AMD Processors
Decrease font size
Increase font size
Topic Title: Cooling mod info needed.
Topic Summary:
Created On: 05/07/2006 02:17 AM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 05/07/2006 02:17 AM
User is offline View Users Profile Print this message

Author Icon
zaphodb777
Junior Member

Posts: 3
Joined: 05/07/2006

Okay, last time my fan (an actual AMD cooling fan) began to stall, at least I was here to catch it ! It had slowed down to 2500 RPM (1/2 normal/nominal), and I got things shut down before the processor cooked. Cleaned and re-lubed it, but don't trust it /dry.gif" style="vertical-align:middle" emoid="<_<" border="0" alt="dry.gif" /> . (at least it was ball-bearing, if it were sleeved, a new fan would be in order!.

Now have new, larger AMD official heatsink which is going on this CPU, but the fan in this case is a sleeve , deffinately don't trust that. But guess what, I have a somewhat heavier 110 volt muffin fan, that delivers about 120 CFM that will fit on the sink. They don't fail, not like this weak 12 volt junk. So I am going to mount this megablower, and use it. BUT...

What I need to know is, this fan is set up to be grounded, and I was wondering, would grounding the heatsink to the case ground, or power ground cause problems for the CPU? (AMD Palimino core).

(5/7/06 EDIT: Just got ahold of a 2nd Barton, going back up in the world!

I know the CPU has exposed capacitors on it's top, and worry they may be grounded out. Also worried about a possible vcc-ground-loop ?

Any advice on this?

-------------------------
**THE KLINGONS ARE ATTACKING**
**THE KLINGONS ARE ATTACKING**
OWWW! Bad cyberkarma!
http://zaphodb.dyndns.org
 05/07/2006 11:53 AM
User is offline View Users Profile Print this message

Author Icon
CBI Elite
Senior Member

Posts: 1089
Joined: 06/23/2005

Just wondering, you seem to be on the way to fixing the FAN problem, but what material is your heatsink made-of? Alluminum or Copper?


Don't do anything drastic until you've looked into the market for a nice copper HSF combo. Zalman and Thermaltake are big names out there, and there are others. These bad-boys will beat the crap out of most stock coolers, so look around!

-------------------------

AMD Athlon 64 3200+ E3 @ 2.6GHz/1.44V | Asus A8N-SLi V1014
eVGA 7900 GS @ 525MHz/1,450MHz | <a href="http://ww
 05/07/2006 10:40 PM
User is offline View Users Profile Print this message

Author Icon
zaphodb777
Junior Member

Posts: 3
Joined: 05/07/2006

It's aluminum, but it's big. Sinking isn't the problem. Problem is fan bulls**t! Sick of 12V, and the load it puts on the PS, for very little airflow. Sick of chintzy ball bearings or sleeve bearings drying out and giving up the ghost.

Haven't any of these fan manufacturers looked at beveled thrust bearing designs? At least one thrust bearing should be used, because these fans generate, guess what, THRUST! (but too little to be of consequence unless it's thrust 24/7 for years on end. Which in my opinion, for servers, is a fairly realistic thing to expect.)

Problem is too many pointy haired bosses watching the bottom line and sending out crap components. Anyone want to re-cap their motherboard with Nichelon caps? I think not. How about G.E. or Malaroy? (I'd buy a motherboard BECAUSE of good component use like that, forget the benchmark. Does it work? Does it keep working? Oh yeah!

And I am burnt out on fancy coolers. Have a Gigabyte squirrel cage + heatpipe monster copper job sitting on the shelf. The heat pipes out-gassed, and ate my first barton . Will never ever trust heat pipes again. At least not at 5280 feet in elevation. Oh yeah, it will be a cold day in hades before I buy Gigabyte again. (it would have to be dang cold in order for their junk to work with blown heat pipes!

