Testing SMPS

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
I am finally getting back to working on this Plasma TV that occasionally
shuts down.

I would like to isolate the power supply to be certain that it is where the
problem lies. I have tried powering it up using 2 100V light bulbs as a load
on the 60v and 80v rails to no avail. It doesnt attempt to power up unless
it is attached to the Y-sus board.

Im guessing it gets some kind of ok signal from the VSS line?? Or is the
150W light bulb not enough of a load?


Re: Testing SMPS
Usually with a ticking problem, I will pull the cathodes of each diode one
at a time that are located on the secondary side of the SMPS xfmr, and power
up. When the ticking stops, then you have found the DC line causing the
problem.

Quoted text here. Click to load it


Re: Testing SMPS

Quoted text here. Click to load it

that works except for the supply that provides feedback,part of the control
loop.

or you could scope each line,see which one isn't coming up properly.

Quoted text here. Click to load it

you have to have a load on the supply that the PS regulates from.
Otherwise,the control circuit sees that supply rail come up too quick,and
shuts the PS down.
Many PC supplies regulate on the +5v logic rail.

Also,-some- switchers have a "balance node",where all the supply rails are
summed by sense resistors,and there's a "window" of operation and outside
that,it triggers a shutdown.That way,if one rail is too low or too high,the
balance node is outside the window(might only be a few 100 mv) and the PS
shuts down.

Quoted text here. Click to load it

--
Jim Yanik
jyanik
at
We've slightly trimmed the long signature. Click to see the full one.
Re: Testing SMPS

Quoted text here. Click to load it


Well actually the problem is the SMPS comes up but sometimes shuts down
right away.
After it heats up it USUALLY works fine, but occasionally will still
shutdown..

After hearing that the X-sus board may be drawing too much current, I wanted
to see if I could
test the PSU with a dummy load and see if it still shuts down.. As a simple
trail I hooked up
some 100v 150W light bulbs to the 80V and 60V supply.. Unfortunately it
doesnt supply power
unless connected to the x-sus board. I'm guessing it is something to do with
the low voltage
VCC supply providing some kind of OK signal.




Re: Testing SMPS

Quoted text here. Click to load it


That's fairly typical of high ESR/dried out electrolytics.

Some visual signs to inspect for are any electros with bulged tops. Very
occasionally the electros can have slightly concave tops - the cap overheats
and vents some of its electrolyte, when it cools there becomes a vacuum that
sucks the top in just perceptibly.

A failing electro invariably runs too hot, so look for discolouration or
shrinkage of the plastic sleeve - of course you could feel if any electros
are running hot, but the aluminium can isn't isolated so if the circuits
live - so is the capacitor can!

Going over the soldering can be enlightening - sometimes electros corrode
one of the lead wires and leak on the PCB, heating the lead cooks the leaked
electrolyte which makes an evil smell.



Re: Testing SMPS
Quoted text here. Click to load it
quick,and
outside
PS
overheats
that
leaked


Now I own a 0.1 deg Fahrenheit resolution non-contact IR thermometer , the
next problem SMPS I come across will get a scan across before ESRing
18GBP and its already found the awkward problem area of an analogue amp,
well worth adding to any repairman's toolkit
badged in UK as Maplin N19FR
http://www.maplin.co.uk/module.aspx?moduleno22%0790
it has a scan mode , not mentioned in instructions



Re: Testing SMPS

Quoted text here. Click to load it

Hmmm.. I would love to buy one of those. For now though my budget prevents
me from doing so..

I guess I will be removing and ESRing all of the caps again. First time
around did all in circuit and didnt find anything out of spec.



Re: Testing SMPS

Quoted text here. Click to load it

You need *VERY* low ESR for modern SMPSUs.

For some low value electros, like about 1uF or so it isn't too difficult to
find non-electrolytics small enough to fit in the space.

In some of the less critical places you can sometimes bolster an average
electro by adding multilayer ceramic chip capacitors on the print side - you
can get 10uF, maybe more.



Re: Testing SMPS
Quoted text here. Click to load it
line
rails
the
is
down
Very
or
circuits
corrode
the


I imagine with .1 deg F resolution , if you could get inside and get some
sort of decoded feed (LCD to BCD or whatever) out to a pc plus mount on an
X/Z platform driven by geared down pair of stepper motors and a bit of
colouring software you could make a poor-mans IR survey camera system.
Perhaps a scan taking a few minutes rather than near real time.



Re: Testing SMPS

Quoted text here. Click to load it


   You want a thermal imager, not a thermometer.


--
Politicians should only get paid if the budget is balanced, and there is
enough left over to pay them.

