OK - - - I have a notable power on problem with a Dell XPS8900.
* Start button appears to do nothing in all cases.
* Upon connecting power, the computer will "attempt" a power up, fans spin, and about 1 second later everything dies. Upon waiting a period of time (minutes), I get the same result. If I DON'T wait, and reconnect power immediately. . . nothing happens. Note that the power up attempt is happening WITHOUT touching the start button.
* If I remove the CMOS battery and wait a while (minutes), and then install the battery and then power the following happens; The computer does the "power up attempt" three - four times in short succession. On the third or fourth attempt, the PC boots up. I do get a BIOS error for lost clock setting. It **Seems** like the PC is struggling to get data off the hard disk (it is a HDD), and firing up Windows seems to take more time than it used to. Note: Start button is not pressed in this scenario.
* After cycling the installation of the CMOS battery and getting a boot up, I can restart through windows. If I shutdown through windows, and come back to the computer 60 seconds later, it won't boot up (same behavior as noted above).
Troubleshooting to date:
* Replaced the power supply (yikers. . . )(no change in performance)
* Replaced the CMOS battery (first with one from another device, then with a fresh one purchased today) (no change in performance)
* No errors in device manager; HDD and optical disk is accessible (HDD seems slow). No memory error. The video card (GTX1060) seems to be alive. Once running, the computer seems "normal".
Any thoughts? (Yes. . .I have now recovered all necessary data off the machine. . . so I can be brave with experimentation)
Version of Windows?
Can we assume the new PSU has wattage adequate for this system?
If using a power strip, surge protector or UPS, remove those and connect PC directly to a known working outlet.
Did you recently install any updates, eg, BIOS, drivers or other software?
Reboot and immediately press F12. Look for option to run diagnostics and run all of them, especially RAM and extended hard drive tests. Copy error message(s), if any...
Do you have a small (32-GB) SSD in this PC that's used as a cache for Windows, or any other RAID setup with 2 hard drives?
Forum Member since 2004
I am not a Dell employee
Thanks so much for the reply.
I have not run those bios diagnostics. . . I will give that a shot.
As for power cords - - -yes, I have swapped cords and outlets once or twice even before putting in the new PSU. The new PSU is rated for 760w, so I don't think that is the problem.
Software: I am running Win10 from MS installation media; Ran the fall update 2 months ago. The spring update got jammed on me in the last day - - - after the current problems started. I saw it coming, so I prepared. The update ran slow, but ran all the way through. The machine was warm booted 3-4 times during the process. Since the new software went in. . .no change.
Firmware: I think there is new firmware for the 8900 I-5's floating out there? Don't know. . .have not updated the bios.
I will try the diagnostics next time I can. Pretty sure I ran a boot test without the HDD or optical drive; but I can try that again. The RAM is not original, but has been in the system since last July. It "posts" properly . . .but I hear you on diagnostics. Is this a likely culprit? I have the old ram sticks. . . I can swap them out. I even can swap out to a different HD to see what happens.
The power button behavior (i.e. never pressing it) is bugging me.
As an aside;
Don't make any hardware changes until we see what the F12 diagnostics have to say, especially about the HDD... Otherwise you 'll just complicate things even further.
And what was your "As an aside; " ?
Forum Member since 2004
I am not a Dell employee
The "aside" was an editing error on my part. Sorry about that!
I will be at the troubled XPS8900 tonight, and will let you know what the diagnostics turnup.
Thanks for your help.
^- - - my system behavior was just like Mikes.
- - - pulled the CMOS battery and the plug; let the power drain. Reinstalled and booted up WITHOUT the power to either the HDD or DVD drive. No change. . . . .system automatically went through a short 1-2 second power up (and failed) three times. On the fourth time, the system booted to the bios screen and reported "no HDD". . .and wouldn't let me progress with any tests.
- - -pulled the CMOS battery and the plug; let the power drain. Reinstalled CMOS battery, put power to HDD and DVD drive. Upon plugging in. . .. .system automatically went through a short 1-2 second power up (and failed) three times. On the fourth time, the system booted to the bios screen (since I had pulled the CMOS battery) and I ran the diagnostics.
There appears to be two sets of diagnostics.. . . a short set which tests DVD, HDD, RAM etc; - - - > which reported *no problems*. It then gave me the option to run a longer test (estimated time: 9+ hours). I began that test. . . and after 30 minutes, was still running. I will report back results later tonight.
An out of the box thought: I see that BIOS is set to go to "last state" upon restoration of power. (i.e. restart if the PC was running; stay off it the PC was not running). I wonder if there is a counter mis-set, or not getting reset. I have not had a normal "push start button and PC boots up" since this mess started. The times I have gotten windows to load was AFTER going through the CMOS battery pull routine, going through the BIOS screen, and letting the system boot from there. I will try setting this BIOS option to boot the machine upon restoring system power.
