Stuck...power issues...bad board?..thoughts?

[Encrypted Existence]

Hello all. I have an HP Probook 4510s on the bench.
Specs: 4 GB DDR 3 RAM
Intel core 2 duo
250 GB Hitachi HDD

I ran an extended self-test with gsmartcontrol on the HDD it passed. I have been unable to test any other components as you will see below.

Customer informed me that the machine wouldn't properly power on. The first thing that I did was test the AC adapter with my meter and the output was what it should be. I noticed that the power LED was not lit when the AC adapter was plugged in. The next thing I tried was turning the machine on myself with the battery connected (no AC adapter). The blue ring around the power button turned blue for a few seconds then went dark. I didn't see anything on the screen at all when the blue light was lit. Next, I took the battery out and tried to power on with the AC adapter plugged in. I pressed the power button and got absolutely nothing. No blue light like when the battery was connected. My troubleshooting thus far led me to believe that the machine may have a faulty DC jack so I tore the laptop down and pulled out the jack and tested it with my meter with the AC adapter plugged in. The output was what it should be (19.2v). At this point I am stumped. My instinct tells me I may have a bad mobo on my hands but I am still a beginner and I don't trust my instincts all that much just yet so I figured that I'd reach out to my fellow techs. I plan on swapping the RAM for known good but I don't think that is the issue considering I didn't get any beeps or video output that would indicate that it was a RAM issue. One of the board's caps looks ever so slightly more bulgy than the other three but I'm not sure. Thanks in advance for any help/replies.

 

[Martyn]

Hello all. I have an HP Probook 4510s on the bench.
Specs: 4 GB DDR 3 RAM
Intel core 2 duo
250 GB Hitachi HDD

I ran an extended self-test with gsmartcontrol on the HDD it passed. I have been unable to test any other components as you will see below.

Customer informed me that the machine wouldn't properly power on. The first thing that I did was test the AC adapter with my meter and the output was what it should be. I noticed that the power LED was not lit when the AC adapter was plugged in. The next thing I tried was turning the machine on myself with the battery connected (no AC adapter). The blue ring around the power button turned blue for a few seconds then went dark. I didn't see anything on the screen at all when the blue light was lit. Next, I took the battery out and tried to power on with the AC adapter plugged in. I pressed the power button and got absolutely nothing. No blue light like when the battery was connected. My troubleshooting thus far led me to believe that the machine may have a faulty DC jack so I tore the laptop down and pulled out the jack and tested it with my meter with the AC adapter plugged in. The output was what it should be (19.2v). At this point I am stumped. My instinct tells me I may have a bad mobo on my hands but I am still a beginner and I don't trust my instincts all that much just yet so I figured that I'd reach out to my fellow techs. I plan on swapping the RAM for known good but I don't think that is the issue considering I didn't get any beeps or video output that would indicate that it was a RAM issue. One of the board's caps looks ever so slightly more bulgy than the other three but I'm not sure. Thanks in advance for any help/replies.

If I'm reading it right the power supply is clamping down in a fault condition under load. When you pulled the power jack it looks like you were testing with no load is that correct? I would pull out everything you can and see if the power supply lamp will stay on.

 

[Encrypted Existence]

If I'm reading it right the power supply is clamping down in a fault condition under load. When you pulled the power jack it looks like you were testing with no load is that correct? I would pull out everything you can and see if the power supply lamp will stay on.

Yes, you are correct. I just tested the jack while it was connected to mobo with nothing but CPU installed. Still no light. I also went ahead and tested the pins (or leads?) coming from the dc jack connector going into the mobo. I am getting 19.2v. Should I take out the CPU as well and then test again?

 

[Martyn]

Yes, you are correct. I just tested the jack while it was connected to mobo with nothing but CPU installed. Still no light. I also went ahead and tested the pins coming from the dc jack connector going into the mobo. I am getting 19.2v. Should I take out the CPU as well and then test again?

So the light is on with no load and goes dim or off connected? If it goes off you should measure low or no voltage across the pins you were measuring before. Check the usbs i can't remember the exact symptoms but i had one a while ago where the plastic separator had broken and the usb was shorting out.

 

[NYJimbo]

I hate the newer HP jacks. Too many pins and like on this one you have TWO harnesses, one is a 4 wire and the other is the smaller 3 wire.

If this thing is totally dead, why not hot air it and see if it comes back to life. I'm not saying it will be fixed but if they board is dead you can play with it just for future reference.

 

[Encrypted Existence]

So the light is on with no load and goes dim or off connected? If it goes off you should measure low or no voltage across the pins you were measuring before. Check the usbs i can't remember the exact symptoms but i had one a while ago where the plastic separator had broken and the usb was shorting out.

