Precision Workstation 650 Beeping - Power Suply Fan?

I worked on the 650 platform, briefly and with few samples, but I never saw this issue so I can't define it for you.  You have visually confirmed that the fans are operating.  This points in the direction of the conclusion that the alert is extraneous and that there is not actually anything wrong, except the annoyance.

One thing that happens as connectors age, is that low-current signals like the fan monitor signal can build up contaminants on the connector from the air which become resistive enough to block the signal intermittently.  Reseat the connector from the PS to the MB.  If that makes no difference, another step would be reseating the fan connector within the power supply.  That would necessitate disassembly though.

It is uncommon for the fan itself to intermittently stop sending the fan monitor signal.  Typically if that signal fails it remains failed.  But anything is possible.

Thanks for the reply.  I'll give it a try.  I don't mind pulling the PS to reseat the fan connector.  As long as the beeping stops.

When I viewed the event log in the CMOS, this fan error message had been happening since 2003, except that it didn't beep while working in Windows.

Maybe I'm just looking for something to be wrong, but on the visual inspection of the PS fans, one fan could've seemed to be running slower than the other, although i don't know which one is FAN 1 (right or left side).

I'm running 2 HDD totalling 1TB and maxed out on RAM with 4GB.  The last thing I need is to blow the system due to fan failure.  I've even got the Primary HDD in a 5.25" coolermaster enclosure just in case of overheating.

Is there a page in the BIOS SETUP that would tell me the status of the fans?  Someone at work who used to use Compaqs said that his SETUP page used to tell him status off all the fans and other items.

Lastly, if it is the PS fan, any idea on the model # of those PS fans?  Can't find anything in the online manuals?  It must be a standard fan size that fits in there and not a DELL specific dimension.

Fan logs predate beeping?  They may be unrelated.  My Dimension beeps randomly, coincident with the use of arrow keys, and always has (for 9 years).  Pretty easy to ignore, since nothing bad happens.

I only saw a small sample of 650s, most of my time was spent on the bulk products, Dim and O'plex.  I didn't even know that the chipset kept track of PS fans, it doesn't in D or O.  One way to test if those fans are actually producing errors and beeps, is to stop them while they're running.  Stick a q-tip into one, wait a couple seconds for it to register, then do the other one.  See if you get beeps, and if you get two separate log entries separated by a few seconds.  Also watch to see that both fans come back up to speed at the same rate.  If one has a bearing problem, it will startup more slowly.  Don't be fooled by the restart delay.  When the fans are stalled, the electronics cuts power to them, pauses, then tries again.  If your release coincides with the beginning of a pause, restart will take longer than if you release at the end of a cycle.  (You thought fans were simple?)  If the stalling registers two separate events, you will then know which fan it is calling fan 1.  If stalling the fans does not produce a beep, the beep is originating elsewhere.

Unknown as to whether setup displays realtime hardware status.

The fans are off-the-shelf items chosen by the PS vendor.  They will have a model # label on them, and a manufacturer's name.  You can buy them on the open market.

OK, so it's settled, fans are not causing beeping.  And now you know which one is #1 if you want to replace it.

Diagnostic won't catch it, unless it happens while the diagnostic is looking at the keyboard.  So if it beeps every half hour, you'd have to run the kybd diag for at least that long, and even then it wouldn't define whether it was the port or the kybd at fault.

Keyboards are a LOT cheaper than motherboards, so if you want to try replacing something that would be it.

The round connector is PS2.  It is also serial, but monopolar/single-ended discrete send and receive lines.  USB is bidirectional serial differential (data + and data -).  Both are subject to misinterpretation at the receiving end (MB) and garbling at the sending end (kybd).

Could it be a driver issue for the keyboard?  When I installed the OS, I never installed any drivers for it - just whatever WindowsXP uses.  Maybe the driver went bad from an install of something else...

I've gotta find a spare keyboard to do the test.  Maybe I'll switch to a USB one instead of a PS2.

