Capacitors repair on an LCD monitor are these Caps good ones?

  • Thread starter Thread starter Gabriel Knight
  • Start date Start date
Gabriel said:
http://wikisend.com/download/864364/Troubleshooting.pdf

it mentions the crystal as a possable fault and to check if it is
oscillating normally how does one check this?

You could check with an oscilloscope, that a buffered clock signal
coming from the chip, is working. For example, there is an
MCLK signal on one side of the scaler. I'd connect my oscilloscope to
that and check for something running at hundreds of megahertz.
That proves chip is powered, PLLs are working, and there is a
clock signal on output. If the 14.318 was dead, MCLK would be too.

Probing right at the crystal terminals is not recommended,
since the slightest capacitance (a couple pF) can shift
the operating point, detune the thing, change the Q, etc.
You could just as easily stop it from oscillating as anything
else. I've never tried probing a crystal with a FET probe,
just to see what would happen. I wouldn't even try. It
wouldn't be my first point of attack. (If there was no
buffered copy of the clock signal available, then I'd have
to investigate the right kind of probe for the job. A FET
probe is the closest I know of.)

(Down to 0.5pF loading now. Impressive. One of these
would cost as much as a very old used car :-) )

http://www.tek.com/datasheet/active-probe-4-ghz-active-probes

And as a belt and suspenders guy, I don't even like to
design things that way anyway. I like tin can oscillators
with four legs on them, for that "guaranteed to work"
design result. I've seen enough disasters related to
improperly specified two-lead crystals, to not want
any part in it. As long as the boss isn't all "those
are too expensive - get that outta there!", I'll continue
to use the four legged tin cans instead. As long as they
have a tri-state pin on them, for the guys in the factory
to run test, they're great.

I've seen processor designs, that used the two leg quartz
crystal, and the processor crashes every couple of days,
and people are scratching their heads. Turns out, the
oscillator is the weakest link, not the rest of the
hardware. That's where I learned about the value of the
four legged tin cans. No more "grrr... shit crashed again".

(Four legged friend)

http://www.smcelectronics.com/XTALOSC.JPG

One of my managers, took me aside one say, and tried
to explain how to select AT-cut crystals. Went through
the steps you'd follow and so on. I'd do that again
if I had to, but when you buy the ones with the four
legs, you're off the hook in terms of whether it
works properly or not. The ones with the four
legs are digital devices. Feed them clean power
and away you go.

Paul
 
I always like your detailed replys Paul some I understand some I dont I did
search to find answers to test a crystal and you confirmed its not good to
test a crystal with an oscillascope somewhere I read it can do more dammage,
I did remove the crystal as my first suspect component but now I have
soldered it back on and it might be out in pF a bit or a bit out of tune
from what you said in your reply so I have found a replacement crystal
though I havent got the total settings for this crystal. A guy called John
at digikey.com helped to find it going with the "HC-49/S-SMD (14.31818MHz)"
in the schematic and the "14.318180 KONY 0832" on the crystal I have....John
found this:

http://www.digikey.com/product-detail/en/FOXSDLF/143-20/631-1015-1-ND/1024720

Frequency Stability ±50ppm
Frequency Tolerance ±30ppm
Load Capacitance 20pF
ESR (Equivalent Series Resistance) 40 Ohm
Operating Mode Fundamental
Operating Temperature -20°C ~ 70°C


but im not sure the above settings are compatable as I see there are many
options to choose from when in the selection of what type of crystal to
search for on the site. I will buy five if you reccomend them but if there
is a better one you can find I will go with that. Thanks again Paul and
everyone, GK. Oh yes about the S301 sensor when I tested it with the ohm
meter its pins work on its face's

------------
3 = ! ! = 4
! !
1 = ! @ ! = 2
------------

when its horizontal the ball ( @ ) uses pins 1 and 2 and when you move the
screen clockwise 90 degrees it sits on pins 2 and 4 turn again 90 degrees to
upside down horizontal for pins 4 and 3 and lastly clockwise again for 3 and
1. I will have to bridge a link accross 1 and 2 straight and not like a
cross from 2 to 3... Cheers everyone, from my mistakes I have learnt a lot
about component identification and data sheets and not to remove first and
research later.
 
Gabriel said:
I always like your detailed replys Paul some I understand some I dont I did
search to find answers to test a crystal and you confirmed its not good to
test a crystal with an oscillascope somewhere I read it can do more dammage,
I did remove the crystal as my first suspect component but now I have
soldered it back on and it might be out in pF a bit or a bit out of tune
from what you said in your reply so I have found a replacement crystal
though I havent got the total settings for this crystal. A guy called John
at digikey.com helped to find it going with the "HC-49/S-SMD (14.31818MHz)"
in the schematic and the "14.318180 KONY 0832" on the crystal I have....John
found this:

http://www.digikey.com/product-detail/en/FOXSDLF/143-20/631-1015-1-ND/1024720

Frequency Stability ±50ppm
Frequency Tolerance ±30ppm
Load Capacitance 20pF
ESR (Equivalent Series Resistance) 40 Ohm
Operating Mode Fundamental
Operating Temperature -20°C ~ 70°C


but im not sure the above settings are compatable as I see there are many
options to choose from when in the selection of what type of crystal to
search for on the site. I will buy five if you reccomend them but if there
is a better one you can find I will go with that. Thanks again Paul and
everyone, GK.

<<snip>>

I would assume the parts are still working. Unless you know
for sure you've damaged them while removing them.

*******

I'd try taking some identifying information off the crystal
you've removed from the board, rather than relying solely on
the schematic.

Looking at the item in hand, it's either a thru-hole or
a surface mount part. Which makes a big difference when buying
a replacement. The HC-49 is specifying a package type. But
the ones I was looking at, looked like thru-hole. Whereas
your schematic mentions SMD. Perhaps the crystal is dual-footprint ?
That means there are two outlines drawn on the surface of the PCB,
two sets of pads, and the stocklist populates either one location
or the other. This make it possible for the factory to substitute either
form, depending on what is cheapest or available on a given day.

Paul
 
Back
Top