• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

First PSU; Then memory ; now -what the hell??Help please

D

Davbaron

Member
Sorry this may be a bit long , but I have described a sequence of events which is
probably essential in the diagnosis.
My system is Athlon 1000 T?Bird; 512 SDRAM 133; MSI MS6340 Ver 1 mobo;
GTS2 Vid card; Santa Cruz sound card; ATX case with Enermax 350 w PSU

I have changed my PSU from a 235 w to this new one. The reason Ii did it was that ,
after a sudden close down when playing a flight sim, I switched on again and nothing
happened POST wise, all I got was nothing on the monitor. But the CPU fan worked
and so did the switch lights, but that was all. No Bleeps. I suspected the Power_good
signal wasn?t getting thru( having read Scott mueller?s book on PC?s and so ordered a
replacement PSU, as I thought mine was going on the blink.
On one occasion, the PC did boot up and I thought all was Ok , but some two days
later it suddenly went off, again, in the middle of a flight sim.
The new PSU I got was the Enermax 365P - VD. I wernt to fit it today and still a
problem as below.
I have a MS 6340 Ver 1 mobo. It is fitted with a AMD Thunderbird 1000 processor.
I have fitted the plugs of the PSU the same as when they came out of the system and
they are;
1. The twenty pin (Pl) plug which goes into the motherboard.
No other plug was taken out of the motherboard when i disconnected the PSU.
2. The peripheral connectors to Hard drive; CDRom; CD Writer; (4 pin )
3. The floppy connector.(4 pin )
There are two other sorts of plug which I haven?t connected to anything. In the PSU
manual they are described as
a) Aux power connector (Pentium4 )
b) +12v Power connector (Pentium 4 )
In view of the (Pentium 4 ) after their description , I presume they are only applicable
if i had a pentium 4 processor and therefore a pentium 4 motherboard.
I cannot see what else I need to plug in. So, I switched on .
The switch light comes on.
The CPU fan goes round ,
There are three bleeps; then three more bleeps; then three more bleeps and so it
goes on. Can someone confirm that this is a Post Audio Error code. If so, what is
indicated?
My BIOS , according to the round disc on the motherboard says 686
AMIBIOS
c1999
AK87
2796

I hit the switch to turn off , but nothing happens, The switch stays on, the bleeps keep
going.
Now, I remembered that Scott mueller?s book said that there were audio error bleep
codes on a CD with his book, and I found the CD. I looked and behold, three bleeps
for Amibios is a memory problem.So I took out the memory dimms ( two ) and re
located them and switched on and Hurrah - It all WORKED!! Brilliant- Problem
solved and all right, maybe I had changed the PSU unnecessarily , but it now ran very
quiet and I knew I had a quality powerful PSU.
I had several days of good working order from the ?puter, including some intentional
shutdowns at night and restarts OK.
Then yesterday, 13 (sic!) January it stopped again in the middle of a flight sim.
Earlier that day, I had run Norton Utilities and decided to have a hardware diagnostics
check.
It told me there was a fault somewhere on the memory!
Then later, the system stopped as described.

So, I figured that the memory fault had developed and caused the crash ( I think now
this may not be right but see later ) and as my 512 SDRAM memory was in two 256
sticks, I decided to put in one stick at a time and when it ran then I was using the good
stick, Right?----------Wrong!
It wouldn?t boot and although the switch comes on , and the fan on CPU and Video
card goes round, there is no sign of a boot up.
Ominously, there is no bleep as there was last time which worried me as to whether
this was a memory problem like last time. After many attempts there is no luck and
there has never been any bleep-
It?s as if there is a missing connection.
I have done the following;
I have?rung the changes? on the memory- both sticks one at a time in both DIMM
slots; and together in both DIMM slots. No change No bleeps
I have changed the monitor cable . The new one was working on my second
computer OK and the original now works OK on the second computer. No
change- No bleeps
I have swapped video cards ( both AGP ) No change - No bleeps.
I have unplugged and re-located the Power supply 20 pin plug to the mobo.
I have unplugged and relocated the JFP1 plug on the mobo
I have done the same with the only other plug to mobo, which I think is called a
J8 plug.
I have unplugged and re-located the hard drive ribbon plug to the IDE1 socket on
the mobo.
I have unplugged and re-located the power plug to the hard drive..

All resulted in no change and no bleeps.