Standard OS2--or USB for that matter--keyboards do not require a driver to work with XP.  As long as you installed the chipset driver, without which I don't think you would have gotten this far.

Keyboards only require drivers to make special functions--volume controls, etc.--work

ok - here the latest...

I got a USB keyboard and swapped out the old one.

Bad news... Still Beeps.

Back to the old drawing board.

Just to narrow the possible issue... Does a system board beep specifically indicate some type of hardware issue or would it also indicate a software issue (some type of driver conflict or something)?

Another thing that i find interesting... I originally thought it was the system board battery going bad, so I replaced it. I got a POST error on the first boot afterthe battery change to reset the clock settings and stuff (pretty normal stuff). The wierd thing is that the beep after the battery change became a weaker sound.  Kinda like the little speaker was being strangled and couldn't get the full Beep sound thru.  It was a new unopened battery, but it was laying around in my toolbox for I don't know how long.

...maybe the "new" battery was OLD and WEAK after sitting around.  I'm grasping here now, but just maybe the battery dying battery was replaced with a worse battery.

The battery might cause beeping, but I would think that would show up on the log, though not necessarily.  The single beep is non specific.  I don't know why mine beeps when I hit arrow keys.  Never character keys, just arrows.  Did it with 2 different PS2 keyboards.  I don't know why mine does it, and I don't know why yours does it.  I do know that mine never got any worse, so best I know it is not anything to worry about.

just on a hunch, I tried testing the RAM chips.  The 650 uses paired RAM, so I should have 1GB chips in each of the 4 slots.

I initially used Microsoft's RAM diagnostic utility and then found another one called memtest86+. Both runs off a boot floppy.

The MS one found errors in every test run over the course of 6 passes. On the otherhand, the memtest86+ found no errors in I don't know how many passes since I left it to run overnight for 6 hours.  It found none at all.

I haven't tried pulling the RAM Chips to see if any one of them would cause the beeping, but I'm not sure how to do that since they require pairs in the slots.  If one pair does beep, how would I naarrow it down to which one of the two?

Is there a definitive RAM checker Utility?  Not sure why I'm getting 2 opposite results - either all bad or all good???..

Thanks.

I'l give it a try.

if the RAM is error correcting and the beep is the notification of a corrected error, would the RAM just constantly be getting errors and correcting them?

I'm guess the only solution to a bad RAM chip is replacement??

I think I got these Crucial branded RAM maybe a year or a year and a half.  Hopefully there is some kind of warranty on them.

I was told to clear the CMOS report log and also reset/clear the BIOS after the battery swapout.  Something about jumper pins that might have to be pulled for 10seconds and replaced to reset/clear the BIOS because bad batteries can mess up the BIOS. 

Currently, my 650 BIOS is A05

Any idea if this is applies to the 650 Motherboard?

Might make sense to do this then test out each sstick of RAM...

Windows Memory Diagnostic (WMD) is the one I used in the lab.  It would indeed find errors others would overlook.  For example, a system would bluescreen on boot, but the Dell memtest found nothing wrong.  WMD did however.  The explanation I was given is that Windows uses memory differently than the Dell diagnostic does.

There was a provision in the workstation line for a diagnostic mode in which only one memory stick is required, in socket 1.  WMD specifies a location and the exact error, but it may be cryptic.  So, substitute each in socket 1 until you find the one which errors.  Unless it won't run at all that way--not sure how universal that feature is--in which case, number them and rotate each through a single pair until one errors.

Quite likely that is what's happening, as WS use error-correcting RAM, and the beep could signify that an error has been caught and corrected.  Though it would be helpful if it logged the event, wouldn't it?

A failing clock battery can cause all kinds of goofy things related to turnon and startup, but it does not cause memory errors.

It doesn't error 'constantly'.  Just on certain data patterns.  It's called cross-coupling.  If you put in all 1s or all 0s, you get them back.  But for example, if you put in 0101, a cross-coupling error would produce 0111.

Memory cannot be repaired, only replaced.