The light has never come on. It stays off whether there is a load (jack is connected to mobo) or no load at all (jack is not connected to mobo at all). Is there anything left for me to do (test) before I can confidently tell my customer the mobo is dead? Thanks.

 

[Martyn]

The light has never come on. It stays off whether there is a load (jack is connected to mobo) or no load at all (jack is not connected to mobo at all). Is there anything left for me to do (test) before I can confidently tell my customer the mobo is dead? Thanks.

Sorry I misunderstood I though you were talking about the light on the power supply block.

 

[Encrypted Existence]

I hate the newer HP jacks. Too many pins and like on this one you have TWO harnesses, one is a 4 wire and the other is the smaller 3 wire.

I tested for power on the four wire harness only. Should I test the three wire harness as well? As far as hot airing the board...I don't have a gun or a rework station. I plan to get one (station) though.

 

[Encrypted Existence]

Sorry I misunderstood I though you were talking about the light on the power supply block.

I just re read my OP and I def wasn't clear enough. I was referring to the light on the jack itself.

 

[NYJimbo]

I just re read my OP and I def wasn't clear enough. I was referring to the light on the jack itself.

In theory the light on the jack is supposed to be an indicator that positive is getting to the motherboard and its being fed back to the jack positive connection on the LED. On most LED jacks the neg line of the led is a local loop back in the harness. Having said that I have some HP jacks that show the LED on even if you connect the AC Adaptor to the jack/harness assembly without it being installed onto the motherboard. Why the f-ck is that ?

I have to wonder why the laptop makers like HP and Dell just keep changing everything, making jacks so complicated and when something fails nobody really has a 100% way of knowing what the hell to do next. If Sony, Acer, Gateway and other can use two or four wire simple jacks why does HP/Compaq and Dell have to keep expanding. Have you seen the 11 and 13 pin jacks on some newer models ? Somebody tell me why the hell they exist ?

 

[ectta]

back to the basic first

If I'm reading it right the power supply is clamping down in a fault condition under load. When you pulled the power jack it looks like you were testing with no load is that correct? I would pull out everything you can and see if the power supply lamp will stay on.

use a regulated power supply about 30v 5A one , plug in , see any reading first!
check the "common point" voltage with the power supply on. if voltage not found, it's should be related to the pre-"common point" ckt(isolation ckt). the dc jack ckt , eye on those pmos of nmos to see whether "half" working or not?
that's chip-level repairing basic!

 

[Encrypted Existence]

Update...

Well I figured I'd give an update on this machine...turns out that the AC adapter AND the mobo were both bad (I am guessing power surge?). Make a long story short I ordered a replacement mobo off of ebay (used). I swap mobos (reapplied arctic silver thermal paste on the new one) and I am using my sparkle universal AC adapter to power the machine. All goes incredibly well for a short time. I installed SP1 (Win 7) on the machine along with a few other updates and the machine was running great. I had plans this evening so I figured "what better time than now to do a burn in test?". I started the burn in test (passmark) and babysat it for a few moments and all seemed to be going just fine. I left the house for about 4 1/2 hours and came back and the machine was off completely. I hit the power button and the lights came on as if it was going to power on but no video...no HDD activity light blinking...just solid lights (I could hear the fan spinning). I then held the power button down to shut the machine off. After that I tried powering the machine back on but I get absolutely nothing. I not going to mess with it any further tonight and start with a (hopefully) new perspective tomorrow. I am thinking that the burn in test may be what caused the board to fail. Any thoughts? I am very discouraged at the moment and I am considering not doing any type of computer repair anymore as this is just a side gig for me and I don't do much business. Seems that headaches like these aren't worth the money at all. Not to mention that I get to look like a dumb ass giving my customer back the money that they have given me. Again, I am frustrated at the moment but who knows what tomorrow will bring. Thanks.

 

[PCX]

Well I figured I'd give an update on this machine...turns out that the AC adapter AND the mobo were both bad (I am guessing power surge?). Make a long story short I ordered a replacement mobo off of ebay (used). I swap mobos (reapplied arctic silver thermal paste on the new one) and I am using my sparkle universal AC adapter to power the machine. All goes incredibly well for a short time. I installed SP1 (Win 7) on the machine along with a few other updates and the machine was running great. I had plans this evening so I figured "what better time than now to do a burn in test?". I started the burn in test (passmark) and babysat it for a few moments and all seemed to be going just fine. I left the house for about 4 1/2 hours and came back and the machine was off completely. I hit the power button and the lights came on as if it was going to power on but no video...no HDD activity light blinking...just solid lights (I could hear the fan spinning). I then held the power button down to shut the machine off. After that I tried powering the machine back on but I get absolutely nothing. I not going to mess with it any further tonight and start with a (hopefully) new perspective tomorrow. I am thinking that the burn in test may be what caused the board to fail. Any thoughts? I am very discouraged at the moment and I am considering not doing any type of computer repair anymore as this is just a side gig for me and I don't do much business. Seems that headaches like these aren't worth the money at all. Not to mention that I get to look like a dumb ass giving my customer back the money that they have given me. Again, I am frustrated at the moment but who knows what tomorrow will bring. Thanks.

