Flyback Troubleshooting help

From: Jess Askey <jess_at_askey.org>
Date: Mon Jan 17 2011 - 01:26:12 EST

  Hi Everyone...

My Amplifone raster saga continues, I now have two deflection boards and
an original deflection coil (thanks John!) but nothing is working yet.
Im probably going to go into a little too much info here, but I figure
more is *more*. :-)

First Important Note: Im running this deflection PBC off the Atari
switching power supply which puts out 145VDC for the monitor. Oddly,
this is fed directly in where the 125VAC goes, which if you do the
analysis, is just fine... it is actually a little easier on the monitor
power regulator (120V) since it only has to regulate down from 145VDC
rather than 170VDC. The reason this is important is that the switcher
shuts down in an overloaded condition (saving me fuses and more) any
time I power up the monitor. When I first was troubleshooting this
monitor I had a few times where I doubted that the deflection PCB was
the problem so I put it on the regular 120VAC line... well, that proved
one important thing twice (yes, I was stubborn enough to learn this the
hard way twice). I shorted the HOT and the power regulator. So, lets
just say that Im pretty sure that using the Atari switching power supply
is working well as a troubleshooting method. As another note, the raster
deflection PCB has a 1A slow blow fuse in it. I measured the current in
my 1 second power-up and as best I can see, it peaks at around 1.1A
before the switcher turns off.

The Current (not amperage) Situation:

When I power up the game, the switcher ramps up and then goes into
overload in about 1 second. Not much troubleshooting that I can do in
that one second. If I disconnect the monitor, game comes up fine, just
blind.

All of this points to the Raster Flyback (Red just like the vector HV,
different windings tho) being the most obvious problem. However, There
are some more variables. I have another deflection PCB so I have
complete comparison to work with. Both deflection PCB's are doing the
same as described above. I have verified the following on both...

   1. There are no shorts across windings on either flyback
   2. the DC resistance of both flyback windings are the same(but fairly
      useless since they are so close to 0 ohms)
   3. The HV diode is not shorted in either flyback
   4. None of my secondary rectifier diodes are shorted... they all test
      good.
   5. Both deflection PCB's have been recapped
   6. Both deflection PCB's have good HOTs
   7. Both deflection PCB's have good HOT protection diodes (cross
      swapped and also used a spare 3rd diode)
   8. There are no longer any cold solder joints.
   9. There are no physical oddities such as cracks or shorts on either PCB

In a sad footnote, the second deflection PCB actually worked just fine
for about 30 minutes before the switcher started shutting down.

So, with all this... again... sounds like the flybacks are both bad
since those are such likely culprits... I don't have a ring tester, but
I have more evidence that they are somewhat healthy...

Second Important Note:

The Amplifone raster has an interesting design flaw that is pretty nice.
When the base of the power regulator (a TO-3, hfe=40 standard NPN power
transistor) is left floating, the output of the power regulator drops
down to a solid regulated 36VDC rather than the proper 120VDC (30%).
This is sort of like having a built in variac around but I can still run
on the switching power supply as protection. This gives me an
interesting result... I can actually let the deflection PCB run and the
switcher does not turn off due to overload, additionally, the Horizontal
Oscillator runs quite nice.... I have had it running while
troubleshooting for periods up to 30 minutes at a time so it is quite
stable too. This allows me to get out the O-scope and dig deeper...

First Reference: LOPT Testing Summary -
http://www.scribd.com/doc/40027645/Flyback-Tester which is short and
sweet, Sam Goldwassers LOPT FAQ is much more robust but makes the same
points on explaining they theory of flybacks and how to ring test them.

So, with the flyback running with a 36VDC primary voltage... here is a
picture of the collector voltage (normally not readable on a scope due
to the 1.5KV flyback spikes, but at our lowered rail, much better.

http://gamearchive.askey.org/Video_Games/Manufacturers/Atari/monitors/amplifone/raster/images/waveform_hot_padded.jpg

Items of Note on this Photo:

   1. I have really good ringing. If I had a short then based upon my
      understanding, ringing wouldn't be so good?
   2. Oddly, my output voltages are indeed concerning on the
      secondaries, they are the expected ratio?
         1. +180VDC (tap of primary) = 56V (this is about right based
            upon the ratio expected)
         2. +24VDC = 5.2V = 21%
         3. Filament (19VAC) = 2.5VAC = 13%
         4. Grid (?) = 19.4V
         5. Focus (?) = 25.5V
         6. Anode (25KV) = didn't measure
   3. Concerning: Is that Im only getting a 90V Peak reverse spike off
      the 36V primary... not good.. it should be higher... like 300V
      right? Something *is* damping this down it seems!!!

So.. all in all... it really does seem like my flybacks are done (both
of them). Still looking for more potentially good flybacks if anyone has
one laying around!!!!

Any comments or insights are welcome... yeah, I know I spent too much
time on this, but Im sort of aspergery!!! :-)

jess

PS - again the whole enchilada is here...

http://gamearchive.askey.org/Video_Games/Manufacturers/Atari/monitors/amplifone/raster/

---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Mon Jan 17 01:27:42 2011

This archive was generated by hypermail 2.1.8 : Mon Jan 17 2011 - 14:50:01 EST