Sony GDM-20E01 flyback drive voltages
Kevin Buhr
handy can help me.
I picked up one of these monitors from the trash. On power up, after
a short pause, it displays flashing power and power save LEDs and a
solid GEOM LED. From archived posts, it looks like this is just a
generic HV failure condition. Often, it means a bad flyback.
However, the flyback *seems* okay. As I say, I don't have a
schematic, but I can guess some of its function: pins 1 and 2 are
obviously the primary drive; pins 5 and 6 are apparently a small
secondary used to detect overvoltage conditions; pin 11 is the HV
ground return; and pin 12 seems like a ground for the core. Pin 13 is
apparently used to detect undervoltage, and it's somehow connected to
an internal resistance divider on the HV, according to an archived
post, but I'm not exactly sure about the details. I'm guessing pin 14
works into this, too. Pin 15 looks like it has an AC path to HV
through at least a 100k resistance, but I didn't look at it too
closely, and I'm not sure what its function is.
Can someone with a schematic give me more detail?
Out of circuit, there's continuity between pins 1 and 2 and pins 5 and
6. I can also drive 100V from pin 11 to the HV suction cup and get it
back after an appreciable voltage drop across the rectifiers, so
there's continuity there, too. Most importantly, it passes the ring
test: 5V pulses applied to pins 1 and 2 with a .047uF capacitor in
parallel produce beautiful, gradually decaying rings of 10 cycles or so.
In circuit, things aren't so rosy. The peak voltage from pin 6 is
supposed to charge C841 via D824 for overvoltage detection. According
to an archived post, 30V is normal, but I measure only 8V. The same
archived post indicates that pin 13, sampled by the unity gain op-amp
at IC803, should produce a voltage of 9V or so at IC803 pin 1 for
undervoltage detection. I measure a noisy signal at around .75V.
Perhaps the drive circuit is faulty? Q803 and Q804 seem to be working
correctly: Q804 applies 2usec 144V pulses to T801 pin 1 every 10usec
or so and Q803 pulls T801 pins 5 and 6 to ground for that 2usec pulse
and the following 4usecs. Some ASCII art shows what T801 and FBT pins
1 and 2 are doing during all this:
Q804 drain 144V ---- ----
& T801 pin 1 0V ---- ---------------- ----------------
Q803 gate 15V ------------ ------------
0V ---- -------- --------
Q803 drain 300V ++ (320V) ++
& T801 pin 5/6 200V || ||
100V /\ || /\ (80V) || /\
0V ' `-----------' `' `-------------' `' `
T801 pin 10 400V .. (370V) ..
& FBT pin 1 300V || ||
200V || ||
100V /\ || /\ (100V) || /\
0V ' `....-------' `' `....---------' `' `
-100V (-40V)
(140V)
FBT pin 2 100V ++ ++
(noisy) 50V /\ || /\(50V) || /\
0V ' `-----------' `' `-------------' `' `
Do these waveforms look right? Are 360V pulses what's supposed to get
fed to FBT pin 1? What's FBT pin 2 supposed to look like?
Thanks!
Kevin <bu...@stat.wisc.edu>
Steve Bell
Buhr" <bu...@stat.wisc.edu> wrote:
<snip>
> Most importantly, it passes the ring
> test: 5V pulses applied to pins 1 and 2 with a .047uF capacitor in
> parallel produce beautiful, gradually decaying rings of 10 cycles or so.
I usually test the GDM-20E01 FBT's with a DSE/Bob Parker flyback tester out
of circuit or with the solder sucked away from primary pins. This rings the
FBT and lights a series of LED's, up to a maximum of 8. Often I find these
faulty with between 5 to 7 LED's on, to be good all 8 need to be
illuminated. I must measure the inductance of a good one. 7 rings on the DSE
tester indicate good on it's scale, but not with these. You could still have
a defective FBT.
Steve Bell
Andy Cuffe
obvious problems like an open fuse, or bad caps in the power supply, the
flyback is probably bad. If there is HV before shut down, the flyback
is definitely good. I might be able to help you out with a good used
flyback if you determine that this is the problem. I have a couple from
junk monitors that I'll probably never use. Out of roughly 10 of these,
I've only seen one bad flyback.
--
Andy Cuffe
balt...@psu.edu
Kevin Buhr
>
> If there's no HV before it shuts down and you can't find any other
> obvious problems like an open fuse, or bad caps in the power supply, the
> flyback is probably bad.
In my monitor repair efforts up until now, I've never run into
problems more serious than bad capacitors and cracked solder joints,
so I don't have any sort of high-voltage probing equipment. There
seems to be high voltage at the MV input to the focus resistance
divider: with the cable pulled out of the divider, I can detect the
pulsing voltage with a low-voltage probe held in proximity to the
cable, but I can't yet measure the HV directly. I suppose I could use
the wet finger test under the suction cup... (HA HA!---Don't try that
at home, kids!) There's also some high-voltage "whine", but that
could be coming from the separate HOT circuits. My guess right now is
that there *is* high voltage, it may just not be high enough.
Anyway, sometime during my testing, Q804 (the MOSFET that pumps 144V
pulses into T801) shorted and it took PS800(?) with it, so I've got a
couple parts to scrounge up before I continue.
> I might be able to help you out with a good used
> flyback if you determine that this is the problem. I have a couple from
> junk monitors that I'll probably never use. Out of roughly 10 of these,
> I've only seen one bad flyback.
Okay, I may take you up on that.
Kevin <bu...@stat.wisc.edu>