CPU upgrade, external monitor no longer works at BIOS level

Dell LATITUDE D630
September 26, 2013 at 19:28:37
Specs: Windows 7 Ultimate, Intel dual core 2GHz 3,00GB
I've replaced the Intel Core2 Duo T7300 @ 2.00GHz CPU in my Dell latitude D630 laptop with the Intel Core 2 Duo T7800 2.6 GHz Dual-Core Processor(a compatible upgrade, according to Dell documentation). Every thing seemed have gone just fine until I hooked it up my docking station. The CRT external monitor no longer works at all at the BIOS level. I cant view the boot process on the laptop screen but the first thing I see on the external monitor is The Windows 7 Welcome Screen. It works all right in Windows except that I can't get various resolution to work the way they used to using the controls on the monitor (can't stretch to edges of screen). I get the same result whether I plug the CRT into the docking station or directly into the laptop. I don't see any BIOS settings that would be relevant. My BIOS version is A17. There is an update available but from what I've I read, it seems very unlikely to make a difference. Obviously, though, there is something happening at the BIOS level. I'm completely baffled. Does anyone have any idea what might be going on?

message edited by krh


See More: CPU upgrade, external monitor no longer works at BIOS level

Report •


#1
September 26, 2013 at 19:39:39
In Device Manager, are there any devices with yellow exclamation points or anything like that? Are the video drivers up to date?

I don't believe that you'll need to flash the BIOS update, as you have said, both processors were originally available as options when the machine came out so a BIOS update shouldn't be required to make the processor work correctly.

You've been helped by a 15 year old.


Report •

#2
September 26, 2013 at 19:50:13
I cannot see that drivers would have anything to do with what happens before you get into Windows since they do not load until after Windows loads.
I doubt a BIOS setting would have changed by itself since it worked before and a BIOS update would not make much sense either.
In order to change out the CPU in a laptop, you had to disassemble a lot of components, is it possible that there was minor damage to something on the motherboard, the external video port, or a connection or ribbon cable that was not completely inserted. It could even be a wire or ribbon that is being pinched during reassembly. Even possibly extra thermal compound squeezing out from under the heat sink, could have caused a problem, depending on the thermal compound you used (conductive or non-conductive).

You have to be a little bit crazy to keep you from going insane.


Report •

#3
September 27, 2013 at 15:28:27
First, I have something very strange to report. As I said in my original post, my BIOS version was A17, but the more I thought about it, the more certain I was that before the CPU upgrade, it was A18. It's hard to imagine such a thing, but it seems that my version somehow got rolled back! (Possibly when I unplugged the CMOS battery?) At any rate, I did upgrade to version A18, with no change in behavior.

There are no in indications of any problems in Device Manager and according to Windows (Win 7 ultimate 64 bit) my graphics driver is up to date. Please disregard any reference I made to the monitor not working in Windows the way it did before. I simply had to correct my settings. Once the image appears at the Windows Welcome screen, the monitor behaves perfectly normally. It's as if there's simply no connection at the BIOS level (and I'm sure that no Windows setting could affect that).

It's occurred to me also that there may be a problem with the way I reconnected wires when I put the computer back together, although the fact that the monitor works normally in Windows makes me wonder about that (especially damage to the motherboard). This weekend, I'll go back into the PC and check my connections.

There is one other strange thing. In Setup, the graphics controller is identified as "Intel Crestline Graphics". I can't remember what it said before, but I'd swear it was something else. I'll do some research on that to see if there's a firmware update or something. If I could have anticipated this kind of trouble, I would have taken more note of BIOS entries before I started.

I used Arctic 5, which I believe is conductive, as thermal paste, but it's unlikely that any slopped over onto the MB, since I used only a less-than pea-sized dab in the center of the CPU.


Report •

Related Solutions

#4
September 27, 2013 at 21:08:27
It certainly sounds like you did everything right. Keep us updated on what you find. Right now, I cannot think of anything else it might be, but if any other clues turn up, let us know because maybe it will strike a memory in someone that might help you solve this.

