PDA

View Full Version : Start up issues after new PSU.



dami0101
12-14-2011, 03:06 PM
Hello,

I'm hoping I can get a little insight into an issue that I'm having.

I built my computer about 3.5 years ago and it consisted of:
EVGA nForce 780i SLI ATX Intel Motherboard
Intel Core 2 Duo E8400
G.SKILL 4GB DDR2 800 (PC2 6400)
GeForce 8800 GT
Thermaltake W0106RU(Toughpower series) 700W PSU
Seagate 500 GB 7200 RPM HDD
Windows Vista 64 and later upgraded to 7 Ultimate 64

About a month ago I sent my PSU into Thermaltake because it was making a clicking sound, assume bad bearing, it had been doing this for about a year and I had always assumed it was my HDD waiting to die. They sent back that they would replace it with a W0319RU(TR series) but that appeared to have been from a lower model line and got them to send me a TPX-775M (Toughpower XT series) instead.

I put my system back together on Monday evening and it ran fine that night. Yesterday morning before going to work I turned on the machine to let it run the install for TOR while at work and when I got home I restarted the computer and at that point it wouldn't boot. I was getting 1 long beep followed by 2 short beeps and my computer wouldn't even pull up to the bios and my mobo gave me several different debug codes across different resets. I made sure all of the power cables were properly connected to the devices on the PSU since it's a modular one, and I even tried booting with just the HDD being powered, single cable powering HDD and 2 optical drives.

Finally after about 20 resets it booted to my bios and I got an error of Floppy disk(s) fail (40) CMOS checksum error - Defaults loaded. I went WTF since I don't have a floppy disk and when I got into windows I got an error that I needed to format my A:drive. I closed that out, procceeded to play League of Legends for about 3 hours and then went to bed.

This morning I got up and it booted the first time but once again gave me the floppy errors. I played around on it a bit updating my iphone and checking e-mails before playing a game of LoL. Things were going fine until the game was about to be over and then my monitor went into power save mode because it wasn't getting a signal from the computer, but I could still hear some passive game sounds, though not any active battle sounds. I reset my computer and once again got the 3 beeps and no boot. Reset again, got into my bios and looked at my temps which were CPU 47, board 20, MCP 45 and as I was about to write these down my computer did the same thing it did when I was in LoL. Took two more resets before I could get back to a BIOS screen.

So the question of the day, does anyone have any ideas?

zburns
12-15-2011, 09:04 AM
This is a late reply to your post. The primary reason being, that you are describing a multiple number of problems happening sort of 'back to back' with the only 'common thread' being the replacement of the power supply. In troubleshooting any mechanical or electrical problem, there is almost always ONLY one primary cause; there can be secondary issues that 'erupt' due to the main cause of the problem. There is no way for all to happen as you describe it, except for a psu or cpu 'gone mad'. A failing or intermittent performing psu could cause 'weird' performance of the cpu where the +12 volt rail is erratic. ( I had to put together in my mind a senario that fits your description. )

So you have to look at your problem starting with the TPX-775M replacement psu and say 'how can this replacement psu' cause all this 'stuff' to happen.

Evidently, you have some sort of long lasting warranty for the 'original' power supply. You sent it back; they tried to send you a 'cheaper or lower performance level psu' which you would not accept. So they 'upped' the model to the TPX-775M. Question is did they send you a 'absolutely fresh, newly manufactured model that had gone thru all normal mfg processes, quality control tests, etc. OR did they send you a 'refurbished' model that had been returned to you that according to the mfg should be 'a good as new model' -- one that someone else had a problem with, returned it to the company, it was checked out, repaired etc., tested, etc. and then sent out to you. If this is the case, how do you 'know this process is without compromise'.

For example, you do not know if there was some intermittent electrical problem that showed itself only 'some of the time', etc., maybe such a problem was just missed?

Now to the, modular feature of the psu. If they sent you a modular psu, then your original psu had to be modular because the cable arrangement coming out from the psu had to be modular. This had to be the case with your original psu, assuming you were not sent a modular cable set with the TPX-775M which you did not mention (no comment on the modular cables). I also have to assume you reused the original and older modular cables.

Modular cables. (Antec's newest power supplies have some modular features and modular cables; however, their modular cables have 'gold plated' pins -- which should mean that the modular sockets they go into have 'gold plated' sockets). Your modular cables if they date back to the date of your first build are likely 'tin plated' (silver color). My impression is that where 'higher currents' exist, as in the +12 volt rails for the mobo and video cards, a 'fair amount of amps' is drawn thru these modular pins/sockets. Tin plating as opposed to gold plating is prone to 'arcing' because the plated pins/sockets can 'corrode due to the combination of moisture and heat (from the presence of high amps thru them, such as 20 or 30 amps thru a + 12 volt rail output. Once corosion and arcing starts, it becomes a 'runaway electrical problem' -- it feeds on itself and only gets worse).

Separate possible problems. Your cpu temps are too high. In a rest position, computer just turned on, cpu temps 32 to 37 degrees C is what you should expect, not 47 -- however, did you read this direct from your BIOS or did a separate 'app' display it for you. Assuming the cpu high temps are correct, this could be caused if the +12 volt rail feeding the cpu is putting out 'in an erratic fashion' too much voltage some of (or all) the time which would cause the cpu to overheat. Arcing within the modular plug that feeds the cpu power could be causing the higher temps (assuming the temps are real).

I reccomend you do something like this. Get a new psu that is not 100% modular. Antec's latest models are 'partially' modular, cables wired directly into the housing, with few of the 'outputs' modular but gold plated. Try a new psu and see if things settle down.

Assuming you have been careful in your 'repair' effort, including static protection wrist strap, etc., you have to start with a new psu -- because you 'have no history on the replacement psu sent to you' -- I assume it is not a freshly manufactured psu -- but a so called refurbished model. Get a new psu with new cables installed (soldered) inside the psu housing, hook up only the mobo with HD, ram and cpu first, look at temps for the cpu in the BIOS, then add video card if all else looks ok!

Question; after upgrading to Win7, -- DID any of the 'weird stuff you describe' occur after the upgrade but before the psu problem surfaced? Final comment. You cannot have multiple problems all occur at the same time. If at one point everything was ok, then everything goes haywire, only one thing 'started it' (for example the psu); other failures may be the result of the first failure, etc. All you can do (my recc) is to put in a psu with new cables as the first step, then with a minimum of components, turn the computer on and go from there!