Dell W3000 Problem with Power Supply

On Fri, 8 Aug 2008 22:52:43 GMT, snipped-for-privacy@repair.org put finger to keyboard and composed:

No specific help, but some general ideas ...

Assuming the Schottky diodes are on the secondary side, then it might pay to see what's connected to those rails.

Otherwise, if you think the uP may be shutting down the supply, then try to locate a PS_ON pin.

If it's a thermal problem, then try to narrow it down with some spray freeze and a hair dryer.

As a matter of course I'd check the ESR of all the electrolytic caps.

What is the part number of the PWM controller IC? Perhaps its datasheet has an application circuit.

- Franc Zabkar

The primary side controller is a MC33262PG.

Switching MOSFET: 20N60S5

The secondary schottky diodes (2): DSSK60-018A 180v 2X30A (I used two DSSK60-015A 150v 2 x30A) I could not find the exact replacement in stock.

One other schottky diode in secondary: 20CTQ150PBF (I ordered it but did not replace it because it was not shorted)

Bridge Rect.: D5SBA80-7000 6A 800v (I used a D5SBA60 6A 600v) At the time I ordered it, I could not find the exact replacement in stock.

I guess that could be part of the problem, but the replacements are very close. As you know, parts are hard to find.

Thanks for you help!

On Mon, 11 Aug 2008 14:46:34 GMT, snipped-for-privacy@repair.org put finger to keyboard and composed:

That's a power factor correction controller:

There must be an additional PWM circuit to drive the chopper transistor for the switchmode transformer.

Could it be that an excessive load on this particular rail is the root cause of your problem?

- Franc Zabkar

It also uses a least one ICE2AS01 controller IC. It may have more, but I don't remember right off.

On Wed, 13 Aug 2008 14:59:06 GMT, snipped-for-privacy@repair.org put finger to keyboard and composed:

That IC has its own internal overload protection but no external on/off control. I was wondering whether the PSU had some control pin that was being triggered by a fault condition on the logic board. I notice that a lot of people are having power-up problems with your particular model. One vendor is offering a replacement logic board as a solution.

Here is the datasheet for the PWM controller IC:

- Franc Zabkar

This unit does have powering on issues.

Can you tell me which vendor is offering a replacement?

Thanks

On Thu, 14 Aug 2008 03:51:35 GMT, snipped-for-privacy@repair.org put finger to keyboard and composed:

This supplier is offering an exchange logic board for US$150:

The symptoms are:

Power, Sound But No Picture Have to press power button numerous times to get a picture Power, Picture But no Audio Output

Not the same as yours, but it seems to be a common problem. I'd still be looking around the power supply area, though, before taking a chance on the logic board. If you do go for the board, I'd be keen to know what the fix is.

The same supplier is also offering a PS for $120:

- Franc Zabkar

I have two W3000's and one of them has had the problem since day one. It usually comes on first try, but often requires two tries. On one occasion, it took 12 tries. Anyway, my guess (and I stress the word guess) is that the backlight takes about 5 or 6 seconds to fire and they have some logic that gives up if the backlight hasn't lit after a certain amount of time. Unfortunately, some backlights takes just about the same amount of time as the timer allows. If the timer expires before it lights, the logic gives up and turns it off. (Occasionally, it lights just as the logic turns it off so I see a flash of picture). In other words, the only real problem is the length of the timer. I have no idea if the timer is implemented in firmware of some sort but it is frustrating to know a simple change of a timer could fix the whole problem. My other W3000 (a year older and therefore a lot more expensive has never failed once.) The problem one was purchased "on sale" right before they stopped making them. Hmmm.

...Pat

On Thu, 14 Aug 2008 12:40:57 -0400, greenpjs put finger to keyboard and composed:

I notice that some (most?) lamp controller ICs have a timing capacitor that is set to charge up to some trigger point in one second or so. If the feedback voltage from the lamp's current sense resistor does not reach 3V within that time, then the lamp controller shuts down. I suspect that the "fix" in this case would be to increase the value of the timing cap. Having said that, I can't see how a logic board swap would cure such a problem, unless the uP was able to detect the lamp in some way. It may be worthwhile to identify the functions of the pins in the harness between the logic board and the inverter/PSU, or at least to identify the lamp controller chip and study the application circuit in its datasheet.

- Franc Zabkar