Seriously Hot Voltage Regulator - Faulty?

Thread Starter

James Farrell

Joined Oct 3, 2015
3
Hi guys. Long time reader and first time poster here. I'm hoping that someone can assist me in diagnosing a faulty motherboard. I'm getting the dreaded power on but not booting + black screen issue. It's for an all in one terminal that has been discontinued, so replacing the motherboard isn't an easy option.

I've noticed that a voltage regulator chip on the board is getting seriously hot. 93 degrees to be exact. Now could it be a matter of swapping the chip, and hopefully having a working board again if the heat hasn't done damage elsewhere, or is it more likely to be a case of something else faulty causing it to heat up.

The regulator in question is a CMPWR 280TN, and the datasheet for it can be found at http://pdf.datasheetcatalog.com/datasheet2/a/0scojciwx83zwoj076q14e1qkspy.pdf

I've measured the voltages at each pin and they are as follows:

1 (VSBY) - 5.08v
2 (VCC) - 5.04v
3 (GND) - 0v
4 (VOUT) - 2.45v - (is this a concern?)
5 (DRIVE) - 5.02v

According to the data sheet in the link above, VOUT should supply 3.3v right? Could this be where my problem lies?

If anyone can shed any light on the situation, or maybe suggest any other tests to do, it would be much appreciated. I have quite a successful history of repairing boards through component replacement, but when it comes to knowing exactly how particular components work and what they are supposed to do, my knowledge is a little limited.

Thanks in advance.

James
 

Dodgydave

Joined Jun 22, 2012
11,284
3.3v is usually for the memory chips, remove the plug in memory cards, see if it comes up to 3.3v.
Or trace what the output feeds and remove it.
 

ian field

Joined Oct 27, 2012
6,536
Hi guys. Long time reader and first time poster here. I'm hoping that someone can assist me in diagnosing a faulty motherboard. I'm getting the dreaded power on but not booting + black screen issue. It's for an all in one terminal that has been discontinued, so replacing the motherboard isn't an easy option.

I've noticed that a voltage regulator chip on the board is getting seriously hot. 93 degrees to be exact. Now could it be a matter of swapping the chip, and hopefully having a working board again if the heat hasn't done damage elsewhere, or is it more likely to be a case of something else faulty causing it to heat up.

The regulator in question is a CMPWR 280TN, and the datasheet for it can be found at http://pdf.datasheetcatalog.com/datasheet2/a/0scojciwx83zwoj076q14e1qkspy.pdf

I've measured the voltages at each pin and they are as follows:

1 (VSBY) - 5.08v
2 (VCC) - 5.04v
3 (GND) - 0v
4 (VOUT) - 2.45v - (is this a concern?)
5 (DRIVE) - 5.02v

According to the data sheet in the link above, VOUT should supply 3.3v right? Could this be where my problem lies?

If anyone can shed any light on the situation, or maybe suggest any other tests to do, it would be much appreciated. I have quite a successful history of repairing boards through component replacement, but when it comes to knowing exactly how particular components work and what they are supposed to do, my knowledge is a little limited.

Thanks in advance.

James
Are there any aluminium electrolytics in the immediate vicinity of the regulator?

High ESR can cause failure to stabilise the regulator - if it bursts into oscillation, it will get hot even with no overload.
 

GopherT

Joined Nov 23, 2012
8,009
@James Farrell
I think that some part downstream of your voltage regulator is drawing too much current (amperage) and the voltage regulator is reducing voltage as part of thermal protection feature.

It could be a tantalum capacitor, resistor or transistor.

Also look for Tin Whiskers. Certain solder made from 1995 to about 2007 has a the dances to make whiskers (fine fibers of tin that form from smooth solder joints over time and thermal cycling). The whiskers can form short circuits.
 

Thread Starter

James Farrell

Joined Oct 3, 2015
3
Just realised there's a transistor getting seriously hot too. Might this be something to look in to or is it more likely to be a result of the problem that I've still yet to find?
 

GopherT

Joined Nov 23, 2012
8,009
Just realised there's a transistor getting seriously hot too. Might this be something to look in to or is it more likely to be a result of the problem that I've still yet to find?
if it is an NPN or PNP type (bipolar) transistor, then the transistor should be supplying current to a resistor or capacitor with a short (much less resistance than expected. The flow of current through the transistor should be limited by a load of some type or a resistor. There is likely something wrong with the load or the current-limiting resistor connected to the Drain or Collector.

If it is a MOSFET transistor, then it is also possible that the driving signal supplying the gate of the MOSFET, then that part of the circuit may be oscillating too fast, or may have lost control (PWM control to maintain voltage). Alteratively, the MOSFET may be damaged.

Good luck.
 
i had a 7805 regulator that got well over 110c (yes that's Celsius and the regulator was on a heatsink!) when it was connected to 12vdc and an electric motor drawing 0.8a on the output.
that is normal for linear regulators though because of how inefficient they are.
 

GopherT

Joined Nov 23, 2012
8,009
i had a 7805 regulator that got well over 110c (yes that's Celsius and the regulator was on a heatsink!) when it was connected to 12vdc and an electric motor drawing 0.8a on the output.
that is normal for linear regulators though because of how inefficient they are.
Inefficient? That is the definition of what they are: energy loss devices.

Datasheet
=======
[(V(in) - V(out)) * I(load)] dissipated as heat: 100%
 
Top