You have to be a little bit crazy to keep you from going insane.


Report •

#5
September 28, 2013 at 12:46:00
I hope you got the T7800 really cheap because I doubt you'll notice much of a difference except until certain circumstances. The proper thermal paste application for mobile CPUs is the 'surface spread' method: http://www.arcticsilver.com/pdf/app...

A BIOS has to be flashed to another version, there is no way it can roll back on it's own, so you must have had A17 the whole time. The latest is A19: http://www.dell.com/support/drivers...

As for your display problem, are you saying the Dell splash screen is no longer displayed? Have you tried pressing any keys at startup - Esc, Tab, Pause, etc to see if it will bring up the splash or POST screen? Did you change any BIOS settings that may have disabled the POST screen from being displayed?

"the graphics controller is identified as "Intel Crestline Graphics""

That's what you have, Intel GMA X3100, code name: Crestline. It's being reported correctly.

http://ark.intel.com/products/coden...

http://www.notebookcheck.net/Intel-...

Here's your laptop specs: http://www.dell.com/us/dfb/p/latitu...


Report •

#6
September 28, 2013 at 15:16:04
I upgraded the CPU in hopes that the added power would improve the performance of some of my peripheral devices and it seems to have done that. My external DVD writer, which was behaving very erratically, now works perfectly. Most USB devices now require a minimum of 2.4 GHz. I bought the DVD writer because it allegedly only required 2 GHz, but it really didn't work right until I upgraded.

I would agree that BIOS can't roll back on it's own, based on my experience--but my experience until now has been mostly with old legacy machines. My most recent experience updating BIOS was with our old Gateway 2000. But I'd swear that it did! When I got the machine, it had A18. I was avoiding an upgrade to A19 because I didn't seem to need it and I had read negative things about it. I used the non spread method for applying the thermal paste because of relative pro and cons I had read. Perhaps I should have spread, but I doubt if its a factor in my problem

My display problem is that nothing appears on my external monitor during boot-up except a "no sync" message at the bottom (which, of course, conveys no information except that there's no signal) until the Windows Welcome Screen appears. At that point, the monitor crackles and comes to life. After that, I log into Windows and everything is normal. If I leave the lid of the laptop, which I'm using as a desktop, open, I can watch the whole boot process, from the Dell splash screen, on the laptop screen, so there's nothing in Setup hiding the POST (such a setting doesn't even seem to be available). I should probably mention that my external monitor is an old CRT plugged into the VGA connector on my docking station, although the behavior is the same if I plug it directly into the VGA connector on the laptop. One of the few configurable settings in Setup that seems even remotely related to the monitor is the default controller, which offers the options of docking station or on board. The default is docking station, but I've tried both with the same result. I should also stress that the problem only started after the CPU upgrade.

I accept the assertion that "Intel Crestline Graphics" must have been there before. It's not something that would have been significant to me at the time.

It's as if there's no support for the external monitor until the Windows driver loads, strange as that seems. Pressing the key-stroke combination for switching monitors (<fn> <f8>) does switch the display from the laptop to the external monitor before the appearance of the Welcome Screen, at the boot menu, but that only works after the hard drive has been accessed. And of course, as I say, it's a brand new problem. Judging from the lack of problems once Windows starts, It seems that both the computer and the monitor are OK, and I just don't see anything at all in Setup that could affect the situation. In your opinion, Is there something I might have done in disassembling and reassembling the computer that could have caused it? I need to go back in and check the wiring, but since the computer seems to be working fine otherwise (better than ever) I'd like to do as much as I can to see if there's another explanation.

Thanks for the links. I'll look through it all.

message edited by krh


Report •

#7
September 28, 2013 at 16:43:33
I have a D630 of my own, running BIOS A18. I tried putting it in a docking station and using a VGA cable to hook it up to an external monitor, and with the lid closed the BIOS and boot process was displayed on the external monitor. However, with the lid open, there was no video output to the external monitor. Check to make sure all of the cables from the LCD are attached correctly, as some of the cables that indicate whether the lid is closed or not may be lose.

