View Full Version : Info about the 5v numbers...
There's a thread going on right now, where someone has what some people call a "low" 5v value.
My question is, What is Bad about having a low 5v value?
yeah....i read it on here too. This guy posted his benchmark in Sandra or Mobo Monitor and someone pointed out this low +5 number. They said that it should always be greater than the number so that it provides sufficient voltage to the chip or wherever the power is directed to.
AAAhhhhh.....found it
Here is the Thread (http://forums.overclockers.ws/vb/showthread.php?threadid=45308)
To quote Alkaline:
"The problem with the 5 volt is that it is BELOW 5 volts..
(4.86) if anything it should be above 5 volts.. (5.04) or something. Having voltages right on spec, or a little above helps improve stability. If your lines are belowe spec like that, u might want to buy a new power supply.
Here are my lines with 320 watt power supply and 6 fans! "
Hope this helps
Hehehe yeah I know :) I'm a part of that Thread....
But my question is... What does a Low number result it? Does it destroy the CPU? Does it just sit there and look sad? ;) hehehe
There has to be more concrete information than just saying something about Stability... What exactly would a low 5v do?
I guess that its safer ( not neccesarily better) to have less power. See, the way I see it is that if you dont have enough, your cpu will just shut down and try to restart your computer if your lucky. But, if you have TOO much power, you might run the possiblity of frying and burning out your cpu. I think its not a matter of stablility but rather a matter of reliablity.
Thelemac
11-12-01, 04:52 AM
That's the line that the CPU runs off of...so if it's low, you're not really getting enough voltage (at least, I'm pretty sure that's why)
If thats the line, then what is the VCORE? VCORE is the voltage that is adjustable through the bios. I thought that would be the one that the Cpu ran off of. But, I dont know too much about this so....
VCORE is the voltage going to your processor. For example, the default vcore for most AMD processors currently is 1.75v . This can usually be raised in the bios up or down 0.1 . 5v is the power going to all your 5v stuff like cdrom, harddrive, and any fans you may have on 5v. 12v is the power going to your 12v fans ect...
Thelemac
11-12-01, 05:21 PM
You'll notice that the vcore is adjustable...which isn't done at the PSU. The vcore is regulated by the motherboard, which I'm pretty sure uses the 5V rail for that. I suppose it could be the 3.3V, but I doubt that for some reason.
I don't know exact reasons but I know this: I had an old psu that only sent out 4.75 to 4.85v and my Athlon 1000 crashed a lot.
I got a new psu, 5v line puts out 4.98 to 5.05 and it didn't crash anymore.
I read a thread that said low 5v lines (under 4.8v) caused instability in Athlon systems in particular.
My guess is that it supplies either the power or control voltage to the cpu.
So does having a low 5V affect your cpu primarily or your system components (cd-rom, hd, cd-burner, etc). My 5V is running at 4.95, is that too low, or is that acceptable?
Originally posted by Leeua
So does having a low 5V affect your cpu primarily or your system components (cd-rom, hd, cd-burner, etc). My 5V is running at 4.95, is that too low, or is that acceptable?
Do you get a lot of crashes?
My symptoms were: oc(1000@1400) that ran for weeks stable, then started random crashes. When not crashing, all components worked fine. Temps were low 40s load (seti). I happened on a post that mentioned low 5v rails and did some more research (most of which I've forgotten) but the upshot was - low 5v can lead to instability of cpu, especially on oc'd systems.
Thats all I really know for sure, but I'd say that if you aren't having unexplained crashes you must be ok! Someone else mentioned that a .25 variance is nominal, so there seems to be no hard and fast rule about what is ok.
I even know of 1 guy with a dual mp rig that has like 3.8v on the 5v rail and runs about a week at a shot!
wow....3.8!! Hehe, that is LOW!! Well, im running pretty stable so Im not too worried. Keep you guys updated though if anything starts going wrong.
alkaline
11-12-01, 11:43 PM
OK, Here is my experience with the 5V issue.