Re: Testing SMPS

Quoted text here. Click to load it

I have a Sears Craftsman model 50466.  It resembles the "Infrared
Thermometer" shown below the pocket version in your reference, and has a
laser for aiming.  "Scan" mode apparently means that it reads
continuously as long as you hold the button.  When you let go of the
button, it holds the last reading for 7 seconds.

I've noticed that a GFCI outlet may be about 8 degrees F higher than
room temperature.  The doorbell transformer is 7 degrees higher.  It has
proven quite handy when looking for hot components.  However, you need
to be aware that the sensor reads a spot that has finite size.  You
could be reading an adjacent resistor instead of a capacitor!

Fred

Re: Testing SMPS

Quoted text here. Click to load it

You do realize that a cold light bulb's resistane is very low...
Lightbulbs are not terribly good as a dummy load--use power resistors
instead.

Quoted text here. Click to load it

Probably the light bulb, cold, is too much of a load. Check the cold
resistance of your light bulb, the apply ohms law, and you'll find the
current--bet it is much more than the PS is capable of delivering.

Re: Testing SMPS

Quoted text here. Click to load it

Well the PSU will start with them as well as the X-sus board connected.
So I think the problem is related to something else. But I had that same
thought..

I am actually wondering if they are not enough of a load. Power resistors
would be the best
route, and I will be getting some soon, as soon as I can. I'm kind of short
supply here of everything.. I'm used to having a workshop full of tools at
my disposal. Here all I have is a DMM soldering iorn and my Bob Parker ESR
meter. I guess one of the prices to pay for living  abroad for a while..




Re: Testing SMPS
On Sat, 23 Oct 2010 00:14:22 +0900, "Michael Kennedy"

Quoted text here. Click to load it

I'd say the extreme variability of the light bulbs resistance makes
trying to accurately match them to a given load condition is going to
be difficult--hence the suggestion for resistors.

If nothing else, oil-cooled wirewound would work, as long as you can
keep the temperature stable.

Re: Testing SMPS

Quoted text here. Click to load it

don't forget you can parallel resistors to get higher wattages.
I've seen double-sided copperclad used to build dummy loads of considerable
wattages from 2W carbon comp resistors,and immersed in tin cans full of
mineral oil for cooling.

--
Jim Yanik
jyanik
at
We've slightly trimmed the long signature. Click to see the full one.
Testing SMPS Video
Here is a video of this thing clicking while warming up and cooling off.


http://www.youtube.com/watch?v=ckW8XSZ7T4A



http://www.youtube.com/watch?v=urx3JrsxFmo


I have gone through this thing with my ESR meter and havent found any bad
caps. I changed some suspect caps because of their location near hot
resistors.  To make things more difficult the inside of this tv is coated
with cigarrette tar... So some caps and things look blackened but in reality
its form smoke residue..


by the way thank you everyone who has made a suggestion  on this.  I am very
thankful everyone here is so supportive to amatures like myself..


Re: Testing SMPS Video

Quoted text here. Click to load it

My 32" LCD makes all kind of clicking noises while it warms up and
especially when it cools off after it is shut off. But it is mechanical
noise from the expansion and contraction of heated surfaces. Are you sure
you're not hearing some of this?



--
Live Fast, Die Young and Leave a Pretty Corpse

Re: Testing SMPS Video

Quoted text here. Click to load it

Hmm.. It is very possible the noise is expansion and contraction noise,
although it
can be very loud. When I first heard it I thought it was fairly big relay
clicking. I was unaware that plasmas usually make noise since this is my
first experience with the technology.

Since I became less focused on the clicking and more on looking for the
problem I believe that I have solved the problem. After thourghouly checking
the PSU and noticing that it wouldnt run without an OK signal from the X-sus
board I decided to give the X-sus a once over again.

I found a 1uF 50v mini alunimum cap that had an ESR of 9ohms.  It is
attached to an M3356 transistor. The board also has solder landing pads for
a SMD ceramic cap which was not populated. I think it would have been a
better choice becasue IME these small aluminum caps seem to go high in ESR
quite easily.

The only bad caps I've found in this thing have been 1uF on the Y-sus board
and X-sus.  Changed the Y-sus caps before without any improvement.

Anyhow.. I would like to thank everyone for all of their suggestions and
advice. I hope that this thing is fixed now. I guess thats the challenging
part about intermittant failures.. You don't really know if its fixed or not
unitl it stops working again..


- Mike



Re: Testing SMPS Video

Quoted text here. Click to load it


You're right on the money there, the smaller the electro (physical size) the
more likely it is to go high ESR - higher voltage types even more so!

Shouldn't be too hard to find a non-electrolytic 1uF that fits in the space,
or have a rummage for any forgotten about very old HDDs - they can be a good
source of large value multilayer ceramic chip capacitors.



Site Timeline