Precision 690 ram issues

There's nothing that really leaps to mind here; it sounds as though you're doing everything right. When you removed the memory risers and installed the RAM in slots 1 and 2, there's a chance that it will overheat since you've moved them out of the air stream, but I suspect that the system would have at least attempted to boot before the memory got too hot... and you're saying you got the beeps/LED codes right away, correct?

Either way, it might be worthwhile to temporarily unscrew the fan from the rear panel and set it inside so it's blowing directly on the memory when you're set up this way - fully buffered DIMM memory (as I'm sure you know) runs -really- hot.

When you tried the memory in the risers again, did you install each DIMM in slot 1 of the "riser 1" and "riser 2" cards? And you got the same results? Even though the manual says this isn't recommended, it ought to work. And did you remember to reconnect all the power leads to the risers?

What's the exact part number from those Kingston DIMMs?

That's about all you can think of.

On the smoke-damaged memory; how "damaged" is it? If there is no obvious visible damage to the DIMMS - especially in the area of the gold plated fingers, I'd be tempted to try it.

The kingston part # is KVR667D2D8F5K2/2G. As far as the smoke damaged ram, it doesn't look too bad, in fact everything in that shroud looked OK, just a little tarnished, these systems stand up to fire and smoke really well. I am trying to use the processors from the same smoke damaged computer, but I figured these to be OK (hopefully) because they look brand new, and the heat sinks didn't get any damage. In fact the only thing that really took a hit was the video card, because it had a little heat damage I can lift the shroud assembly from the old case (750w) to better cool the memory until I buy more to put the risers back in. Also I did install on slot one on risers 1 and 2, and I just tried it with the old ram as well in the other two risers, all slot 1, but still get the beeps. I tried all four without the risers in a few combinations so far, but still nothing. If the system has just one stick of good ram in slot 1 it should boot right? Also I read that the system won't boot if the memory risors are unpluged, thats why I reset the Bios, are the 750w and 1000w boards the same? And could it be something else conflicting, like the gamer card?

Hmm... I'm really stumped now. That memory ought to work.

I agree that your smoke-damaged processors are probably good; the fact that they can execute enough of the BIOS software to operate the beep codes and front-panel diagnostic lights is a good sign.

Using the memory fan from the damaged system is a good idea; it should work fine with the risers removed. The motherboard itself won't know the difference if the risers are there or not (nor does it know the difference between the 750W power supply and the 1000W power supply). The only things you can't do are (1) run with one pair of risers installed without the other pair installed too, or (2) run with the risers installed but with memory also in any of the slots 5 - 8. You're not doing either of those.

If you remove the risers, then try to install only one memory DIMM in motherboard slot 1, the system should boot although it will warn you that you really should install more than one DIMM.

I don't see your problem being connected to any other peripheral devices such as graphics cards, etc, since the system never seems to be getting that far in its boot sequence.

Are you certain that both power connectors are attached to the motherboard? Hate to ask, grasping at straws... Or maybe it's upset because you're running 667 memory instead of 533? That should be fine, but I really can't think of anything else. Sorry I'm not more help.

You have been a great help, I still can't get the thing too work. Both power connectors are indeed plugged in and I get an orange light on the motherboard. Looking at things there is a connector (p4) that looks lonely, Its probably for the video card riser (don't have this) but I just want to check. Also prior to turning the system on I get a 4 on the front panel. I tried removing everything but ram and processor, and now I am going to switch the processor for another one, that is the last thing I can think of. I wonder if it could potentially hurt the power supply if I plug it into the old motherboard that is potentially damaged, I had cleaned it up ever so carefully over the summer with swabs and rubbing alcohol when I was trying to make that computer work. It looks fine and there are no bulging or leaking capacitors. I reread the return policy on this PC and I think I am not in a good place, so I have to make this computer work.

Doing some research, it seems that the memory controller on my P690 is probably bad, anything I can do to check it? anybody?

The new Ram is 667, and the old I am not sure, but it reads on the stick:1GB 2Rx8 PC2-5300F-555-11-BO. The system tries to boot the same with the vid card as without, and I should note that it sounds the POST errors rather quickly, within a manner of seconds either way. Also the fans all seem to work fine, and try as I might I have not been able to produce a different POST code (unplug fans, remove video card, unplug optical and disk drives, try to start with intrusion monitor not depressed, having CPU in slot 1 instead of 0) The only way I could get a different boot was to remove the CPU altogether, and no Bios code was run, and the system simply shut itself off (although with nothing but a stick of ram installed on the system, the orange light remained solid and I got the 4 light on the front panel. I am going to try one last thing that I just thought of, to try and boot with no ram installed to see if I can get a different code, I will post back my results shortly. Oh yea, thanks again for helping out, I wish there was a way to give rep in these forums, cause I would be giving you a ton for stopping to help me

Lessee; jumping back to your earlier message first... The orange/yellow light on the motherboard and the "4" on the front panel are both saying the same thing; that your power supply is producing the auxiliary, or standby, power that the motherboard needs when it's supposedly off. So that's a good thing.

