Please help a newbie (me)!

[scottkw]

I just upgraded my 4200 to a 9800 Pro and now I'm having CTD issues in every game I play except World of Warcraft (I wish someone could tell me why that game doesnt have any problem).

Naturally I suspected my PSU might be the problem, so I started running ASUS Probe and MBM 5 to monitor Voltages at 5% changes. After running all day I had 3 alarms, 2 from MBM which said my +12v dropped to 11.37v briefly and one from ASUS which said my 12v hit 12.50 briefly, but it only happened once, and while I was just reading the forums (go figure). Typically while I'm playing something and it crashes there is no voltage alarm.

I don't know if my PSU is junk or not, the sticker says its made by UNEEC, 400W, it lists:

+3.3V/28A +5V/30A +12V/15A and -5V/0.5A -12V/1A +5Vsb/2A

Unfortunately those numbers don't mean anything to me, so I'm pleading for help. As I understand the voltage shouldnt exceed a 5% variance, but if it's a relatively rare event is it still a big deal?

I tried unplugging my soundcard and CDRW, so I only have my 2 HD's, kb/mouse and the 9800 Pro connected but the problem persists.

I'm 99% sure it's not a cooling issue, especially since the CTD's can occur long before anything has a chance to heat up.

I would appreciate any help determining if my PSU is just junk or not, or if its possibly a bad 9800 thats causing all the trouble.

Thanks in advance.

 

[Kain]

What are you powering?

run the numbers

 

[Ice Czar]

^ couldnt have said it better myself
(thanks kain)

15A on the +12V is very low for most modern configs and its very likely our supply simply isnt up to your vidcard upgrade

your assorted componnts will have a maximum draw they are capable of
this is expressed as amps (A) so you add all those up
the 5% variance applies to the voltage on each of the main rails +3.3V, +5V and +12V
it needs to stay within a 5% varience because the circuits on the boards\cards ect expect that voltage range (though some are tested outside of it)
Basic 5% range is
+12V....11.4V to 12.6V
+5V......4.75V to 5.25V
+3.3V...3.135V to 3.465V

the reason a supply might not be able to maintain those voltages is because its overloaded with not enough amps on a given rail, its a cheap supply that is deceptively misrated, or its unable to deal with a serious sag from the AC source power (outlet)
the way that is measured is call the Transient Response
since all your components arent drawing maximum power at the same time, the load is dynamic, the transient response measures how good a job the supply does to adapting to that
and is generally expressed like
+5V,+12V and 3.3V outputs return to within 5% in less than 1ms for 20% load change.

better supplies exceed spec and have load regulation (just another way to say transient response) of 3% or 1%
meaning there is very little to no fluctuation of the reference voltage (3.3, 5, 12) under normal circumstances
but even the best supplies can be overloaded with too many components
(generally that would appear as a "low" voltage for a given rail, say 11V on the +12V rail)
or have issues if there is a great enough brownout, at which point they should safely shutdown without allowing a overshoot or undershoot through

an overshoot or undershoot, is them passing on power outside of that 5% range above
and overshoots especially are what kill mobos, RAM, or any IC (Intergrated Circuit) chip (CPU, GPU, RAM Chip, ect)

so run the calculator for what your planning on powering now and in the near future
and post back with the amps per rail (3.3, 5, 12) and total watts
and we will help you factor in some additional considerations and review possible supplies

note the total watts it will spit out are the theoretical maximum draw of all the components
something that never really happens but that is offset by the likely situation that the rating figures for a given supply are greatly exaggerated, largely by the difference between the temperature the supply was rated at vs the real world operating temperature, but also how the supply was tested by a less than honest PSU manufacturer

 

[scottkw]

I've been doing some testing with my ASUS monitor instead of the MBM since I've read it's not very reliable. When I experience CTD my 12v is reading 11.904, which seems alright to me, am I wrong? That's pretty much the average reading I get under load, it never goes below that according to ASUS.

It also still makes absolutely no sense to me why some games can run fine all day, but then others CTD.

I used the spec creator that was linked, and it came up with the following:

3.3V Max 2.4A
5V Max 18.7A
12V Max 15.6A
3.3+5V Max 101W
Total output 300W or above

Comparing that to what's listed for my current PSU it seems that the 12V is lacking, however the monitoring doesnt seem to reflect that, so I really don't know... very confused by all this. Of course all this is happening the week HL2 comes out! Why me!

On a side note, I noticed with some games disabling AF seemed to fix the problem or at the least reduce it greatly, does that mean anything? Or was it probably just a coincidence?

 

[Kain]

You used ASUS PC Probe? Try to get ahold of a Digital Multimeter. Software monitors are good for general values only.

 

[scottkw]

Well, my dad is an electrician and my brother is an electrical engineering student, but neither of them knows anything about computers... and I know nothing of testing power.

Can you tell me what they should be testing, and how? Or do you know a link that might help? I'd really appreciate it, I'm near the point of just RMAing the card and also buy a new PSU, but it took me 9 months to get the money for a 9800 Pro, I'm not really looking to spend more.

