Re: G-08 fire prevention...

Scott Swazey wrote:
> I've been looking at the schematics of the G-08. I'm trying to
> "harden" the G-08 so it won't catch fire.
>
> I think I have figured out the problem, it's the 3A slo-blo fuse
> called out in the schematic. It protects the Yoke from a Transisor
> Vce short. I believe that a Slo-blo is the wrong type of fuse to use.
>
> Here's why. The most common failure on a G-08 is a Vce Short in one
> of the power transistors (e.g. 2N6254) and this happens when large
> currents are flowing through the yoke. This means that, at the point
> of failure, there is +63 volts across the series connection of the
> yoke, the fuse, and a 1.5ohm 10Watt resistor to ground.
>
> First Simplification: Assume resistance of the yoke and the fuse is
> zero. At the point of failure, the current in the resistor is 63V/1.5
> or about 40amps. The 10W resistor will have to dissipate 40*40*1.5 =
> 2400Watts! until the fuse blows. From the littlefuse 3A Slo-Blo
> datasheet <http://www.littelfuse.com/data/en/Data_Sheets/313_315.pdf>
> at 40amps, the 3A fuse will blow in about 60ms.
>
> Realistic Simplification: assume the resistance of the yoke to be
> about 1 ohm. This drops the current in the resistor to 63/2.5 =
> 25amps. The resistor will have to dissipate 25*25*1.5 = 780Watts!
> until the fuse blows. From the littlefuse 3A Slo-Blo datasheet
> <http://www.littelfuse.com/data/en/Data_Sheets/313_315.pdf> at
> 25amps, the 3A fuse will blow in about 200ms. 200ms at 780W is plenty
> of time to smoke 10W 1.5 ohm resistor.
>
> Therefore, I believe that the 3A slo-blo is the wrong type of fuse.
> Amplifone monitors use a 3A fast blow in the output stage. So I'm
> thinking a 5A or 7.5A fast blow fuse could work. In the circuit, the
> fuse is primarily protecting the yoke windings as the transistors are
> the ones that are going to fail.
>
> Thoughts... Comments...
>
> Thanks,
>
> -Scott
>
> PS I've significantly simplified this analysis to keep it brief. I
> can go into much greater detail if necessary (e.g SOA vs fuse size,
> effect of yoke inductance on peak current and back EMF voltage, etc.)
>
> FYI
> At 03:00 AM 10/20/1999, Al Kossow wrote:
>> Date: Wed, 27 Aug 1997 15:53:15 -0700 (PDT)
>>
>> If anyone is curious, I measured the inductance and DC resistance
>> of [the G-08] XY deflection yoke today. Here's the measured
>> values:
>>
>> type wire color L DC R notes
>> ----------- ---------- ----- -------- ---------
>> Triad-Utrad blue .41mH .43 ohms
>> 21-143-01 red
>> 17-8121 green .52mH 1.0 ohms
>> Electrohome yellow
>> G08-003 Color
The biggest cause of GO-8s blowing out is unbalanced common (often
referred to as ground) connections between the monitor, power supply and
logic board. If you make a very secure connection between the three you
are less likely to have any serious problems. Almost all of the blown
GO8s that I saw back in the days when these games were operating had to
do with toasted connectors at the power supply. This would lead to
ground shifts relative to the monitor/PCB common and drive the monitor
outside its design parameters. Its also the problem that a number of
pinball manufacturers had with coils slowly melting or staying energized
- Gottlieb in particular. Again if you fix the commons the problems go away.

So the simplest solution for protecting a GO8 based game is to replace
the linear 5VDC supply with a good switcher and run extra common wires
(18 gauge!) between the power supply, logic board and monitor commons.
Solder the connections at the monitor and game board is best.

No monitor game combination that I did this to had a monitor failure
that I ever heard about later. And as the only shop in western Canada
that still fixes vector video games I think I would have by now...

You can't have enough common lines between logic boards and the things
they run...

John :-#)#

-- 
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