Someone correct me if I am wrong, but your not really supposed to run load / stress tests for that long. You can kill a new board that way if It is not properly cooled. Anyways, the rule of thumb that I went by (from researching this matter in the past) was 45 - 90 minutes and that is while keeping an eye on the temperature. I have heard other doing it for a few hours, but you are really taking a chance then. You got to remember, the chances of your customer stressing out that computer in the same manner for that long period of a time if extremely unlikely. Finally, I suggest never using used boards for repairs.

 

[red12049]

Someone correct me if I am wrong, but your not really supposed to run load / stress tests for that long. You can kill a new board that way if It is not properly cooled. Anyways, the rule of thumb that I went by (from researching this matter in the past) was 45 - 90 minutes and that is while keeping an eye on the temperature. I have heard other doing it for a few hours, but you are really taking a chance then. You got to remember, the chances of your customer stressing out that computer in the same manner for that long period of a time if extremely unlikely. Finally, I suggest never using used boards for repairs.

Back in the old days, the "gold standard" for burn in testing on a completed system was twenty four hours. Most all system makers advertised that.

No idea what the standard is now, but I doubt that forty five to ninety minutes will accomplish much, although you probably should stay around and monitor temps at least that long.

Agree on the used boards, except in unusual circumstances.

Rick

 

[PCX]

Back in the old days, the "gold standard" for burn in testing on a completed system was twenty four hours. Most all system makers advertised that.

No idea what the standard is now, but I doubt that forty five to ninety minutes will accomplish much, although you probably should stay around and monitor temps at least that long.

Agree on the used boards, except in unusual circumstances.

Rick

I guess it depends on the purpose for running the test? When I do load / stress tests, I usually do it for the purpose of checking for overheating and stability. 45 - 90 min is usually long enough to tell you if there are overheating issues. You can then fix the problem and then run the test again. Other than running these tests for over heating issues, what else would you really run them for other then to check for long term stability?

 

[Encrypted Existence]

Someone correct me if I am wrong, but your not really supposed to run load / stress tests for that long. You can kill a new board that way if It is not properly cooled. Anyways, the rule of thumb that I went by (from researching this matter in the past) was 45 - 90 minutes and that is while keeping an eye on the temperature. I have heard other doing it for a few hours, but you are really taking a chance then. You got to remember, the chances of your customer stressing out that computer in the same manner for that long period of a time if extremely unlikely. Finally, I suggest never using used boards for repairs.

The test that I ran was a basic 15 min. test (I am running it on my personal laptop now).

 

[PCX]

The test that I ran was a basic 15 min. test (I am running it on my personal laptop now).

ah, got it. Well, at least now you know.

 

[red12049]

I guess it depends on the purpose for running the test? When I do load / stress tests, I usually do it for the purpose of checking for overheating and stability. 45 - 90 min is usually long enough to tell you if there are overheating issues. You can then fix the problem and then run the test again. Other than running these tests for over heating issues, what else would you really run them for other then to check for long term stability?

On new or refurb systems, early failures, or imminent failures. On other systems, random, unrepeatable errors. I have a Toshiba laptop in now that the screen randomly goes grey, according to the customer. Running HW test on it to stress the video.

This may help a bit: http://en.wikipedia.org/wiki/Burn-in

Rick

 

[pc142]

I know it's discouraging but don't quit now. I've replaced a number of motherboards with the "refurbished" ones off of ebay. I offer customers to replace with new or refurbished. They almost always choose the refurbs. Only had two bad ones but found the sellers were more than happy to replace the failed board. Success is sweet! If you choose to quit do it after a success not a failure. Good luck.

 

[Encrypted Existence]

I know it's discouraging but don't quit now. I've replaced a number of motherboards with the "refurbished" ones off of ebay. I offer customers to replace with new or refurbished. They almost always choose the refurbs. Only had two bad ones but found the sellers were more than happy to replace the failed board. Success is sweet! If you choose to quit do it after a success not a failure. Good luck.

Thanks for the encouragement. I will contact the seller about a return/refund.