Jump to content
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble

MSFN is made available via donations, subscriptions and advertising revenue. The use of ad-blocking software hurts the site. Please disable ad-blocking software or set an exception for MSFN. Alternatively, register and become a site sponsor/subscriber and ads will be disabled automatically. 


Sign in to follow this  
TrevMUN

RAM Errors (Update: Now It's Much Worse)

Recommended Posts

Posted (edited)

Hey gang,

I've got a rig with an ASUS P5Q PRO TURBO motherboard. I've been using this setup since 2009. Since May last year I've been having a problem with the RAM; far as I can tell through hours of MemTest86+ runs and seating/swapping RAM sticks, it's not the sticks themselves that are causing the problem, but one of the RAM slots.

Specifically, it appears that the slot closest to the CPU is no longer keeping the stick locked in place. When I start experiencing memory errors, reseating the RAM makes the problem go away for a time; before reseating the RAM I'll see errors come up right away on MemTest86+, but after reseating I can run the program for an entire day (six or ten passes or more) without a single error ... and a few weeks (or even days!) later, the errors come back.

I'll have to pay attention the next time this happens but I think the PCI-E side of that particular slot I mentioned has a lever that isn't closing all the way shut, or won't stay closed shut. I noticed this when swapping the sticks out just earlier today when I came home from work to find the errors were happening again. I had to press down on that side of the stick to get that particular lever to snap into place.

Does anyone have any suggestions on how to remedy this problem? Far as I can tell everything else is still working great and I'd like to reuse these parts for a different PC in the near future, once I can finally overhaul this rig.

EDIT: Things have gotten worse and it doesn't seem isolated to that single RAM slot now.

Edited by TrevMUN

Share this post


Link to post
Share on other sites

Posted (edited)

It's queer. :dubbio:

Do check (with a magnifying glass) that the inside of the slot is clean, that pins are not bent and that the lever is not damaged.

It is similar to this kind of lever/slot:

https://www.compuram.de/blog/en/installing-new-memory-modules-a-short-guide/

halterung-dimm.jpg

right? :unsure:

The lever is self-locking, i.e. when you insert the stick, it makes pressure on the lower part of the lever and makes it close, it is possible that something prevents the stick from going "full in" (and thus locking the lever) or the lever (or the hole on which the lever pivots) is damaged.

In theory the lever is more an "extraction" lever, i.e. the RAM stick should stay in place even if the lever is broken, but it is entirely possible that the slot pins are not (anymore) as "springy" as they should be.

You can replace the lever (taking one from a dead motherboard), but of course you cannot easily replace the connector.

The poor man solution is to wrap an elastic band around the slot, keeping the levers closed, but it is only temporary, as the rubber will soon "cook" itself and break.

A more proper solution is (it depends on the specific slot, thickness of the ram module, etc. ) to build an external clamp of sorts.

jaclaz 

 

Edited by jaclaz
  • Like 1

Share this post


Link to post
Share on other sites

Yeah, that's the type of lever on the RAM slot. Rubber wouldn't be a danger to the other parts in the CPU when it breaks, yeah? How long would a rubber band hold out?

What would be my options to build an external clamp? The sticks I've got are Crucial Ballistix Tracer 1024 MB DDR2-800s, the kind with the red and green LEDs that cycle when the RAM's in use.

Share this post


Link to post
Share on other sites

An update on my situation here; what I assumed might be the RAM clips on that slot might not be the case.

I had been testing to see if keeping the slot I described bereft of any RAM sticks would confirm whether or not my issue is that specific slot. Up until now, it had seemed like reseating the stick in that slot would solve problems for a time.

Today, I came back from work to find my computer BSoD'd. I ran MemTest86+ as soon as I could, and in the first 30 seconds, found over three thousand errors.
When I removed a second stick (the one adjacent to the slot closest to the CPU) and tried again, I still had tons of errors, thirty in the first 30 seconds. Far as I could tell, in both cases each slot's levers were still closed. They did seem to be a little wiggleable in their slots, though I don't know if that is normal.

So now I'm down to one stick of RAM in the slot furthest from the CPU. Neither the two middle slots nor the sticks that were in them ever showed a problem before, so it's curious I would suddenly have such an explosively bad episode happen now that the slot I thought was a problem had been left vacant.