You've been helped by a 15 year old.


Report •

#8
September 28, 2013 at 23:10:54
"Check to make sure all of the cables from the LCD are attached correctly"

Are you referring to wiring inside the laptop (and not the VGA connector)? If so, and if there really is supposed to be a signal that tells the system whether or not the lid is open or closed, it would certainly be worth while to go back inside the box and check the wiring. I can't remember the exact behavior before I swapped out the CPU, whether or not I ever tried to boot with the external monitor attached and the lid open. I hope I didn't break anything. I've built several desktops but this was the first time I'd ever gone inside a laptop. Hopefully, checking all the wires and re-seating all connections will help.


Report •

#9
September 29, 2013 at 13:21:19
Yeah I'm referring specifically to the ribbon display cable attached to the right side of the motherboard (not the black and white wires going to the WiFi card).

You've been helped by a 15 year old.


Report •

#10
September 29, 2013 at 15:24:47
#2 above:
"In order to change out the CPU in a laptop, you had to disassemble a lot of components, is it possible that there was minor damage to something on the motherboard, the external video port, or a connection or ribbon cable that was not completely inserted. It could even be a wire or ribbon that is being pinched during reassembly."

You have to be a little bit crazy to keep you from going insane.


Report •

#11
September 30, 2013 at 15:56:07
I went in and checked my wiring and connections. Everything looked fine except that at a point in the routing channel right next to the connector for the LCD, the WiFi wiring passed under the LCD cable. Thinking that that might be preventing the LCD connector from seating properly, I changed it, but it had no effect. So I went back in, disconnected everything, removed the display, reseated it, and re-routed and reconnected all the wires. Everything looked good but it still doesn't work. At this point, all I can think of is either some subtle, very selective damage to the motherboard or some kind of damage to the LCD cable. Replacing that would be pretty drastic just for the sake of troubleshooting. If I knew what kind of mechanism tells the computer the lid is closed, it would be very helpful. That seems to be where the problem lies (or maybe not).

The problem occurs whether the monitor is attached to the VGA port on the docking station or directly to the one on the computer, so the port is ruled out.

Late edit: There's one other thing that needs to be mentioned, something that had slipped my mind. Replacing the CPU involves turning a locking screw 180 degrees, then turning it back again to re-lock it. After seating the chip, the screw would only turn back about 3/4 of the way before encountering resistance, so I left it that way. In retrospect, I should have tried taking it out and re-seating it. It seems strange that if the chip isn't properly seated, it would result in just one narrow problem, but there might be others I haven't found out about yet. I hate to do it because it means cleaning the chip and risking damage, but I guess I'm pretty much obliged to try it, aren't I?

message edited by krh


Report •

#12
September 30, 2013 at 20:14:32
Yes, it shouldn't have even booted without the processor fully secured. You're gonna want to tighten it.

You've been helped by a 15 year old.


Report •

#13
October 2, 2013 at 19:07:02
I reseated the chip and also tried another monitor, but no change. The only other trouble-shooting measure I can think of is re-installing the old CPU to see what happens. It might take awhile to report back on how that goes.

Report •

#14
October 5, 2013 at 15:16:57
Well, I re-installed the old CPU but it didn't restore the old behavior. Apparently, I did damage something when I initially worked on it, though I can't imagine how. I was gentle and didn't touch any circuitry. I'll continue to do some research and see if replacing the LCD cable might help. Maybe I damaged that taking it in and out of the routing channel. What stumps me is how the problem can be so isolated and singular. Once into Windows, everything works fine. In the meantime, I guess I'll just have to live with it, using the laptop screen when I need to get into Setup. I am able to at least access my boot menu by continuously hitting <scroll lock> <f8>, but that's a kind of hit and miss (literally, in the dark). It's easier to just open the computer lid for that, too,

Thanks to everyone for your help.


