• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Safe voltage for Phenom II??

Joined
Feb 19, 2007
Messages
12,453 (1.85/day)
Location
Yankee lost in the Mountains of East TN
Processor 5800x(2)/5700g/5600x/5600g/2700x/1700x/1700
Motherboard MSI B550 Carbon (2)/ MSI z490 Unify/Asus Strix B550-F/MSI B450 Tomahawk (3)
Cooling EK AIO 360 (2)/EK AIO 240, Arctic Cooling Freezer II 280/EVGA CLC 280/Noctua D15/Cryorig M9(2)
Memory 32 GB Ballistix Elite/32 GB TridentZ/16GB Mushkin Redline Black/16 GB Dominator
Video Card(s) Asus Strix RTX3060/EVGA 970(2)/Asus 750 ti/Old Quadros
Storage Samsung 970 EVO M.2 NVMe 500GB/WD Black M.2 NVMe 500GB/Adata 500gb NVMe
Display(s) Acer 1080p 22"/ (3) Samsung 22" 1080p
Case (2) Lian Li Lancool II Mesh/Corsair 4000D /Phanteks Eclipse 500a/Be Quiet Pure Base 500/Bones of HAF
Power Supply EVGA Supernova 850G(2)/EVGA Supernova GT 650w/Phantek Amps 750w/Seasonic Focus 750w
Mouse Generic Black wireless (5)
Keyboard Generic Black wireless (5)
Software Win 10/Ubuntu
I'm looking for some credible information on Phenom II voltage. What is a safe max 24/7 voltage for these chips?
 
I just read on a Review page that they had one running with 1.57 volts to OC to 3.8Ghz. As for it being a 24/7 voltage, not sure.
 
you will have to wait a few months to get accurate data on the voltage.

Right now not many have tried the proc for long enough to comment on effects of high voltage.

But 1.5-1.6 for a 3.8Ghz OC, seems normal.
 
I just read on a Review page that they had one running with 1.57 volts to OC to 3.8Ghz. As for it being a 24/7 voltage, not sure.

Yeah, the numbers are all over the place, but I can't find anything to tell me what safe limits are. From what I heard, AMD 45nm chips handle voltage much better than Intel chips, but I'd like to find out a fairly safe 24/7 so I can find an OC I'm comfortable with. Right now, I'm sitting at 3.75ghz on 1.49v (real), though it's not been tested stable yet.
 
Well with my Phenom 1.5 volts is cutting it somwhat high, but I think like the safe limit was 1.4-1.48 with P1, P2 should be like 1.5-1.58. Just using logic, with time and as soon as I get mine there will be more detail on the P2.
 
Well with my Phenom 1.5 volts is cutting it somwhat high, but I think like the safe limit was 1.4-1.48 with P1, P2 should be like 1.5-1.58. Just using logic, with time and as soon as I get mine there will be more detail on the P2.

Hmm, not sure about that logic. I would think that since they are 45mn chips they would be more voltage sensitive.
 
As long as your cooling is up to keeping the temps below 70-75C~ full load I wouldn't worry too much about the voltage, from the various results I've seen over at XS you shouldn't need more than 1.5v unless you are planning to run the chip at its highest clock possible, some have managed 3.5GHz with stock voltage!
 
for 24/7 on any chip, be it a C2D or a Phenom, I would never go above 1.45V for 24/7. Its fine saying they have one running on almost 1.6V, but thats for testing, not everyday use.
 
As long as your cooling is up to keeping the temps below 70-75C~ full load I wouldn't worry too much about the voltage, from the various results I've seen over at XS you shouldn't need more than 1.5v unless you are planning to run the chip at its highest clock possible, some have managed 3.5GHz with stock voltage!

Temps are not really my concern. My concern is electron migration destroying the chip at high voltage.
 
Didn't AMD say they mastered the technique of something lithography and these didn't have as much or anywhere near the same voltage rules as c2d or i7.

This is something I read so don't take it to heart.
 
Didn't AMD say they mastered the technique of something lithography and these didn't have as much or anywhere near the same voltage rules as c2d or i7.

This is something I read so don't take it to heart.

Link? :)
 

Let me find it :) I've got a feeling its somewhere on the tpu news.

Ok I googled what was stuck in my head which was AMD masters immersion lithography and the other thing was lower current leakage which I think would mean it could take more volts or it used less volts.

http://www.extremeoverclocking.com/articles/press/Phenom_II_Preview_1.html < bit about the immersion lithography

http://www.amdzone.com/index.php/reviews/60/11033-amd-phenom-ii-x4-940-black-edition < bit about the lower current leakage

Neither of these actualy say it can withstand more voltage but I'm still looking for that source so I will edit this post when I find it because that is also stuck in my head.

Here's a discussion on http://www.xtremesystems.org/forums/showthread.php?t=211571 < wouldn't rely on 100% but it shows I'm not the only person with the same thinking.
 
Last edited:
I'm looking for some credible information on Phenom II voltage. What is a safe max 24/7 voltage for these chips?

Good question... I don't think anybody knows yet but time will tell...

I would guess no more than 1.45v for 24/7 usage...
 
I haven't saw any dead PII chip around any forum yet. Highest I saw was 1.65V water.
 
highest i see (and probably all u see) was 1.9 when amd overclocked it to 6ghz
 
Anyone find anything else regarding PII voltages?
 
Well, if that is AMD's official max, then it probably can take at least 1.6v with good cooling.


The AMD max voltage figure stated probably assumes stock AMD/average performance cooler, 62C max temp threshold seems rather low, maybe a deliberate built in +10-15C margin on that figure so people don't jack up the volts + clocks far too close to the actual safety limit, minimise any necessary RMA returns.
 
Last edited by a moderator:
AMD's specifications...

Voltages 0.875-1.5V
Max Temp 62°C
Thermal Design Power 125W
Process Technology 45nm SOI
 
Yup, seen those. However, anyone who overclocks needs to know the real tolerance of the chip, which is always a good bit above spec. Anyone else with thoughts on this?
 
I wouldn't go to high over the stated 1.5 without some ridiculous cooling. I mean if them temps aren't going crazy at that voltage then I'd say go a bit further, but I wouldn't push to far... for me 1.6 would probably be my max
 
I wouldn't go to high over the stated 1.5 without some ridiculous cooling. I mean if them temps aren't going crazy at that voltage then I'd say go a bit further, but I wouldn't push to far... for me 1.6 would probably be my max

Well, I'm on water so temps are not a problem. I'm currently sitting at 29c while running Boinc, winrar hardware test, and an antivirus scan at the same time. :) I'm guessing that I could go to at least 1.6v before temps would even be a consideration. My concern is at what voltages do I start to be concerned with electron migration degrading the chip. I want to keep this thread going, and add information regarding voltage as we learn more.
 
Look at it this way:

When your CPU reach a certain voltage/speed, your cooling couldn't cool the core fast enough.
Between the core, and the cpu HSF the temp different would be 10-15C. And It will take, maybe 10-15 seconds for the core to cool down to the HSF's temp, which is why, extreme overclocker always have their CPU at negative temperature.

Run your processor with prime, the second option (Maximum heat, and power consuming) at ~1.55v for 8 hours, and to make it a lil harder, do it at noon when the air is hottest.

After 2 hours of running, use a temp stick, and measure the waterblock/heatsink temp. If it's above 55C, it's dangerous, the core temp probably alot higher.
 
Back
Top