When I first tried overclocking my computer with Windows XP, I couldnt get more than 1500mhz. I had about 6 fans in my case, one being a powerhungry delta, not to mention the 1.33ghz athlong @1.5ghz, and only a small 230watt power supply. my voltages were about 11.68v/12v and about 4.75v/5v.
After I got hooked up with a new 350watt powersupply, My new voltages were about 12.12/12v and 5.03/5v, and my stability jumped a whole bunch, and i was able to overclock at 1.566ghz, so i think that u would be ok with having low voltages, but if u want to maximize your overclocking potential, having enough voltage is a really good plus.. this is just my experience, im no expert.
I thought that the 5v and 12v voltage rails were used in supplying perepheral components EXTERNAL to the mother board. (correct me if i am wrong here) The MoBo only monitors current to these rails and they are used to power Floppy, HDD etc.http://koti.mbnet.fi/antiqser/smilies/teleport.gif
alkaline
11-13-01, 09:40 AM
Yeah, but isnt having enough voltage going to your peripherals, like your hard drive important? especially when you are upping the FSB which ups the speed that your HD is going at?
Just been reading your comments on 5v Rails etc. & thought I`d give some input. Don`t no to much about this my self either.
But I was running a Athlon 1000mHz on a 300 psu & had a reading of 5.08 & 12.463 I was running 4 extra fans at the time.
Have since updated to a XP1600+ & was told to also up date the psu. So I`ve bought an Antec 350 psu & now have readings of 5.053 & 12.646 still running the same 4 fans.
Does`nt mean to much to me but it looks future proof for a few up-grades yet.
Acko
XP1600=
Asus A7V266-E
512 Crucial DDR
Alpha 8045 / Y.S. Tech fan
Ge-Force 3
30 Gig Diamond Max +
Sound Blaster Live
Plextor 12/10/32A CD/RW
Iiyama 19" Monitor
Antec 350 psu
Temps 38-39 C idle / 42-43 C load
wizardloader
11-13-01, 02:45 PM
Now I see da light!!!!! My older rig has a 300 watt PSU running a 1.2 Bird on an Asus A7A266 (board sucks) and my 5v is 4.85 and it crashes.......................My new has a 350 Fortron Source 350 PSU and is @4.98 with no problems
I have had a lot of problems with PSU's and did some research on the subject some time back.
As a general rule the closer the +5V rail stays to the 5V the more stable the system will be. I have found that once the +5V drops below 4.9 then problems start occuring in the form of blue screens etc. What I have also found is that certain brands of reputable PSU's seem to "go off". I have had a 300W Sparkle and a Macron and both were fine for a month and then just degraded. Have a 350W Enermax which has been fine for a few months but I notice that the +5V is starting to drop.
The following may help;
http://www.procooling.com/articles/html/picking_psu_s_for_athlon_syste.shtml
But basically the story is that it all relates back to the power that each rail can supply. The more devices that are attached to each the more the power drain P=IV P=Power (Watts) I=amps V=Volts. So P is directly proportional to V and I is a constant. If the power requirement increases then the volts will drop.
kevin_bouchard
11-13-01, 06:08 PM
So that would explain why when i manually bump up my vcore my 5v line drops. Has anyone noticed if there 12v line goes up if there is more current being drawn from the 5v line. i change my vcore to 1.87(1.6 default), my 5v line goes to 4.83 but my 12 line jumps from 12.30 to 12.50. One thing I want to know is if a high 12v line can damage hardware, if it can be manually lowered(moded) and if you can bump up easily the 5v line(by means of modding), I was thinking on the lines of a pot or something like what you would use to make a vcore mod for a mobo.