As I say, the fans on the CPU HS and the video card are working.
There is no smell of burning from the CPU area.

Conclusions?? - In view of the lack of bleeps I think the memory is not the problem?
But could both sticks have ?blown? - I would have thought that at least one of the
sticks should be Ok and there would be a bleep if memory problem?

It is as if there is a dry joint on the mobo or an intermittent connection somewhere has
maybe become permanent.
Or, for some reason, there is no Power_Good Signal getting to the mobo. Unlikely
though in view of the one week old Enermax 350 w PSU?

So, I?m going to attempt to fully strip the ?puter in a day or so- ( I?ve never done that
before ! Oh My Gawd! ) and then subject to advice from the good old internet, I
might have to put in a new mobo and CPU!

What would you check next to cure this, and how??
What really is niggling is that it may be hardly anything wrong, just a connection!!!

Advice would be greatly appreciated - but be gentle on this near 60 yr old.
Thanks in advance.

 
Well, if both sticks were bad, you would get the series of beeps indicating that there is no memory detected.. It sounds like your Mobo is going bad
 
The best bet would be to try your CPU in another motherbaord... if you dont have access to one, go to a place like Best Buy maybe? that has a GREAT return policy, buy a board, try out your CPU, and then if you find out its a bad board, then well.... problem solved, if you find out its a bad CPU, take the board back for refund, and then order a new CPU
 
Also, pull your CPU off of the motherboard, and see if thier is any discoloration to the thermal paste.. if there is, your CPU may be damaged
 
Wageslave - Thanks a lot for your replies. I will add here what has happened since your replies. In particular, can you describe what I SHOULD find when i remove the processor if all is fine. I have only done it once ( this morning ) and I have found some little discolouration on the white heatsink. On the bottom of the heatsink ( which is about 2" square) , is a half inch square black mark!! Is this meant to be paste ? and what colour should it be ?
"Thanks I appreciate your replies. The tale gets curioser- You may be interested. !
This immediate problem under discussion was solved, thanks to your and other suggestions, particularly Scott mueller's book which told me what the three bleeps in my Ami bios meant. It was memory, so when I re seated the memory everything then worked. So that meant that the Original problem was the PSU and then the memory had become loose when fitting the new PSU , Yes ???- DONT KNOW, READ ON!!!
Everything was fine for four days, then the 'puter stopped again in a flight sim.
The important thing now with this not booting is the same as before i bought the new PSU in that THERE ARE NOW NO BEEPS !
I have since several times re seated the memory, the vid card the PSU plug.
I have swapped vid card, monitor cables - all which work with my other computer. Still no go.
I have today re located the processor and heatsink assembly. No go.
So I then ( a first for me ) completely stripped the 'puter. I took out the mobo ( still with memory and processor on) and relocated it.
No Change.
I think there are still three possibilities, about which I would like opinions please.
1. The mobo has packed up . However, I am surprised that it could pack up Twice!! But I noticed a strange thing when taking the mobo off the case. I expected some plastic washers to insulate the mobo from the case?? There were none . Only "raised bumps" in the case into which to screw the mobo! Have I got this all wrong (that the mobo should be insulated) ? Or should the washers be there and there has been a mistake or reliance on the mobos paint (sic ) to insulate?? I might have this all to cock, but if this is the case it could explain the intermittent, and now apparently permanent, shorting ?? No doubt i will be enlightened🙂
2. The second PSU has packed up. -Apparently - it isn't necessarily true that because there is power to the fans etc on mobo, that the PSU is OK. It seems, according to Scott mueller's book, that the Power_Good signal could still be absent in those circumstances, and it is that signal that boots you up!!
3. The processor has packed up- It seemed pristine (the pins etc ) when i took it off, the white heatsink had traces of browning but no signs of anything burnt out?
What about the switch? On the case ? If that comes on, then it is OK is it? Or are there "stages" it uses and could one of those "stages " be not happening? - This is pure speculation by a desperate man🙂

I think next step before ordering a new mobo is to check PSU ( What's the odds on a duff new PSU ) in another system and if possible the processor( easier said than done)
Suggestions please??
The only good thing about this is I'm learning , I'm learning - ( Education is a ah heck sometimes !🙂) "

Comments from all welcome - Thanks
 
Hi!

I've read your posts carefully multiple times.