 

[Kain]

There was a link on how to do that here but I forgot which thread. Ice'll be able to help you more than I can.

 

[Ice Czar]

http://www.hardwaresecrets.com/faqpowersupply.html

 

[scottkw]

Ok, I tried running some games with the multimeter connected, idle voltage is 12.22, during game its 12.10/12.11. Of course I get this thing hooked up and now nothing will crash...

We also checked the 5 volt which was about 5.10/5.11 during gaming. There really was not much variation at all, seemed very stable. Also tried running Prime95 in max power test which it stays at about 12.08 while the test is running.

I think it's just trying to drive me insane, and it's working! But what do you guys think, am I any closer to ruling out the PSU as the problem now? Or is there something I'm overlooking?

 

[Ice Czar]

well then next biggest variable is heat
effecting not only the CPU\GPU but the supply as well (3 below)
followed by the AC Source range (2 below)
so ideally you employ a UPS

POWER SHMOWER by Mike Chin

or How PSU Power Ratings Mean Almost Nothing

A frustrating fact about PSUs is that there does not appear to be a stringent or regulated standard for reporting, advertising and labeling rated power. This is despite the existence of standards like ATX2.03 or Intel ATX12V.

There are well-established standards for measuring and rating HDD capacity, an engine's horsepower, or the heat generated by a furnace... but not one for how much power a PSU can deliver. There are so many cases of people with "450W" PSUs having power stability issues running a system that can't possoibly draw more than 150W. And "300W" units that keep running where the "450W" units are faltering.

It's not just about bad PSUs vs better ones. It's a dumb situation caused by uncontrolled marketing competition. Real regulation would bring PSUs out of snake oil territory and into a more sensible consumer-friendly terrain.

There are many ways PSU makers fudge to make their units seem more powerful.

1) Out and out lying. You add up the power on all the lines in many PSUs and they fall short of the rated power by 10, 20 30W or even more.

There are more sophisticated ways:

2) Limit the AC input voltage to a very narrow tolerance. The best PSUs are able to deliver their rated power given a decent range of AC input power, say 90~130V for a 120V unit. It's much more demanding to produce 300W w/90VAC input than with 120VAC, so what some PSU makers will detail in their tech specs (usually not in their consumer brochures) is to specify 115-120VAC for input power. A PSU specified this way will not deliver full power if the AC voltage sags, if there is a brown-out. Surely it causes instability more often than a PSU rated to deliver full power with 90-130VAC.

3) Specify a low operating temperature for rated output. This is quite common, but again not often seen in consumer brochures, but rather tech spec sheets provided usually only on demand by engineers or corp buyers. A typical PSU operating temp statement is somthing like this:

0ºC ~25ºC for full rating of load, decrease to zero Watts O/P at 70ºC

Examine what that says. Full power (let's say 400W) is available when the unit is at 0ºC ~25ºC. Hmmm. Think about this.

Have you ever felt air blown out of a PSU in a PC running absolutely full tilt (which it would have to do to get anywhere near 400W output) that felt cool to the fingers? 25ºC airflow would feel exactly that: Cool, given that normal body temperature is 37 °C.

So this PSU cannot deliver full rated power when its temperature goes over 25ºC. OK, what happens to the max power output capacity above that temp? It decreases gradually so that by the time the PSU temp reaches 70ºC, the PSU cannot deliver any power at all. So if you assume that this power drop as temp rises is linear, then max power capacity will drop by ~9W for every degree over 25ºC.

Now having examined as many PSUs as I have over the last 2~3 years, I have to say there's not a single PSU in ANY PC I have ever used or examined that would not measure at least 30~35ºC almost anywhere inside the PSU under almost any kind of load. And if/when it is pushed, 45ºC is nothing at all, especially for or near hot running components like voltage regulators.

So let's say 40ºC is a fairly typical temp inside a PSU. This 400W rated unit would actually be able to deliver a max of just 220W at that temp. Hmmm. Interesting, isn't it? At 50ºC, the available power would drop to just 130W. No wonder some PSUs have 3 fans each capable of 50 cfm!!

Here's a simple fact: Really high quality PSUs are actually rated for full power output at as high as 40ºC. The trick is get a hold of the spec sheets that tell such information so you can compare apples to apples. Or ask.

what supply are we talking about here?
is this it? > http://www.chenmingusa.com/Uneec.html#400
which would make it a rebranded HighPower (HPC) \ SirTech
here is a competent independent review of its little brother the HPC-300
the HPC-400-101 has been replaced in thier line by the HPC-420-102DF (the Sirtech link)
which is a little unusual but I cant definatively say it really means anything, but the are no longer making your PSU

what are your room, case and CPU temps?
(CPU under both idle and load)

best guess as to what is happening
either you have a thermal problem with the GPU or CPU
or the PSU is getting too hot to effectively deal with a sag from the AC source
you are very close to the maximum load on the +12V rail in theory actually over it
since the PSU capacity drops as it gets hotter, and since its harder for the PSU to maintain a stable voltage when the AC source drops too low
I think you might have an issue under those circumstances

PSUs are all about the worse case senerios
which is why they are infamous as a source of intermittent problems

simply having the side of the case off for testing might be enough to not make it crash

 

[scottkw]

It's about 70F in the room, CPU temp is around 34C idle, up to 52C under max load. Case temp is reported at 33C idle, about 36C under load. P4 2.53 Northwood chip, not OC'd.

