 PSU plugged in, what is the voltage reading off the leftmost leg (As you are facing the rear of the mainboard) when it is powered on with a 2600 game in it because that should be at 9v. Then power it off quickly and plug in a 7800 game and power it on and get another reading. 

 

 

Now it's not powering on at all. ? I'm wondering if there is a cold solder joint or something only loosely connected.  Going to have to go over the board with a magnifying glass, I think.

I have a bench power supply arriving on Friday.  Providing I solve this issue, where do I connect on the board to provide +5V to the system main logic and bypass the power circuit (as you mentioned in a previous post)?

[ChildOfCv]

When you first apply power:

 

While you hold the power button:

 

Release the power button:

*Increasing to 5

 

See how much of this matches your readings.

Edited October 27, 2021 by ChildOfCv

[+-^CrossBow^-]

3 hours ago, Rastamafugg said:

I have a bench power supply arriving on Friday.  Providing I solve this issue, where do I connect on the board to provide +5V to the system main logic and bypass the power circuit (as you mentioned in a previous post)?

I had to make a quick correction to what I said above. You should show +5 off that left leg of the voltage regulator with the back of the system board facing you. 

 

To answer you other question, anywhere along the fat trace down the middle I have with a Red line. There are several unused vias along that trace you can solder a clipped lead or piece of wire to. That is one of the 2 main +5 rails through the system. The other large trace to the right of it is GND.