I'm at a loss as to what's going on here, and if it's even fixable. A friend had considered the possibility that my motherboard might be missing a standoff and that the board is flexing. We also checked to see if there are any bulging or leaking capacitors, but there aren't any.

Anyone have any ideas? Is this setup salvageable, or is it time to junk these parts?

Share this post


Link to post
Share on other sites
20 minutes ago, TrevMUN said:

Neither the two middle slots nor the sticks that were in them ever showed a problem before, so it's curious I would suddenly have such an explosively bad episode happen now that the slot I thought was a problem had been left vacant.

Electronics rarely suddenly fail. If they do, it is usually due to static discharge or some other foreign event such as lightning or Pepsi. :w00t:

Components work because they operate within an allowable tolerance. Over time, the actual tolerance (or range of possible electronic signal) for a particular component may extend beyond the acceptable range, which then creates undesired effects. Timing being wrong, voltage being wrong, etc. What happens suddenly is that you notice that something is bad and in all likelyhood a part of collection of parts within the circuit have been bad or going bad for years, if not straight from the factory. Most people end up ignoring warning signs, even if they know what they are, so it seems weird that a failure could come to some great surprise. And besides, it wasn't very sudden, by your account you've been having a problem for 8 months now.

I would test the RAM in another known working board. I would also replace the existing board, and then use it for testing purposes only.

Share this post


Link to post
Share on other sites
Quote

it seems weird that a failure could come to some great surprise

It's a great surprise when parts which seem to not be having an issue (that is, the middle RAM sticks/slots) suddenly have it so dramatically. Usually when I catch the RAM in that one particular slot misbehaving, it would at most a dozen errors in a single pass. Even one error is bad, of course. None of the other slots/sticks showed signs of problems until today, and the result was quite severe.

I don't have any other boards which can accept DDR2. I could ask around and see if anyone has a spare motherboard to test the RAM's integrity, but at this point if the motherboard is a write-off then it's not much use keeping the RAM or CPU, either.

Share this post


Link to post
Share on other sites

Update on the situation: The last stick of DDR2 RAM left in the machine after I pulled out the other three finally developed RAM errors 11 days later. Because it's taking time for me to earn the money I need to buy parts for an all-new setup, I am continuing to try and determine what's causing the problem and if it is salvageable. (Condescending ":w00t:" replies aside.)

A friend of mine suggested earlier this week that I try leaving the machine completely off (power supply switched off and unplugged) until the whole thing cools off and then try testing the RAM again. Strangely enough, the errors vanished when I did that, and no errors appeared after leaving MemTest86+ to run overnight. Today, after repeating the cool-off process, I reinserted the three sticks I took out two weeks ago, and ran MemTest86+ again. So far, no errors.

I'm not sure what this means. The issues weren't relegated to one slot, but eventually affect all the sticks no matter how many I pull out. Yet, the RAM errors go away when the sticks are reseated, and also go away when the computer is left off long enough to cool entirely.

Is it possible that the power supply could be causing this? I guess I could test by making a power supply my next priority purchase, testing it on my current setup, and if the problems do not go away that would rule out the older power supply as the problem ...

Share this post


Link to post
Share on other sites

First make sure to clean with a soft eraser the golden pins of all RAM boards and also clean the RAM receptacles on MB with a brush and latter pressurized air.

If this doesn't fix the problem and as you said you have been using that PC since 2009, it is now 10 years old, so my best guess is: there are internal failures on some or all components (MB and RAM), wich are more evident when they are hot.

Anyway if you don't find the source of the failure, only final solution is build a new PC, if you as me prefer to install Win7 on your new rig I recommend to see this thread.

alacran

  • Upvote 1

Share this post


Link to post
Share on other sites
On 2/6/2019 at 11:35 AM, alacran said:

First make sure to clean with a soft eraser the golden pins of all RAM boards and also clean the RAM receptacles on MB with a brush and latter pressurized air.

Any specific eraser types you'd recommend for the job? When I think "soft" I think of those kneadable erasers you can get at arts and crafts stores ...

Also, I was talking with some friends about my plans to use the PSU I buy for the new setup to diagnose whether or not the currentsetup's PSU is the culprit for the hardware failure. One of them warned me that if there is hardware trouble it can affect the PSU. That seemed a bit weird to me. Is it possible for failing hardware to damage a PSU like that? I would have imagined it needed to be a catastrophic failure, not ... something like this.