It's OK that the P4 connector is open; it's sort of an accessory connector that isn't always used. I think that you may be right that it powers the dual graphics card riser.

I'm not so sure I'd plug in the old motherboard just yet (unless you already have, heh). Although it's not likely to hurt your power supply I'm not sure you're at the stage yet where I'd suggest that.

OK, catching up to your most recent message... your memory controller is embedded in your Intel "north bridge" chip. That's the chip that's buried underneath that metal heat sink near the center of your motherboard (near the memory). This chip controls a bunch of other stuff, so while just the memory portion could be bad, I'd be surprised if that was all that was bad in the chip (and all your other stuff seems to be working).

Remember when I said that this problem couldn't be connected to your graphics card? Forget that (heh); I did think of one possibility. It's kind of an 'out there' possibility, but there is one way the card could interfere with the memory. Have you tried to boot without that card in place? Did the system behave any differently? I know it won't boot, but does it fail to boot in a different manner without the card than it did with it?

Lastly - is all of your memory 667 MHz (PC 5300)? Or do you have any 533 MHz memory?

OK I tried with no ram installed and got the same audio POST but a 1 on the front panel, so it can at least see that no memory module is connected, I am going to go out and get a card cleaner and clean inside the slot of Dim1 just in case there is a problem with debris blocking the pins from contacting. Also the north bridge looks fine, is there a way however to remove the heatsink, I want to have a look, see if it was cracked, maybe a heavy hand installed the heatsink. Is it a matter of pushing the clasps down and rotating? I was terrible with these types on my old 486 machines, I could get the heaatsink on, but those suckers never came off without grief and a pinched finger.

I know I am getting a bit longwinded here but I wanted to add that I cleaned the channel for Dim1 with no change in anything. reset the bios both by jumper and a battery pull for 10 mins (I have read that some users had a problem with new processors until the battery was pulled to reset the Bios). Could power be an issue? I really know nothing on the subject, but I do live in an older house, that had poor wiring originally, could my P690 not be pulling adequate power? Also the cord I am using is an IEC c19 to 5-15p 15a common pin. I am pretty sure this is the correct cord, but I really just want to cover my bases here. Also is it common for Dell support to put one on hold and hang up on them? I had this happen 3 times when calling, it really is disconcerting.

Boy, I sure wish you'd throw me a bone here. :) Everything you describe is normal, except that the system won't boot!

Your old smoke-damaged RAM is 667 - the 5300 gives it away. Each DIMM is 8 bytes wide so it tranfers 8 bytes on each clock tick... 8 X 667 = 5300. I think this may be significant but I need to think about it a little more. 667 RAM ought to work with your Dempsey processor but, what if...

Regarding post codes - your inability to influence the post codes makes sense. Memory is one of the first things to be checked; if the system can't set up memory, it can't really do much at all so it throws up its hands and says "I give up". That's the beep code you're hearing. It never even tries to set up the graphics cards, check for fan operation, or any of those things. That all would come later.

With no processor, you can't even get that far, so the code is different in that case.

The north bridge chip isn't quite like the ones you remember. I'm guessing you're thinking of a chip with a small, say, 3/8 in square piece of silicon on top, right? This chip is kind of a beast - encased in metal much like the processors themselves. So you can remove the heat sink if you wish but unfortunately you won't find much under there but metal. Watch your fingers, heh.

This doesn't act like poor power in my opinion. There are failure codes to indicate that a power supply rail has failed and you seem to be getting past those. And until it boots, the unit probably couldn't pull more than maybe a couple hundred watts from the wall. I believe you're OK there. Of course, I'm sure that statement would carry more weight if I knew where you -weren't- OK...

One more question - your processors from the smoked system - were they also 5060 Dempsey's?

They were, I am trying to use them... I am thinking that maybe they could be the problem, but I would think that they would pull a POST code before the RAM, and I have tried both with the same results. Also cause I can maybe see where you are going the old 750w system mobo is the cn-0d029-13740-68M-0023 Rev A00. and the 1kw is cn-0dt029-13740-67v-00zj rev A00. I know the first is the 5xxx series, as in it doesn't support the newer quad cores, not too sure about the second, the listing where I got it advertised it as quad core, and they assured me that it was despite being listed as 5xxx series it was quad core, not an expert here, but are the new ones 51xx series or something like that?

Yeah; you said that early on about where your processors came from. Somehow I decided you had one or more from each unit. Sorry about that.

I think you'll find that both of your motherboards are "DT029" - and thus the same. They should be fine with the processors you have.

There are three series of Intel processors that can work in the various versions of the 690:

- 50xx "Dempsey" dual core, based on the older P4-style architecture
- 51xx "Woodcrest" dual core, based on newer notebook-style architecture (really!) for power savings
- 53xx "Clovertown" quad-core processors, basically two Woodcrests in one chip.

I think your motherboard will work with the first two but not the third.

Hmm. I'm starting to believe in the bad memory theory, finally. Lastly, for now, please check your private messages (top of the page).