Thanks :D
bigfoot
11-13-01, 06:53 PM
Well, It's been said that if you crack off the case of your PS, you can sometimes find some pots somewhere in that jungle, I used to have a heck or a time OC my 1Ghz T-bird then I learned about this hole 5V thing an noticed Mine was something like 4.6:eek:, after I twiddled with one of those pots in the PS I had it running at 1.2Ghz with a 5.06/5V:)
kevin_bouchard
11-13-01, 06:58 PM
thanks alot i guess i know what i will be doing now:D
There are only a few manufacturers using pots in their PSU's. Just a warning (without wanting to start a new debate on electric shocks and their potential to kill), there are some serious capacitors in there that are still carrying their load and are capable of doing some potentially lethal damage if touched. I have heard of peps getting some heavy jolts from them even after a year of non-use.
kevin_bouchard
11-14-01, 01:02 AM
thanks for the warning, thats exactly what my dad told me when I opened up my tv, now half the muscles in my left arm are fried:eek: ,j/k
No pots in my psu :(, but does any one know how to manipulate the voltage regulator for a L&C model lc-250atx psu?
One thing that does concern me is that the pcb board is a darkish color (almost looks burned) under a resistor and a diode.
Thanks
!-=sky=-!
11-14-01, 08:58 AM
i m running the 5v@4.7v and my computer is still stable
My old PSU, 300W, & new PSU both had their 5V rails in the up side so I never really had an issue with it. Did you guys notice that the 3.3V, 5V & 12V rails fluctuate when your heating up the toaster(CPU) but not the VCore? So if they dip below their spec then you will get instability. The problem is that the tolerances these manufacturers are given are way too big, 4.75V - 5.10V(Enermax e-mail), for people like us who like to push our hardware to its limits so test before you buy & get the shop to agree that if the product does not meet your expectations you can return them in good order. Sometimes monitoring software can be off so it's always best to check with a tester. My reading are close to the software monitor I use so it's safe to base my reading on it.
C0nDuCt EviL
11-14-01, 03:05 PM
Could that be the thread I started..lol
bigfoot
11-14-01, 06:07 PM
sorry about that, I ment to warn him about that capaciter issue. Even though I was awear of that when I opened up my first PSU,(A-Open even used pot's)I got one heck of a shock:eek:. When i was holding it I acidently hit a capaciter pin with my screwdriver
WarriorII
11-14-01, 11:06 PM
Not sure if this is related in any way. I had posted this in the AMD MB section 1st.
Iwill KK266+ 2.5v prob
Got a problem with my 2.5v on the mobo. In Via Hardware monitor(V1.07) it says the 2.5v line
reads 0.25v to 0.27v.
I am not familiar with what the 2.5v line controls.
ANY info would be appreciated.
Everything "seems" to work fine.
Vcore = 1.81v
2.5v = 0.28 (right now)
3.3v = 3.32
5v = 4.95
12v = 12.01
Thanks,
Ok, all is well with my system as far a crashes etc..are concerned, but My 5V sit's at around 4.87+V.
Now my major concern is if it what's got my VCore Up so High? My Vcore is sitting at 1.79-1.81 (fluctuates constantly) and I have my bios set to 1.75V. I'm personally worried it will fry my CPU due to the VCore Fluctuations..as well as the fact tht the VCore is quite a bit higher then my actual settings.
Could the increased VCore also be what's keeping my system temps up?
Your Vcore should be ok, even 1.85v is normal for some. As for your Temps being affected, yes. The lower the Vcore, the lower your temps will be. But don't go too low or else your system will probably become unstable and start blue-screening all over the place.
Kawalski
11-15-01, 04:27 AM
Look at it this way if it helps. That 5+V line is the lifeline of you CPU whilst its in action. If the 5+V is running weak, your starving the processor. It'll run smoother if you supply what it requires.
Originally posted by ROH
Ok, all is well with my system as far a crashes etc..are concerned, but My 5V sit's at around 4.87+V.
Now my major concern is if it what's got my VCore Up so High? My Vcore is sitting at 1.79-1.81 (fluctuates constantly) and I have my bios set to 1.75V. I'm personally worried it will fry my CPU due to the VCore Fluctuations..as well as the fact tht the VCore is quite a bit higher then my actual settings.
