greetings

so a few days ago I wanted to turn on my Viewsonic VP191s LCD display but it would not, not even the power led would light up. So I took it apart, looked for burned caps but found none. Next I took out the PSU and checked the output voltage: When turned on there is a faint clicking sound coming from the PSU, ca. 3/5 per sec, the voltage fluctuates between 12.1 and 12.3 V.

If I add a load, a 12V 5W bulb,

either of the following happens:

1) the voltage drops instantly, the bulb does not light up, and the clicking sound speeds up, about twice as fast, or

2) the bulb lights up and the output voltage stabilizes at 12.3V.

I have only tested the PSU for less than a minute.

Now I'm not sure what to check for next, any ideas?

thanks

 

Replace the small electro cap on the PCB (bottom of your photo) next to the heatsink.

 

Replace the small electro cap on the PCB (bottom of your photo) next to the heatsink.

Hi,

that is a 50V 22uF cap. The other small cap is also 50V 22uF cap. The two bigger caps are 16V 1000uF, they are actually soldered in parallel, I don't understand why they just didn't simply use a single larger cap....?

And the biggest cap is a 400V 120uF.

btw, why do you think it's the small cap by the heat-sink that is the culprit?

 

What test equipment do you have to test the electrolytic capacitors?

Do you have an ESR meter?

Do you have an oscilloscope?

Do you have a lighter load, say 220 ohm resistor 1W ... to be used for testing?

 

I don't have an ESR meter but I do have an oscilloscope and many resistors and and a few multimeters.

 

... that is a 50V 22uF cap.
...
btw, why do you think it's the small cap by the heat-sink that is the culprit?

Thirty five years experience and a big brain.

 

Thirty five years experience and a big brain.

hmm, I thought you might have dealt with this specific PSU in the past.

 

Did you check that little capacitor to confirm or exclude the_RB's advice?

 

ok, I replaced the small 22µF cap, turned the PSU on, it made a very faint high frequency noise, I turned it off anf on again, the noise was gone. It's been on for about an hour now with the 5W bulb load, so far so good.

I guess I can call it repaired? Or should run some tests b4 installing it back into the LCD?

 

Seems like it is fixed.

The fault "clicking" you described is usually called "hiccupping" and it signifies a fault shutdown in the feedback loop once the SMPS tries to power up. The fault shutdown repeats, making a "hiccupping" sound.

Prime suspects are small value caps in the feedback loop usually connected with the optocoupler. The cap failure mode is "high ESR" which means its voltage ripple is quite high so the opto feedback loop thinks the ripple voltage peaks are an overvoltage situation. That causes overvoltage shutdown hiccupping.

Small caps in hot SMPS PSUs are prone to "drying out" or going high-ESR, and this is especially common with a small feedback cap close to a heatsink.

It has been suggested by some people that the manufacturers deliberately place the critical cap(s) next to heatsinks even when there is no need (heaps of PCB space) as a way to reduce the life of their product. Under normal operating temps they can place that cap so that the unit fails after approx 12-18 months, which suits their sales figures just fine.

(If you look at your photo that small cap seems deliberately close to the heatsink, it could just as easily have been placed 1/2 inch further away when the unit was designed).

 

thanks for the help and info!

It has been suggested by some people that the manufacturers deliberately place the critical cap(s) next to heatsinks even when there is no need (heaps of PCB space) as a way to reduce the life of their product. Under normal operating temps they can place that cap so that the unit fails after approx 12-18 months, which suits their sales figures just fine.

(If you look at your photo that small cap seems deliberately close to the heatsink, it could just as easily have been placed 1/2 inch further away when the unit was designed).

Now that's interesting, reminded me of the movie http://en.ekopedia.org/The_Light_Bulb_Conspiracy

When I was desoldering the faulty cap the copper tracks detached from the PCB so I had to solder two thin wires to which I soldered the new cap. Now I'm thinking of extending these wires to place the cap further from the heat-sink. Would 1/2" be the max distance from its original placement?

Another thing that pissed me off is the goo-cream that I had to remove in order do get cap out, why do they put this goo on the pcb components, to make it a pain to repair?

p.s. this LCD worked ok for almost 10 years until this cap failed. The Viewsonic engineers probably placed it way too far from the heat sink.

 

Haha! Yeah they got that wrong, or maybe you live in a cold area?

I don't think its a paranoid conspiracy as much as a fact, over the decades I've seen it get worse and worse, with many devices that have a couple inches free space and room for track routing etc, BUT still put that one cap right next to a heatsink! That was officially considered bad design in the old days. These days it is standard practice.

Re the cap legs good work moving it away from the heatsink, I normally do the same if there is room and lay it over away from the sink. Any distance there you can provide will help, and an extra half an inch away from the heat source is excellent but don't go overboard and put real long legs on it or long flywires.

Generally the goo is to save parts from mechanical shock (drop damage etc) and often for vibration, to stop them mechanically ringing at the SMPS frequency.

I also suggest you use a 105'C rated cap for that part, although if you got it away from the heat a bit that might be ok with a standard 85'C cap. I have a collection of thermal probes and infrared themometers etc to measure how hot a critical parts gets after repair. If you are going to repair PSUs that will be very handy, and you might also want to google for "ESR meter" products or kits. They can measure where an electro cap is in its ESR failure ageing, even before the whole device fails.