HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

[vettouch]

HP Compaq 6510B inventec Motherboard not powering and U28 IC getting very hot. Please help me to pin down the fault. see schematic

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Remove all power -> meter in resistance mode -> check the resistance to ground @ L20.

One meter probe @ L20 (either side is ok) -> other meter probe to ground.

Someone is consuming too much current and causing this regulator to heat up.

It is possible that U28 is now defective but start with the resistance check.

Also check the other half of this regulator by checking the resistance to ground @ L21.

Post the readings.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@mon2

metre @ 2000 ohms;
L21 = 774 (each of the sides)
L20 = 1386 (each of the sides)
L15 = 005 (each of the sides)
L16 = 005 (each of the sides)
L8 = 042 (each of the sides)
L12 = 764 (each of the sides)
L7 = 165 (each of the sides)
L6 = 1784 (each of the sides)

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Someone please, help me resolve this issue.

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Remove all power. Measure the resistance to ground of U28, Pin 19.

Then repeat to check U28, Pin 21.

These are fixed LDO rails that may be the issue.

If heating up too much then there is a fault linked to this device or a transient has now killed this component.

The posted resistance on the coils for this part are ok and not a short.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@mon2

resistance checks on pin 19 and 21 of U28

metre readings @ 2000 ohm setting
pin 19 = 811
pin 21 = 793

Thanks and please, assist me more.

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

pin 19 = 811 ohms
pin 21 = 793 ohms

The LDO resistance readings to ground are not considered shorts but the resistance is low.

Any chance you have a thermal camera ?

If not, you can try to locate the faulty part by removing components like the capacitors, one at a time. After each removal -> check the resistance on the LDO rail pin again. If the low resistance remains, then place the removed part back onto the PCB.

For example consider to remove the 4.7uf caps on these same voltage rails; the 1uf cap on this rail for starters. Voltage injection is another process but this method is much more safe.

So remove C336, C333, C335 but place each back immediately if the resistance is not any higher so that you do not lose the parts nor the location from where they were removed.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@ mon2. I discovered 3.8v and 5.6v around this U28. What could be causing it? Also, most of the caps, coils and the likes around ICH are shorting on test with metre. Could the SIO be bad?

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

I discovered 3.8v and 5.6v around this U28.

Can you be more specific on where these voltages were measured ?

Which pin # ?

Is your multimeter with fresh batteries and is accurate ?

If you are truly reading ~3v8 on the 3VA rail, that may cause damage to other parts on this rail as that is very high to parts expecting ~3V.

The voltage output of each of the buck regulators inside U28 are defined by the FB pins (feedback pins). Respectively, each resistor on the FBx pin must match the values shown in the schematic.

To confirm, remove all power / no battery -> meter in resistance mode.

Check the resistance of each of the resistors on the FB1 pin. Then repeat for the FB2 pin. Be sure that each resistor matches the schematic values.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@mon2. I can't locate R425, R424 resistors. I feel they are not labeled. Please, how can I locate them?

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Locate pin #1 of the ic -> then find pin #6 on the device.

This the the feedback pin for this voltage rail.

With no power, meter in resistance mode, one meter probe on pin #6. Other meter probe to a close by resistor. When you find a resistor with 0 ohms, that will be the feedback resistor circuit.

Post close and clear pics of the area on your logic board.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@ mon2. While diagnosing this board yesterday, I suddenly discovered a burn on Q43. So, I found a replacement for it. After that, I also discovered shorts on Q49, C290, C291, and C336. What have I done wrong? See the attached picture of the board.

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

What was the part # on the original Q43 ?

What part # was used to replace the burnt Q43 ?

With no power, what is the resistance to ground of the parts you are observing to be shorted ? Select the lowest resistance scale on your meter for these measurements. Post the readings.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

original part # for Q43 = FDS8884
replaced it with same part # from another board

resistance to ground for parts observed to be shorting
metre @ 200 ohms
Q49 = 16.2
C290 = 16.9
C291 = 16.7
Q51 = 16.5
C336 = 17.9
C308 = 38.9

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Review, without power, in resistance mode on the meter,

the resistance across the mosfet:

source (1-2-3) & drain (5-6-7-8)
source (1-2-3) & gate (4)
gate (4) & drain (5-6-7-8)


You may have to use a higher resistance meter scale for a measurement of the mosfet pins.

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@mon2

Resistance check on Q49; metre @ 20k ohms
source and drain = 0000
source and gate = no reading
gate and drain = no reading

metre @ 2000 ohms

source and drain = 0001
source and gate = 1558
gate and drain = 438

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

With care + flux + low air pressure (so other parts do not go flying with an air ticket); remove Q49 off the board.

Before doing this, make a note on the pin #1 location.

You can also view the pads UNDER the component for guidance as:

pins 1-2-3 = SOURCE
pin 4 = GATE
pins 5-6-7-8 = DRAIN

Then check this part when it is out of circuit. What are the resistance values?

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

@mon2

Resistance check on Q49; metre @ 2000/20k ohms
source and drain = no reading
source and gate = no reading
gate and drain = no reading

[mon2]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

Can you clarify?

The actual mosfet was removed off the board and now the new measurements are of the PCB pads or of the part that is on your table?

[vettouch]

Re: HP Compaq 6510B inventec Rev 1.0 - U28 IC GETTING HOT

The new measurements are of the part that is on the table (mosfet). The PCB pads still read source and drain = 0001
source and gate = 1558
gate and drain = 438