PSU Model is HPC-400-101, it's pretty old, at least a couple years... I think it came with the case I ordered, I don't remember picking it out.. not surprising to be discontinued.

What you are saying makes sense about the heat, I don't believe the in case temp is too bad, but the PSU might be getting too hot. I don't believe the GPU heat is the issue, wouldn't it normally come with artifacts or other signs? Also, I tried drastically underclocking the card and it still crashed.

I had the window open for a while before I tried to get it to crash again, so the temp was probably considerably lower, I also had the case pulled out further from under the desk so it was getting a lot more ventilation. Maybe I just need to give it more room to breath. Are there any decent cooled PSU's for reasonable prices? I took a look at newegg since they have been pretty good to me in the past but I guess I'm just not entirely sure what I should be looking for.

edit: Haha, I read the review for the 300 model, and I guess now I know what that annoying noise is....

 

[Ice Czar]

well it wasnt discontinued as much as it was replaced with another 400 watt
with better cooling (conforming to the AMD builder spec that called for a 2nd fan)
so to my mind that might suggest heat issues may have been a consideration

as far as the temps go, definately give it some room to exhaust behind the case
and find it as cool a home as possible
(cant tell you how many times Ive seen n00b friends computers next to or near the heating vent)
increasing the airflow in the case might just do the trick, fans are far cheaper than PSUs
however if you are interested in a PSU, Id likely recommend the
Fortron Blue Storm as the lowest priced but reliable entry level ATX12V v2.0 supply
(the spec has changed from yours v1.3 to v2.0) it would easily power your config
and into the next generation as well
has a 24 pin to 20 pin adapter so it will work fine with your current board

there would be other likely cheaper supplies that would do the job,
but then they are sort of dead end supplies,
not making it an infrastructure investment, just a stopgap solution

PS a CPU temp of 52C is a little high to alot of folks in here
but within the high end of spec I think, which CPU is that?

 

[scottkw]

It's a P4 2.53 Northwood, with stock cooling. I think it is a little high, but I haven't had any instability with it (...until now). I have some arctic silver 3 I could apply to replace the thermal pad, but I am always afraid to mess with something I can't afford to replace. If there is a good guide to doing that I might take a look though.

In regards to the PSU, that's a lot more expensive than I was hoping for, I don't think I will be able to build a new PC for a long while so something cheaper just to get the job done might do more for me, if you have any suggestions for that I'd appreciate it.

I have to say your research skills are amazing, thanks for all the great information. I hope I have found the source of these CTDs, the air from the PSU fan is rather warm even while idle... I am almost afraid to check under load.

 

[Ice Czar]

well generally I like to deduct 1\3rd the rated amps per rail of a supply
to reflect the higher operating temperature over the temperature it was rated at
which would mean we are looking for around a 24 amp supply on the +12V rail
however the components dont ever use all their rated maximum draw at the same time
(which is the true safety margin I generally employ by deducting that 1\3rd)
for instance a HDD might use 2 Amps to overcome inertia when it spins up, but only a quarter of that once spinning, same goes for a fan (to a less draw), the opposite is generally true of the CPU GPU, at startup they arent drawing at full load

in addition there are some brands that are known for being conservative in their rating, and Fortron is one of them, so...

the FSP400-60PNU-R at 18A would likely do it $70 but for that extra $15 its not that far from the Blue Storm
there are alot of other possibilities but again we are really getting close on that +12V rail
http://www.newegg.com/app/SearchProductResult.asp?Submit=Go&DEPA=1

the problem is that CPUs switched from being powered by the +5V rail to being powered off the +12V rail, and video cards have to a certain extent done the same and far exceeded power delivery the AGP slot is capable of
Id hate to recommed something to you, only to find you in the same boat your currently in
you can cut it a little closer and save a few nickles but considering we are talking about a couple of cups of fancy coffee or a few trips to the local fast food as a difference...
Sirtech\HighPower isnt a bad brand, maybe try just improving the airflow

most ATX12V v2.0 upplies are in the 30Amp range and most are over $100
any other option I recommend is going to be an ATX12V v1.3 and there is sort of a gulf between the target amps we want and the budget, I dont really feel comfortable recommending less than 18A

 

[scottkw]

Ah well, if this is workable until christmas I should be set, I might look into another fan in the meantime, the one that came with doesnt feel very strong at all. I moved the PC around a bit to get better ventilation and it seemed to help, but I'm still crossing my fingers. Maybe I'll just buy one less book next semester! Thanks again for all the help!

 

[Ice Czar]

Your welcome, its what we are here for

Good Luck

look at it this way, its winter soon