Could the increased VCore also be what's keeping my system temps up?
From my own experience when the vcore starts fluctuating it is on the edge and any slight demand can cause a crash. When I see that occuring its time to move the vcore up a notch. I use it as a tool. I set a speed and use a higher vcore than I need then start reducing the vcore until the vcore readings start fluctuating and then increase on step.
Not too sure about 2.5V, but for my .02:
It is my understanding that the VCore powers the CPU,
3.3V powers the I/O - ie. PCI/ISA (although 5V is additional supplied to cards via PCI),
5V powers items such as chipset (Northbridge/Southbridge),
12V powers fans (CPU/Chipset),
The 3.3/5 and 12 are all supplied by the PS, the VCore is derived from the 5V (I think).
The voltage supplied to the peripherals is 12V (which is why you can put 12V fans on these same connectors).
A low 5V signal will thus affect your chipset and some cards, which can cause stability problems. It shouldn't affect the CPU directly because this is self-regulating - what you get on VCore (in the bios) should be the actual supplied to the CPU.
As far as power supplies go, what you will usually find is that a 300W rated PS will only supply a portion of this to each voltage.
ie. 15Amps @ 3.3V = 50W
15Amps @ 5V = 75W
15Amps @ 12V = 175W
Total 300W
(These numbers are not reflective of a real PS, just made up off the top of my head).
If you exceed a rated current of any one of the above, the actual Voltage for that nominal voltage will drop. As 5V is used to supply current to the chipsets, CPU and some PCI cards it is the one most often at fault.
The reason for the above is that (usually, and I'm going back to power electronics at Uni - which I didn't like) the DC voltages are derived after the AC voltage is passed through a transformer. Each voltage starts off by being tapped at different stages of the transformer winding ie. 240V to 12V is tapped @ 20:1. The voltages are then run through a rectifier (AC to DC) - which is the stuff on the PS PCB. These are generally made of some diodes and power transistors, which are only rated at Certain wattages - one stage for each voltage - and will limit the total power supplied. Thus, each rated voltage is derived by a different series of electronics and acts (almost) independently.
Sorry for my diatribe (I hope it's not too inaccurate and provides background to some of the questions).
For one, external perepherals can be either 5v or 12v. The yellow cable on a power cable is 12v, the red is 5v. Also, a sensor chip might support 2 core voltages but you may only have 1 cpu so only 1 valid core line. Often if this is the case the board maker uses the other -spare- core line to display the 1.5 or 2.5 voltage line.
As to the external stuff, I stand corrected. Late at night and I should have picked that myself.
As to the 1.5/2.5 line, I've never had a dual-CPU board, so this may explain why I've seen this - thanks for the info.
Originally posted by Leeua
wow....3.8!! Hehe, that is LOW!! Well, im running pretty stable so Im not too worried. Keep you guys updated though if anything starts going wrong.
difer from realyty we dont know how accurate is mobo mesuring voltage so u have to use tester so u get u real voltages :)
Originally posted by Tiger
I have had a lot of problems with PSU's and did some research on the subject some time back.
As a general rule the closer the +5V rail stays to the 5V the more stable the system will be. I have found that once the +5V drops below 4.9 then problems start occuring in the form of blue screens etc. What I have also found is that certain brands of reputable PSU's seem to "go off". I have had a 300W Sparkle and a Macron and both were fine for a month and then just degraded. Have a 350W Enermax which has been fine for a few months but I notice that the +5V is starting to drop.
The following may help;
http://www.procooling.com/articles/html/picking_psu_s_for_athlon_syste.shtml
But basically the story is that it all relates back to the power that each rail can supply. The more devices that are attached to each the more the power drain P=IV !!!!soposed 2 be (U) !!!!!! P=Power (Watts) I=amps V=Volts. So P is directly proportional to V and I is a constant. If the power requirement increases then the volts will drop.
man P=I*U (not v=volts u=curenci in volts) and R=U/I cant get more cleare no englishman :D
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.