Share this post


Link to post
Share on other sites
5 hours ago, TrevMUN said:

Any specific eraser types you'd recommend for the job? When I think "soft" I think of those kneadable erasers you can get at arts and crafts stores ...

Also, I was talking with some friends about my plans to use the PSU I buy for the new setup to diagnose whether or not the currentsetup's PSU is the culprit for the hardware failure. One of them warned me that if there is hardware trouble it can affect the PSU. That seemed a bit weird to me. Is it possible for failing hardware to damage a PSU like that? I would have imagined it needed to be a catastrophic failure, not ... something like this.

Actually you don't want a " too soft" eraser.

You want a (very mildly) abrasive one, the best being one of those (that you can find not easily anymore anywhere :w00t: :ph34r:) round "wheel" ones once used for typewriters.

This sort:

https://contrapuntalism.blog/2016/01/21/faber-circular-erasers/

traditionally the white ones are softer/intended for pencils, whilst the red and blue ones are a bit stiffer and more abrasive (intended for ink/typewriter),

Nowadays the more suitable ones are this kind of bicolour ones:

http://www.foxystudio.com/wp-content/uploads/2013/09/187040_1.jpg

and you want to use first the blue side and then the softer pink/red side.

jaclaz

 

 

  • Upvote 1

Share this post


Link to post
Share on other sites
9 hours ago, jaclaz said:

the best being one of those round "wheel" ones once used for typewriters

Now, that's something I don't see in RL for at least 30 years... and if someone still has vintage ones in stock, unless they were stored under nitrogen or in vacuo, they must be too oxidized to be of any use...

Share this post


Link to post
Share on other sites
1 hour ago, dencorso said:

Now, that's something I don't see in RL for at least 30 years... 

Sure, it has become an "exclusive", "design" item.

Look at the price of this (OK, it seemingly includes a sharpener, No, it is just the eraser):

http://www.fabercastell.com/products/more-products/eraser/EraserUFOingiftbox/188306

The sharpener is another item:

http://www.fabercastell.com/design/products/ufo-sharpener-eraser/SharpenerUFOingiftbox/188305

They should have IMHO have called it iEraser instead of UFO, but probably Apple wouldn't have been happy about that, and this is the other one, platinum plated AND fulfilling its task perfectly:

Quote

Each of the accessories in the Graf von Faber-Castell Collection fulfills its task perfectly. The round eraser with platinum-plated grip pieces allows smudge-free rubbing out.

 https://www.graf-von-faber-castell.com/products/Eraserroundplatinumplated/188502

and now look at this latter's price:

https://www.thepencompany.com/product/graf-von-faber-castell-round-eraser/

jaclaz

 

Edited by jaclaz

Share this post


Link to post
Share on other sites
On 2/7/2019 at 10:02 PM, TrevMUN said:

Any specific eraser types you'd recommend for the job? When I think "soft" I think of those kneadable erasers you can get at arts and crafts stores ...

I always have used for more than 20 years the white "Pelikan" erasers or similars, they do not make excessive wear on golden pins, the idea is to remove the patina and let them almost shiny again.  This eraser do not last very long (as it is soft) but I prefer to spend the eraser and not the golden coating on pins.

About changing the PSU, it is not a bad idea, and not an expensive one.

alacran

Edited by alacran

Share this post


Link to post
Share on other sites

I'll see if I can scrounge up one of those erasers you guys recommended.

5 hours ago, alacran said:

About changing the PSU, it is not a bad idea, and not an expensive one.

Speaking of which: from the power supply calculators I've been looking at for my new setup, I can get away with a 750W supply, but if I plan to do any overclocking I'll need an 850W supply to handle the added power load and any additional cooling. (The calculator estimated a peak load of 802W if the CPU I'm considering were overclocked to its most stable setting.)

The only problem is that my uninterruptible power supply has a capacity of 1350VA/810W. Even if I never actually use anywhere near 810W, would having a 850W PSU installed still cause problems for the UPS?

Edited by TrevMUN

Share this post


Link to post
Share on other sites

Hmmm. :dubbio:

Actually *needing* a 850 W PSU sound like a high-high end gaming machine, and overclocking is so '90's :w00t: :ph34r:.

Are you really sure-sure you need such a powerful PSU?

About the eraser, *any* will do nicely, we were just being a bit nostalgic of the good ol'times ...

jaclaz 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...