HP tx2500 hard drive pause

[Th3_uN1Qu3]

I have this HP tx2500 that has been given to me to troubleshoot some power issues. I've been told it works off battery or of AC alone, but it'll lock up if the battery is inserted while on charger, or if the charger is inserted while on battery. Also the side near the power jack gets pretty warm while the battery is charging, even with the laptop OFF. I've been told the charger IC is at fault, and i've also been given a second mainboard to take parts off of, which apparently has a dead GPU.

I was able to see that the charger will not start charging and enter protect mode (LED near the charger blinking, and a ticking noise can be heard from inside the laptop) when the charger is connected with the battery in. If i pull the battery and put it back in, it'll start charging.

I wanted to run some tests on the thing to reproduce the lockup behavior. However, the machine was given to me with no hard drive, so i put one in and attempted to install Windows 7. I can't. When it reaches "expanding windows files" the hard drive STOPS (parks its heads), and it sits there at 0% for a good few minutes, with the mouse cursor hung. After a while it resumes and starts copying. This happens several times at various percentages of file copying. It eventually reaches all the way to the end of the installation (when it's supposed to boot to desktop) when it hangs on a blank screen and stays there.

It does this at random places so i'm sure it's not the HDD, also i've tried running HDD Regenerator and it does the same - it pauses at random areas, sometimes for seconds, sometimes for minutes, then it resumes. Tried 2 drives and it does the same thing, both drives are good tested in other machines.

I'm trying to install XP now and during loading of SATA drivers, it seems to have done the same thing, off the CD... there's definitely something fishy about this board. Any pointers would be welcome. Oh, one more thing. Sometimes when i power it on it'll give the 2 blink error code (BIOS corruption). I'll try to reflash the BIOS, but for that, i need a Windows OS installed as the stupid Insyde flashing utility won't work under DOS...

Edit: "Saving Settings" now at XP install, let's hope it works.

[Tuomo]

Re: HP tx2500 hard drive pause

Do you have schematics?

[opamp]

Re: HP tx2500 hard drive pause

Firts give the southbrigde a reflow to fix the harddrive. If it works update the bios and then troubleshoot the power issue.

[Th3_uN1Qu3]

Re: HP tx2500 hard drive pause

Power issue fixed, it was glaring from the moment i pulled the thing apart. Shorted MOSFET on the DC-in. Also the pauses and the hangs on black screen were due to the NB not the SB, as the last crash was on a green screen with lines - clearly a video issue.

It's been worked on before, the absence of red epoxy was proof of that (the dead board had the red epoxy around the NB, the working one did not). Reflowed NB and it's now working alright.

I also had to go back to the guy who gave it to me (it's not his laptop, he disassembles laptops and sells them for parts, so he's always got stuff floating around) and get a replacement heatsink assembly for this thing, coz the one that was on it was ghetto'd in the worst way... I'll post some pics later.

Right now the board is running out of the case and i'm probing around with my temperature probe. I don't really like what i'm seeing. The heatsink assembly that i got has a copper shim for the GPU, but there's still a thermal pad between that and the heatpipe, and i think that needs replacement. Either way, the TX series is a huge fail in thermal design. The CPU and NB share the same heatpipe, and since the CPU runs so hot, most of the time the heatpipe actually increases the temperature of the NB/IGP instead of cooling it.

[opamp]

Re: HP tx2500 hard drive pause

Power issue fixed, it was glaring from the moment i pulled the thing apart. Shorted MOSFET on the DC-in. Also the pauses and the hangs on black screen were due to the NB not the SB, as the last crash was on a green screen with lines - clearly a video issue.

It's been worked on before, the absence of red epoxy was proof of that (the dead board had the red epoxy around the NB, the working one did not). Reflowed NB and it's now working alright.

I also had to go back to the guy who gave it to me (it's not his laptop, he disassembles laptops and sells them for parts, so he's always got stuff floating around) and get a replacement heatsink assembly for this thing, coz the one that was on it was ghetto'd in the worst way... I'll post some pics later.

Right now the board is running out of the case and i'm probing around with my temperature probe. I don't really like what i'm seeing. The heatsink assembly that i got has a copper shim for the GPU, but there's still a thermal pad between that and the heatpipe, and i think that needs replacement. Either way, the TX series is a huge fail in thermal design. The CPU and NB share the same heatpipe, and since the CPU runs so hot, most of the time the heatpipe actually increases the temperature of the NB/IGP instead of cooling it.

Great you got working, add some locktite to the northbridge to keep it in place otherwise it will do the same thing again once it gets hot and don't forget to post some pics of that heatsink to get a good laugh.

[Th3_uN1Qu3]

Re: HP tx2500 hard drive pause

What kind of loctite do you mean?

Running FurMark with no issues. Had to use 1.7.0 as the latest version lags too badly on this IGP. Maximum measured temperature on top of the NB shim is 58-59C with just FurMark running, 60-61 with IntelBurnTest on top of that. So that confirms my theory that sharing the CPU heatpipe actually hurts the NB instead of helping it.

This NB doesn't have a hardware sensor so i can't find out its actual die temp. It's a few degrees lower on the heatpipe right above the shim, so transfer capability of that thermal pad isn't stellar. Maximum reported temperature of CPU die is 74C. Of course, these temps are with the mainboard out of the chassis, and are likely to climb quite a bit once i put it back in...

I think we got a dud CPU as well. According to HWMonitor, Core 0 is stuck at 1.1GHz, while Core 1 moves all the way from 850MHz to 2.2GHz like it should. Also, XP only reports 560MB RAM with 2x 512MB sticks installed. The Radeon 3200 IGP reports 320MB. It has a 64MB chip of its own and steals 256MB from system RAM, so we should have 768MB usable by Windows, but for some reason we do not.

Oh yeah, and here the pics. Yeah, really. Btw, i'll have to lube the fan that came with the "good" heatsink assembly i got, i don't like how it sounds. But at least it's a normal looking heatsink not like the one i found on it.

[opamp]

Re: HP tx2500 hard drive pause

Holy sh!t!!! now that's ghetto there's nothing that a piece wire and some duck tape can't fix.... definitely needs to be replace. Here's a link for the glue.

http://www.ebay.com/itm/30ml-Red-Glu...item41648888fe

[Th3_uN1Qu3]

Re: HP tx2500 hard drive pause

Has been replaced already, don't worry. Just have to lube the fan, as it sounds a bit like it's rubbing. Thanks for the info on the glue.

[Th3_uN1Qu3]

Re: HP tx2500 hard drive pause

Installed Win7 with no glitches now. Core 0 being stuck at 1.1GHz seemed to have been an XP thing, both cores now go from 550MHz to 2.2GHz under Win7. Unfortunately, that also means the CPU runs hotter, it hits 92C maximum under IntelBurnTest, and this is with the board still out of the chassis. Undervolting this CPU is a must.

Still getting only 560MB (well, 561) of usable RAM tho.