Report •

#15
October 6, 2013 at 04:25:48
At least it is something you can live with. One more thought that would be impractical to prove or disprove is a small ESD (electrostatic discharge) from you or a surface you placed a component on when setting something aside.

You have to be a little bit crazy to keep you from going insane.


Report •

#16
October 7, 2013 at 13:00:01
I thought I was done here but there's been a new development. I couldn't help thinking about RMT2's observation that the POST will only display on the monitor if the laptop lid is closed. I began to think that the heart of the problem might be that the computer doesn't know the lid is closed, so I tried a simple test. In power options in Windows, I chose "shutdown" for what happens when the lid is closed, then closed the lid. Nothing happened! I tried it both docked and undocked, on AC power and battery, with the same result. It does seem, indeed, that the computer doesn't know when the lid is closed. Now I need to find out exactly what it is that lets the computer know it's closed. Is there a light sensor or a gravity sensor? Is there a hidden button? Is it the position of the hinges? If it is just a matter of a wire in the LCD cable (and, hopefully, not a problem with the connector on the MB), then the cable is relatively cheap and seems fairly easy to replace. Since this is starting to look more and more brand specific, I'll have to try the Dell forum also, but I've always liked this forum, and I hope someone here has some info.

Report •

#17
October 7, 2013 at 14:58:39
I have replaced the LCD panel on an HP laptop and I do not remember any switch there so I reread those parts of the manual and there does not appear to be a separate component switch for this or any wires from the hinges and I doubt that they are using an angle or inertia switch. I would assume that it is part of another component like the switch board (power on/off button) which on the HP is right up near the hinge for the top on the right and appears to have a 'leg' that goes up near the bottom of the monitor. If the switch board is the right component, and it is even slightly out of place then this might prevent the lid closing from making the signal to the motherboard, it does not have to be a broken or disconnected wire in this case. That i where I would look first if it was me.

You have to be a little bit crazy to keep you from going insane.


Report •

#18
October 7, 2013 at 15:37:52
I Googled and also search the Dell forums and the most common answer seems to be a magnet located some where on the lid assembly that interacts with something on the motherboard. The only magnet I'm able to locate, using a paper clip, is under the bezel directly under the lid latch. It actually seems to be part of the latch itself, part of the "spring action". I'm going to have to go back into the computer to see whats on the MB opposite that magnet. I'll report back when I find out.

Here's a link to my post in the Dell forum:

Need to find and fix lid sensor on Latitude D630 - Laptop General Hardware Forum - Laptop - Dell Community
http://en.community.dell.com/suppor...


Report •

#19
October 14, 2013 at 17:48:39
OK, here's where I'm at. A magnet on the laptop display interacts with a reed switch on the body of the computer to communicate that the lid is closed. According to several sites selling the palmrest, it includes "coin cell, palmrest, and reed switch" so that switch seems to on the palmrest. I took some pictures of the relevant areas of the computer.

The first one shows the front edge of the underside of the palmrest lying under the top edge of the display (the gray object on the bezel of the screen is a row of steel staples sticking to the magnet underneath). As can be seen, the area of the palmrest that would be directly under the magnet with the lid closed contains nothing at all. If the reed switch is under that plate covering the mouse pad buttons, perhaps the magnet affects it by creating a magnetic field adjacent to it? (The steel plate can be removed but I'm reluctant to mess with it).

http://imageshack.us/a/img36/34/cms...

The second picture shows the front edge of the top of the palmrest and the corresponding area on the motherboard underneath. Again, nothing under the magnet but the battery compartment.

http://imageshack.us/a/img833/8620/...

Since a reed switch is activated by a magnet, it has to be somewhere in the area opposite that magnet. I should go back in and remove that plate to see exactly what's under it and, unless someone knows the answer, I guess I'll have to.
I really don't see how I could have damaged anything in that area by removing and replacing the palmrest except possibly the connector from the palmrest to the motherboard. If that's the problem, the damage could be either on the palmrest or the motherboard side, so replacing the palmrest would give me only a 50-50 chance of fixing it. Although I doubt that it plays any part in the process, I suppose the problem could still be in the LCD cable. Or some tracery on the motherboard. I hate to think that the only way to troubleshoot this is to keep replacing parts.


