Pan. PT61HX40 shutdown help JURB

[JURB6006]

PT61HX40B

Hi;

Me again. This is a post about a problem, but maybe also for a solution for all
of us. This particular model is an HDTV and the symptom is that it shuts down
after a few hours. For those who don't know, this set shuts down through the
uProcessor on a line called SOS. Printless at the time, and due to the fact
that the set would restart on <u>one</u> push of the power button, I told the
powers that be it could be shutdown or loss of standby or the uProcessor. I now
have a print but due to the way the set shuts down, the lack of a DSO poses a
bit of a challenge.

I would like to call your attention to:

<a href="http://members.aol.com/JURB6006/panscan1.jpg">panscan1.jpg</a>

or, if you would rather download the drawing and open it in an image editor so
you can zoom etc., click:

<a href="ftp://members.aol.com/JURB6006/panscan1.jpg">panscan1.jpg (public
FTP)</a>

It's about 180K.

At the lower left, the collector of Q1581 feeds the SOS line. Second from the
left is the "ABL" voltage, which can't really be called ABL anymore. It used to
be clamped and if it ever dropped, video drive would be reduced. In the newer
sets it is a HV current feedback line, and is used by the sweep circuits to
improve the stability of the raster. The third line from the left is the
filament voltage, and if it exceeds the limit, the set will shutdown.

Measurements at Q1581 would be useless, but at the two points I marked, it can
be determined whether the current or voltage is tripping it. The problem is
that you would have to be there when it happens. the way I see it there are a
few possiblities. One is that regulation is failing and tripping it through
Q1582, or a fault at or near Q1582. Another possiblility is that beam current
is going high and tripping it through D1581 &amp; D1582, or that one of those
zeners is leaky, or intermittently leaky. This is a higher probability than
most think because they are conected end to end and as such the juncture
between them is highly susceptible to static. Along with their proximity to the
flyback, they are suspect. As such, I'll order a pair Monday.

I'm not being pessimistic, but I cannot go on any assumption that D1581 &amp; 1582
will fix the problem. It's just that they are so cheap and they <u>might</u>
fix it. I must still troubleshoot it.

It's also possible that the zeners are doing their job and there is a loss of
ABL signal to the video circuit. I've had this in a different set and it was
really a fun one to run down.

Note that due to the highly intermittent nature of this problem I really really
want to avoid disabling any shutdown lines. In fact, I almost never do that
anymore. Too much can happen too fast.

What I would like to do is to build a little "tattletail" circuit. At the marks
I made on panscan1.jpg, either the cathode of D1581 or the collector of Q1582
is dropping <u>before</u> the other, and before the 12V line. The problem is
that the 12V line goes away when it shuts down, so this doohicky would have to
be self powered. I'm thinking wallwart, SCR, LED resistor and maybe an opto and
another resistor. I'm trying to keep it K.I.S.S.

Also, the presence of C1581 and R1582 puzzles me, unless there is something
else also feeding the SOS line. At the very least this would be sweep failure
detectors. Note that nobody in the shop has observed it actually going out, it
takes too long.

I get all the fun ones. (not whining, that is one of my functions there) So
nice to be able to share them with you though .

Thanx in advnce for any ideas, and when I do get this beast fixed, you will be
among the first to know. (of course with details)

JURB

 

[JURB6006]

OK, nobody has anything on this right now. No problem, I realize this is
something very specific in a plethora of subjects.

At this point I intend to build my little tattler and I'll keep you posted. I
was off today, but I called the shop and told them to order about 5 or 6
ECG5400s. I've not addressed the LED end of this, but I could always measure
voltages to see which one is in the on state. At this stage of the game I'll
just use one of the benchtop PSes we have kicking around.

Looking at panscan1.jpg, it looks like I should use some PNP bipolars to detect
the signal. I don't want to pull the hold current for the SCR from the circuit
points other than the 12V line and I don't feel like looking for anode fired
SCRs right now.

Of course if neither is tripped I must look elsewhare on the SOS line. I only
have a partial print handy, so those details will be subsequent. At this stage
I need to severally confim or disprove each of these first two possibilities.

I actually considered a multistable flipflop, but that won't work in the case
that neither of my test points have caused the trip. A no <u>and</u> no
indication must be available. We are back to the SCRs.

Two channels, one at the cathode end D1581 and the other at the collector of
Q1582. Both are referenced to the 12V line. This is accomplished by connecting
the emitters of the PNP transistors to the 12V line and feeding the bases
(though resistors of course) from the test points I indicated on panscan1.jpg.

The way I see it, this circuit is designed to give just a bit of leeway for
minor momentary lapses of reason by the HV regulation or the video drive. With
a 2.2K Rbe (R1586) and a 5.6K puller from D1581, they may intend for those
Zeners to be about 3V into conduction before the trip. What's more, they bypass
it hard with a 220uF.

I could address a couple more engineering issues right here, for one, R1585 is
a 2.2K. No current is being pulled to ground, except for the charging current
for C1584. D1583 easily handles any emitter currnet needed by Q1582m but the
use of a 47uF ( ! ! ! ) bypass makes this current nessecary. Shit they could've
used a 1 meg and a .0001 ! They're wasting power, not much, but sometimes I
think they do it to assure that the parts go bad sometime. Even a 220K and a
..47uFwould be better.

Anyway, sometimes I don't get any responses. It's either because some people
don't like me or something and/or that I do try to cover all the bases. At this
point I guess I'd like an opinion for a response, from those who agree with me
and especially those that do not. Am I nuts persuing this or is this a good
direction to take ?

Top techs in here, you know who you are, what would you do if the DSO broke
down ? You got ideas ? I got ears.

Thank you very much for anything.

JURB