Re: Tracking down a voltage dependent chip

Hi David,

No self test that I've ever heard of. Good job for the Fluke 9000 series
tool.

How about trying this - use a few independent power supplies, connect
the commons together to the logic boards and supplies, then separate the
+5 and +12 and -5 for one of the boards and run that board up/down on
the three voltages to see if it is sensitive. If not then move on to
another board...

I'm working under the assumption that you have a multi-layer boardset
here...

You can also isolate the 5V bus and power it separately if you have only
a single board (cut and add remote power).

If you can identify the board then you are that much closer to finding
the culprit! Heat gun (tiny nozzle) and cold spray can help too..

John :-#)#

On 09/21/2013 11:39 PM, David Shoemaker wrote:
>
> Well that didn't get me anywhere, still digging.
>
> Any idea if these boards have a self test mode? Would help to have a
> stable screen to debug.
>
> Thanks,
>
> David
>
> *From:*owner-rasterlist@vectorlist.org
> [mailto:owner-rasterlist@vectorlist.org] *On Behalf Of *Rodger Boots
> *Sent:* Saturday, September 21, 2013 3:42 PM
> *To:* rasterlist@vectorlist.org
> *Subject:* Re: RASTER: Tracking down a voltage dependent chip
>
> Bipolar (TTL (7400 series) and ECL and the old frame buffer RAM)
> speeds up as it gets warmer. MOS (large RAM/ROM/EPROM) slows as it warms.
>
> Everything gets faster with increased voltage.
>
> So you can lower voltage until it starts acting up then either warm
> bipolar or cool MOS a bit and note the effect.
>
> Also, ECL was used in early Nintendo and it ran from a separate -5.1
> (?) supply, make sure to check that. It doesn't sound like an ECL
> problem, though.
>
> On Sep 21, 2013 5:24 PM, "David Shoemaker" <davids@oz.net
> <mailto:davids@oz.net>> wrote:
>
> I have a DK which if the voltage is low at all (like 4.95v) will start
> to display graphic corruption (Kong gets stripes and princes hair
> starts to go to the wrong place). Any suggestions on finding the chip
> involved? I have done this before by piggybacking the ram / buffers
> around the sprite but in this case I get no change.
>
> Thanks,
>
> David
>

-- 
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
Call (604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames)
                  www.flippers.com
"Old pinballers never die, they just flip out"
---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com