All my 130XEs (3) getting bad video suddenly

[Bee]

I have a new Sony 55 inch flat screen TV using Composite Video in.  When it came in December 2020 my 130EX on a Lotherk composite cable was Blue.  When I booted it the other day I noticed the Blue is decidedly Pail Green now.  I'm seeing the same thing on 2 other 130XEs.  2 out of the 3 are having issues with the option key.  One of these is about 16 months out of Best NOS Mianboard.  I have a possible keyboard fix from another thread, this is mainly regarding the shift in video.  On my XEGS, same screen, on a double shielded composite cable it's blue as can be like a new unit out of the box back in the day.     

 

Any ideas?  Would A UAV fix  it this or is it a sign these are going south?

 

Thank you

[+Nezgar]

Check your power supply voltage - measure it loaded from the SIO or joystick pins too. Undervolting may also be related to marginal function keys...

 

Too far below 5.0V will cause issues.

 

[Bee]

I'm using a Brewing Academy USB power cable on a 1A usb powers supply.  I will take a look at 5v on the ISO. I should mention the same composite cable on a XL with the same power source is Blue background not Green.

 

Thank you

[TGB1718]

Personally I would use a power supply with at least 1.5A, although the user manual says 1A my Atari PSU 

is 5V 1.5 A (even if it is the size of a brick  )

[ivop]

About Lotharek's cable: https://atariage.com/forums/topic/310402-quick-review-of-lotharek-hq-video-cable/

 

But a (sudden) shift in colors cannot be blamed on the cable. Have you checked the Color Adjustment pots?

[Bee]

I have not looked at the color pots, to be honest I didn't know they were there.   I will now.

 

Thank you

[Mathy]

Hello guys

 

When I had the screen go green on one of my computers, the old Atari power supply was outputting a lot MORE than 5 VDC.  Not sure if switching power supplies will do that too.

 

Sincerely

 

Mathy

 

 

[Bee]

I still have to pop the hood and check the pot but SIO shows 4.653V from FujiNet.  Same as my XEGS.  I will find an original power bick and try it as well.

 

Thank you 

[ColleenLover]

Sherlock Homes once said, "Once you have eliminated the impossible, whatever is left must be the truth"..   

 

The odds of all 3 of your machines doing this at the same time are slim at best.

 

What else do they have in common?

 

a) the PS ... Try a genuine Atari PS.

b) the composite cable itself?  (have you tried --them on monitor-svideo yet?)

c you said the TV was new?????  could it be the TV????   composite is an afterthought today by the manufacturers.

 

 

Edited April 11, 2021 by ColleenLover
spelling lol

[Bee]

OK the question is are we learning something or am I learning something?  I have found really interesting results.

 

USB cable on a 1A power charger = SIO 4.653V pail green background jail bars

 

Atari Brick 1A power supply = SIO 4.757V much more blue with jail bars

 

Atari Brick 1.5A power supply = SIO 4.757V Blue Blue no jail bars ( 2nd one still has them but can't see them unless I'm right on top of it so Improved greatly, 3rd one is somewhere between 1 & 2)

 

So it's not just the voltage alone.  I want to thank you all for your input.  I wouldn't have gotten here without it.

 

Thank you

Edited April 11, 2021 by Bee

[+Nezgar]

5 hours ago, Bee said:

I still have to pop the hood and check the pot but SIO shows 4.653V from FujiNet.

4.7V is barely enough. That's low enough to make SIDE2 not read CF cards. I wonder if measuring from the Fujinet is giving you a lower measurement that from the SIO or joystick port directly. Try measuring between pin 7 & 8 (careful not to short them together )

 

[StickJock]

 

1 hour ago, Nezgar said:

Try measuring between pin 7 & 8 (careful not to short them together )

 

Yeah, who approved of this design with the power & ground pins right next to each other?  They should have at least swapped pins 6 & 7, and maybe 5 & 8 as well.

[+DrVenkman]

7 hours ago, Nezgar said:

4.7V is barely enough. That's low enough to make SIDE2 not read CF cards. I wonder if measuring from the Fujinet is giving you a lower measurement that from the SIO or joystick port directly. Try measuring between pin 7 & 8 (careful not to short them together )

 

The FujiNet measurement may well be incorrect (too low). See this post and following for technical discussion. 

[+x=usr(1536)]

14 hours ago, Bee said:

I still have to pop the hood and check the pot but SIO shows 4.653V from FujiNet.  Same as my XEGS.  I will find an original power bick and try it as well.

Don't use FujiNet for the measurements; it can only tell you what it thinks it's seeing at the SIO port.  It's also not 100% accurate.  Fine for a rough idea of what might be going on where it's plugged in, but not really intended for use beyond that.

 

Measure voltage at the power adapter.  With the 8bitclassics / TBA power cables, it's a lot easier to do this at the DIN connector on the cable.

 

As recommended, swap out for a 1.5A (minimum) power supply.  I'm using 2.5A RasPi PSUs on the machines here and they work great.  They're also inexpensive and easily-obtained, though you may need to add micro-USB to USB-A adapters to them.

[ivop]

The power supply has an effect on the color. The Color Adjustment circuit is driven by phase1 of the CPU, a ±5V signal. After that, it is run through a charge pump to create ±10V, which goes through a potmeter, which eventually creates the voltage that GTIA sees. Check those color pots. After all these years, they might need a good turning back and forth, and then leave them at the right spot. It's all analog.