I think everybody agrees that your first PSU could not handle the load and reacted as expected. That's ok. You got a bigger one. That's
ok also. I think the following scenario happend: While you were changing the PSU (or moving the memory) the HSF got touched somehow by your hand or another object and it moved a little on the CPU. That can make some gaps between the core and the HSF and can lead to overheating later....

I assume that the HSF was installed by the vendor you bought the PC. In this case it has a thermal pad 1 inch square of white or pink color and it feels like a thin rubber if you touch it with your nail. In the center of this thermal pad must be a mark that the cpu core was left by touching the HSF (aprox. half inch square). The mark must be of the same color as the PAD but a little thiner (the pressure has squeezed the most of the pad out of the contat area ...). I had never seen an black thermal pad thought but it doesn't mean that it do not exists.
You can be in problem if you removed the HSF and installed it back latter because the thermal pad can be used only once !!! The second time the same pad makes a bad contact and it lets the CPU heat up to dangerous levels. When you install the HSF the second time you must remove any trace of the thermal pad (from both CPU and HSF bottom) and use a thin layer of ArcticSilver of regular silicon paste (white color).

So if you moved the HSF or reseated with the same PAD it is possible that the CPU went bad by overheating 🙁 Decolorations or black collor stuff do not mean a good thing either...
Please inspect the bottom of the HSF and the cpu core more and post your remarks again.

If you suspect that the mobo had made a short-circuity with the case then inspect those places better for signs of scraches or burnt. Every mobo I've seem have some plastic tabs that keep the mobo at a specified distance (0.7 cm) from the case. The only places that the mobo cames in direct contact with the case is trough some screws near the PCI slots or near the keyboard plug. And even in thse cases the screw touches only places on the mobo that are grounded or are at 0V all the time. Put the mobo outside the case to remove the posibility of touching it.

For testing use only Mobo/CPU/psu/videocard/keyboard. Anything else can be removed to narow the problem better.
If you install the CPU use silicon paste or arctic silver.

I think the PSU is OK, but the CPU is bad.

Keep us posted with the findings!
Cheers,
 
You don't need to isolate the motherboard from the case when mounting it. The area around the holes in the motherboard for the screws is already isolated from any trace paths with current flowing through them.

When diagnosing your system you seem to have failed to replace the one item you were originally having problems with... the RAM. I would suggest you try to find some different sticks of RAM to test with.

I fried a stick of RAM once while upgrading my computer. (I dropped it on the floor and that was all it took.) After reinstalling it into the new computer I didn't get any error beeps while booting, but once inside Windows I would get random lockups. I then noticed that during bootup my system was reporting an odd number for the RAM size, like 212MB instead of 256MB. Now, if I took out the "good" stick of RAM and left the "bad" one in there my system wouldn't boot at all, but during all this I never got any beeps from the BIOS. So, just because you're not hearing any beeps doesn't mean your RAM isn't fried. And yes, it's possible you fried both sticks. Whatever caused the one to fail could just as easily have damaged the other (defective part, power surge, static electricity, etc).

I also agree with cookieman that the black area you saw on your CPU is a bad thing. This normally is representative of excessive heat buildup, possibly indicating a dead CPU or at the very least a high potential that your heatsink is not removing heat properly or the inside of your computer is getting excessively hot. He is also correct in that those thermal pads on the CPU are only good for one application. If you remove the CPU for any reason you must replace the thermal pad with a new one or else use a thermal paste in place of the used pad.

Another thing to try is taking your suspect RAM and trying it out in another computer.

Yet another thing to consider is that playing a flight sim requires a lot more out of your processor, RAM, video card, and power supply. This means they are going to generate more heat when you're playing a flight sim then when you're surfing the web. Any weak link in a computer is going to be stressed a lot more during heavy usage, such as your flight sim gaming and increase the potential for failure.

Places to look for problems in my opinion are first the RAM, second the CPU, and last be sure and check for good airflow inside the computer case.
 
Hi!

FluxCapacitor is right, thanks for correction! Somehow I misunderstood and thought that the memory problem was resolved....🙁
Check the memory in another system. It's the easiest thing to check upon. Run test that search for potential hardware problems in memory. Also you could take a look at the socket in which the memory resides (with a strong light). Maybe there is some bent contact. I saw such problems in boards that accepted EDO memory.

Cheers,
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top