Report •

#20
October 15, 2013 at 05:41:29
Quite a number of years ago I made something using a magnetic reed switch that I purchased at Radio Shack. This reed switch was a double ended glass vial an inch or inch and a half long, a little thicker in diameter than a USB cord, and had a flat metal contact at each end to solder to. Inside the clear glass you could see the two metal strips go past each other with a small space between. When you passed a magnet near the glass, they stuck together, completing the circuit, making a light tick sound when the touched.
Your switch may be smaller than that one (about 1976) which for about 3 months resided in the narrow space in a pinball machine between the top and bottom at college and a pocket sized magnet could pop off a free game in the right place. It started with three of us knowing about it but eventually the machine made no money so they had to find it. It took one man almost 3 hours to find it. We did not tear off the panel that exposed the area the wires passed through (that was someone who took night classes), we just replaced it and made sure the schematics were where they always were kept.

You have to be a little bit crazy to keep you from going insane.


Report •

#21
October 15, 2013 at 16:46:59
I wrote that I didn't see how I could have damaged anything in the area opposite the magnet, but upon closer reflection, I do see how! At one point, while initially removing the palmrest, I remember briefly prying a little with a flat head screw driver, and I think that it was right in that area. I immediately realized that that was probably not a very good idea and stopped, but now, the most likely explanation for this whole problem seems to be that the reed switch is under that metal plate covering the mouse buttons and that I put enough pressure on the plate in that brief moment with the tip of the screwdriver to damage it. I could remove the plate and hopefully find a crushed glass tube underneath to confirm the diagnosis, but I feel confident enough that that's the problem to go ahead and get a good used palmrest from a reputable seller on eBay. I'd eventually have to try that, anyway.
I'll report back on whether or not that fixes it. If I'd mentioned the screwdriver in the first place, it would have made a diagnosis much simpler, but I'd half forgotten it and it didn't seem significant. At the time, I'd never heard of a reed switch and had no notion of the importance of the computer knowing that the lid is closed in order for the monitor to work.

message edited by krh


Report •

#22
October 28, 2013 at 16:27:52
Bummer! Swapping the palmrest had no effect. Since I don't want to sink a bunch of money into replacing parts just for the sake of troubleshooting, I guess I'll have to basically give up for now. I did find out that the metal plate covering the bottom of the mouse buttons is permanently attached to the circuit board underneath, so undoing the screws just removes the whole assembly. If the reed switch is truly under there somewhere, it must be really small and jammed under or between the mouse buttons somehow.
If I didn't damage the reed switch with that screwdriver, I'm sure I didn't do anything to damage anything else, except possibly the cable connecting the palmrest to the motherboard. I'm not sure if that can be replaced but I'll check it out.
I'm beginning to think the problem may be in the lid somewhere. I don't know if you can see it in the picture I posted, but there is a crack in the bezel of the screen right next to the location of the magnet. That was there when I got the computer and I never had a problem until I disassembled and re-assembled the computer when I originally changed the CPU, but it might have something to do with the issue. Ideally, I'll meet somewhat who has the same computer who'll let me temporarily swap the display to see if that helps. In the meantime, thanks to everyone for their help, especially Fingers, but I guess that if I have to select a best answer (I don't really like that feature of this forum) even though the issue is not resolved, I guess I'd have to choose the observation by RMT2 that the monitor will only display the POST if the laptop lid is closed. That's kind of the key to the problem.

message edited by krh


Report •

#23
October 29, 2013 at 04:53:46
Occasionally, there is no best answer. You do the best you can. Since your system is usable and you can access BIOS and boot options with the built in monitor, it is workable, though admittedly not ideal. We can only hope that someone else reading this will also get some help out of it.
Good Luck

You have to be a little bit crazy to keep you from going insane.


Report •

Ask Question