info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Alba LCD19advdw, LCD TV and DVD, 2009
28 views
Skip to first unread message
N_Cook
Apr 25, 2016, 10:40:35 AM4/25/16
to
Failed backlighting. Firstly 2 off 8pin SOIC with 4 pins commoned for
heatsinking, to drive backlight Tx, on the mains SMPS LV side.
Markings look like
PRAAN
FOS
8862C
under 30x microscope,
google comes up with nothing, anyone suggest likely mistrancription
alphanumerics?
DVD plays and sound emerges, not tried a torch on LCD as all a lash-up
at the moment.
That Tx powers 2 separate tubes with only 2 3KV caps the other
components, 1 of 3.9pF and other 4.7pF , different size tubes likely?.
Both separate secondaries of that Tx measure 830 ohm each
500V Insulation testing showed >200M for each of those 3KV caps.
Not knowmingly have any puffy 3KV caps around and tried 2 x 27pF 12KV
ones instead. The whole unit went into a protection mode it seems, no
DVD or sound. Replaced the original 3KV caps and back to the previous
situation. Does this suggest the 2 drivers are working, they cold test
DVM-D pin to pin combinations matching across each ?
Unplugging each lamp in turn makes no difference.
Only the second time I've peered into backlight operation , so somewhat
unknown territory, previous time was leaky HV caps.
heatsinking, to drive backlight Tx, on the mains SMPS LV side.
Markings look like
PRAAN
FOS
8862C
under 30x microscope,
google comes up with nothing, anyone suggest likely mistrancription
alphanumerics?
DVD plays and sound emerges, not tried a torch on LCD as all a lash-up
at the moment.
That Tx powers 2 separate tubes with only 2 3KV caps the other
components, 1 of 3.9pF and other 4.7pF , different size tubes likely?.
Both separate secondaries of that Tx measure 830 ohm each
500V Insulation testing showed >200M for each of those 3KV caps.
Not knowmingly have any puffy 3KV caps around and tried 2 x 27pF 12KV
ones instead. The whole unit went into a protection mode it seems, no
DVD or sound. Replaced the original 3KV caps and back to the previous
situation. Does this suggest the 2 drivers are working, they cold test
DVM-D pin to pin combinations matching across each ?
Unplugging each lamp in turn makes no difference.
Only the second time I've peered into backlight operation , so somewhat
unknown territory, previous time was leaky HV caps.
N_Cook
Apr 26, 2016, 9:26:42 AM4/26/16
to
I've some tubular ceramic caps, 1p8 unknown voltage, scraped back the
covering and 2.4mm electrode separation , so perhaps 3KV or so.
Tried 2 of those on each secondary and no change
2 extenders underneath, so I tried swapping tubes over , no change.
Wired in the tube of a CFL onto the plain unmonitored side of the Tx, no
glow. Removed it and fitted it on the monitored side and 6 seconds after
standby to on, 1 second maintained burst of lamp light. I could repeat
this procedure but only ever 1 second , presumably before the monitoring
decides there is an internal tube short or something.
These CFL tubes have about 10R heaters? either end, but nothing <30 Meg,
over the 4 ends of the original tubes.
So assuming the burried tail connections are ok and the tubes are not
physically broken , what can go wrong with these TV backlight
flourescent tubes, and both of them?
What if any ohmage across the pairs of wires at each end would I expect?
covering and 2.4mm electrode separation , so perhaps 3KV or so.
Tried 2 of those on each secondary and no change
2 extenders underneath, so I tried swapping tubes over , no change.
Wired in the tube of a CFL onto the plain unmonitored side of the Tx, no
glow. Removed it and fitted it on the monitored side and 6 seconds after
standby to on, 1 second maintained burst of lamp light. I could repeat
this procedure but only ever 1 second , presumably before the monitoring
decides there is an internal tube short or something.
These CFL tubes have about 10R heaters? either end, but nothing <30 Meg,
over the 4 ends of the original tubes.
So assuming the burried tail connections are ok and the tubes are not
physically broken , what can go wrong with these TV backlight
flourescent tubes, and both of them?
What if any ohmage across the pairs of wires at each end would I expect?
N_Cook
Apr 26, 2016, 11:40:11 AM4/26/16
to
So it is 2 wires per tube and 4 tubes, not 2 , it seems.
Tried 2 CFL tubes on the monitored side and both light for 1 second.
So I can leave one CFL in there and use the other supply to test each TV
tube plugged-in in turn, surely 4 could not have failed at once, for 1
second.
Tried 2 CFL tubes on the monitored side and both light for 1 second.
So I can leave one CFL in there and use the other supply to test each TV
tube plugged-in in turn, surely 4 could not have failed at once, for 1
second.
N_Cook
Apr 27, 2016, 7:41:50 AM4/27/16
to
Tested each tube and number 4 will not work.
Perhaps dodgey connection of one of the other pair, but now the original
pair of 2 tubes at the top work for 1 second.
Cleaning pcb around HV areas and resoldering HV and monitoring solder
points , still 1 second only.
What sort of compact load to represent the duff tube? a few neon bulbs
in series perhaps.