The plot thickens:
Shortstory: There is a Bios setting that controls behavior after power failure (i.e. pulling the plug). I changed this from "off" to "on". (There is also a "last state" option). This completely changes behavior; and allows boot (if capacitors are drained).
**NOTE: I have not been in the BIOS for a very long time before this**
0) Looking into the drivers, I suspect I have not updated any dell drivers since I loaded Windows 10 in late 2015. Current firmware is 2.0.1. (Dell support software was not loaded until today). From reading other threads. . . I think I will stay away from 2.2.1
1) I have now completed the UEFI BIOS diagnostics that I started this morning. They all passed. I also ran HDD diagnostics from the HDD manufacturer: HDD passed.
2) Did more diagnostics on power behavior; With AC power fail set to "Off"
2a) The computer will do a warm reboot fine (from startmenu, click restart, computer shutsdown and reboots fine)
2b) The computer will not do a cold reboot (From startmenu, click shutdown: Computer shuts down but can't be restarted normally. Must pull CMOS battery)
2c) The computer will LAUGH at a hard reboot. (With computer running, push power button, *no discernable impact whatsoever*)
3) Did more diagnostics on power behavior; With AC power fail set to "On"
3a) The computer will do a warm reboot fine.
3b) The computer will not do a cold reboot, *unless you give time to allow power to drain from capacitors* (From startmenu, click shutdown: Computer shuts down but can't be restarted for a period of time. . . CMOS battery can remain in place). Note: I don't think the startbutton works in this scenario: You must pull plug on the PC to force startup.
4c) The computer will LAUGH at a hard reboot. (With computer running, push power button, *no discernable impact whatsoever*)
4d) Computer will reboot after you pull the plug.
Good Gosh?? This is . . . unsettling.
I believe I can make do with the current state of affairs. . the machine runs. The new power supply I bought actually has an on/off on it. . .so I don't really have to "pull the plug". So my procedure is to "shutdown normally, then turnoff the power supply". It will restart upon turning on the power supply??
The on/off button is now useless?
I have been prowling google. I found another thread from 2017 in this forum. . . describes the same odd power button behavior. They called it a motherboard flaw. Not sure I disagree.
Should I consider a bios upgrade? I won't do the current 2.2.1 version (too many threads containing the word "brick"). . .but I would consider an older version (still newer than the 2.0.1 I am running).
I was strongly considering a SSD upgrade to the system. . . .It serves me well. Now I am not sure I want to invest in this box.
Other thoughts appreciated.
Don't update BIOS now. You'll only complicate things even more...
Have you checked the cable connections from the power-button connector (F_PANEL1) (#9 on page 16, here) to the front panel, at both ends? And I guess it's possible that cable, or the power button or the front I/O panel has malfunctioned.
Are you sure the CMOS battery is any good, even if recently replaced?
Have you checked the Power Button settings options in Windows to make sure the button is enabled, and not set to "do nothing" or "sleep"?
Might not be surprising you have to wait a bit between turning the PSU off and then on again before the Auto Reboot after power failure will sense the outage and kick in.
FWIW, I have my Dell desktop connected to a power strip with its own power switch and BIOS is set to Auto Reboot after a power failure. After shutting down normally in Windows, I turn the strip off which also turns off the monitor and printers. When I turn the strip on, everything powers up and PC boots, all at the same time with just that one switch. So what you're doing is a very reasonable workaround, and I'm just being lazy.
Forum Member since 2004
I am not a Dell employee
Well. . . .
1) I just saw the observation about the power button cable in another post (of yours) about an hour ago. I checked the connection on the Motherboard - its fine. Have not taken the drive bays apart to check the other side where the button side is connected. I probably will in the next few days. Frankly, I have spent too much time trouble shooting in the last few days. I need to do other things 😉
2) Interesting thought on the power button settings. I didn't think of that. (checks settings) They are Not changed. That wasn't it. Wouldn't have surprised me, to be honest.
3) I hear you on not updating bios. I am frankly scared to try it now. I have this sneaking suspicion a setting got changed in my Bios without me knowing it. . .I probably want to prowl through there a bit more.
4) The real bothersome part is if I do a *normal* windows shutdown, the machine can't be immediately restarted. But I guess I will live with that.
5) Yeah, I am sure I have a good CMOS battery. I pulled the second one from another device; then went to the store and bought a FRESH one. I think I want to rule out the battery. (BTW: I had a problem on a Gas Turbine once. . . .the manufacturer has us change out a fuel valve THREE TIMES before we could convince them it wasn't the valve. We ultimately found the problem . . . .it was NOT the valve 😉 )
Thanks for your help. I do appreciate the effort. Hopefully, you or someone else will think of more bright ideas.