I don't understand the monitoring as connects to midway point of 2 tubes
and no meggy resistors around, just 10s of K.
One SMD diode marked A2 is used elsewhere in the LV circuitry, as is
transtor marked 3FW. 2 perfectly normal pitch spacing SO23 Q marked 7.AJ
or 7AJ are not repeated elsewhere but not obviously HV
Perhaps dodgey connection of one of the other pair, but now the original
pair of 2 tubes at the top work for 1 second.
Cleaning pcb around HV areas and resoldering HV and monitoring solder
points , still 1 second only.
What sort of compact load to represent the duff tube? a few neon bulbs
in series perhaps.
I don't understand the monitoring as connects to midway point of 2 tubes
and no meggy resistors around, just 10s of K.
One SMD diode marked A2 is used elsewhere in the LV circuitry, as is
transtor marked 3FW. 2 perfectly normal pitch spacing SO23 Q marked 7.AJ
or 7AJ are not repeated elsewhere but not obviously HV
N_Cook
Apr 27, 2016, 7:43:37 AM4/27/16
to
7AJ are 3 pin diodes, D on the overlay
N_Cook
Apr 27, 2016, 9:50:49 AM4/27/16
to
On 27/04/2016 12:43, N_Cook wrote:
> 7AJ are 3 pin diodes, D on the overlay
not 7AJ but A7J
> 7AJ are 3 pin diodes, D on the overlay
I "replaced" the duff tube with 3 neons in series on the ps, and the
display and backlighting and DVD play works fine. So looks like the
unmonitored side needed a load also to bring the monitoring into valid
envelope. I have to admit its not obvious which area of the screen is
lacking light, needs wrapping up properly and a normal viewing position
perhaps. Would there be 2 long tubes top and bottom and 2 short along
the sides or 2 short each along top and bottom, I've not ventured fully
inside.
The neons run at 24 deg C over ambient, so perhaps 5 or 6 would be
better if more than a temporary fudge.
Going the other way would a CFL lamp make a TV/laptop backlight exerciser?
10 ohms between pins 1 and 2 and another 10R between pins 3 and 4 and
then the tube across 1&4 or 2&3.
N_Cook
Apr 27, 2016, 11:29:05 AM4/27/16
to
Have a small collection of thin tube flourescents from
fax/scanner/laptops of unknown states.
I know at least one is green but I did not label as such, so made an
exerciser for 2 wire tubes , from the CFL board.
Tubes between pins 2 and 3.
2x 10R,.5w soon blew , 2x 0.33R,2W in positions 1 to 2 and 3 to 4
, probably too far the other way, but allowed to run each of them for a
few seconds to check for colour and non-evenness or strobing.
fax/scanner/laptops of unknown states.
I know at least one is green but I did not label as such, so made an
exerciser for 2 wire tubes , from the CFL board.
Tubes between pins 2 and 3.
2x 10R,.5w soon blew , 2x 0.33R,2W in positions 1 to 2 and 3 to 4
, probably too far the other way, but allowed to run each of them for a
few seconds to check for colour and non-evenness or strobing.
Jeroni Paul
May 7, 2016, 8:21:09 AM5/7/16
to
N_Cook wrote:
> I "replaced" the duff tube with 3 neons in series on the ps, and the
> display and backlighting and DVD play works fine. So looks like the
> unmonitored side needed a load also to bring the monitoring into valid
> envelope. I have to admit its not obvious which area of the screen is
> lacking light, needs wrapping up properly and a normal viewing position
> perhaps. Would there be 2 long tubes top and bottom and 2 short along
> the sides or 2 short each along top and bottom, I've not ventured fully
> inside.
> The neons run at 24 deg C over ambient, so perhaps 5 or 6 would be
> better if more than a temporary fudge.
I have been looking for a practical/cheap solution for the same problem on a DELL 1707FPt computer display. It has four backlight tubes two at the top and two at the bottom. One of them was cracked and very darkened inside, the rest were fine. Problem is the inverter is not happy without a load on that output, runs for one second and shuts down. I tried three 3W resistors in series adding 120K and it works fine, yet the resistors get quite hot. The missing tube is barely noticeable, it just causes the top 1/3 of the srceen to be slightly darker but goes mostly unnoticed. I would be happy to make it work with three tubes, installing the resistors permanently is one solution but I wondered if the inverter could be modified somehow to disable or ignore that output. The inverter uses a transformer for each pair of tubes so not sure if there is anything that can be done.
> I "replaced" the duff tube with 3 neons in series on the ps, and the
> display and backlighting and DVD play works fine. So looks like the
> unmonitored side needed a load also to bring the monitoring into valid
> envelope. I have to admit its not obvious which area of the screen is
> lacking light, needs wrapping up properly and a normal viewing position
> perhaps. Would there be 2 long tubes top and bottom and 2 short along
> the sides or 2 short each along top and bottom, I've not ventured fully
> inside.
> The neons run at 24 deg C over ambient, so perhaps 5 or 6 would be
> better if more than a temporary fudge.
N_Cook
May 7, 2016, 3:12:58 PM5/7/16
to
droppers, replacing the failed tube for loading purposes, for 3 hours as
a try out of whether it was possible
0 new messages