Believe massive, and reliable airflow is the trick, yes, copper would help, but airflow will help more. And I am not in it for the sake of overclocking, I need reliability, stability, and durability.

Have you ever seen a 110V muffin fan (this one is small, but the bearings are beefy! Weighs just a bit more than the heatsink itself! fail? I haven't unless lightning, or rain got in it (and then, sometimes, they even work after the rain got to them).

So, I still need to know if the heatsink can be electrically strapped to ground? Have heard it cuts down on RFI instability too. But, I want to hear some more stories about people doing it, and if anyone knows for sure about it.

Thanks,
Zap

-------------------------
**THE KLINGONS ARE ATTACKING**
**THE KLINGONS ARE ATTACKING**
OWWW! Bad cyberkarma!
http://zaphodb.dyndns.org
 05/08/2006 01:21 AM
User is offline View Users Profile Print this message

Author Icon
mtrupi
Senior Member

Posts: 638
Joined: 07/10/2004

quote:

Originally posted by: zaphodb777
So, I still need to know if the heatsink can be electrically strapped to ground? Have heard it cuts down on RFI instability too. But, I want to hear some more stories about people doing it, and if anyone knows for sure about it.


I can address it in general terms and AMD does in some of it's build guides as well. There should be no danger, but if it is without a lid make sure the die is electrically insulated. It should be already. Some motherboards will have a place specifically for grounding the heatsink. Using this can possibly reduce RFI but don't count on that. If your fan is grounding to the AC outlet directly, if it does anything to RFI I might expect it to increase. As for ground loops, plug everything into same outlet and you should be okay, but no guarantees. Can you isolate the fan from heat sink? This might be safest.
 05/08/2006 01:50 AM
User is offline View Users Profile Print this message

Author Icon
zaphodb777
Junior Member

Posts: 3
Joined: 05/07/2006

quote:

Originally posted by: Unknown
Can you isolate the fan from heat sink? This might be safest.



Actually, that would be a pretty mean feat! If I had some nylon machine screws, maybe...

Was going to make a "dongle" cord that went between the power supply and power plug, that provided the fan's power, and ground the fan to the case (which should be an electrically shared ground in the PSU anyway). Then I was going to hot-glue the plug into the PSU, so in order to even power up the machine, the fan must be running. (I knew keeping all those 8088 era monitor cords would come in handy some day!

The only groundloop I feared, was between the top of the chip, and the case ground. The fan itself is insulated, but just incase that insulation failed, I wanted it straped to ground. (This fan was originally out of some piece of IBM equipment, and they ground everything, so I thought, "Why not?"

I will however do a check between the heatsink and the case with a VOM set on the 20 V scale, then the 2V scale. If it shows no voltage, then I will do a check with the machine off on ohms scale, first on the 2Mohm scale, then the 2ohm, and if nothing shows, it would be a real confidence boost.

But I hope someone in here will say they've actually done this to a Barton core, and really put me over the top on confidence.

Thanks,
Zap

P.S. Wish the barton, and for that matter, all Athlons had heat-spreaders/lids like the 64 has.

-------------------------
**THE KLINGONS ARE ATTACKING**
**THE KLINGONS ARE ATTACKING**
OWWW! Bad cyberkarma!
http://zaphodb.dyndns.org
 05/08/2006 07:10 PM
User is offline View Users Profile Print this message

Author Icon
mtrupi
Senior Member

Posts: 638
Joined: 07/10/2004

quote:

Originally posted by: zaphodb777
But I hope someone in here will say they've actually done this to a Barton core, and really put me over the top on confidence.



Sorry, I haven't actually tried this out. Sounds like you have a good approach toward ensuring success. If you try it, post here how it turned out.
 05/08/2006 10:29 PM
User is offline View Users Profile Print this message

Author Icon
stealth13777
Senior Member

Posts: 895
Joined: 12/23/2004

*Quote* P.S. Wish the barton, and for that matter, all Athlons had heat-spreaders/lids like the 64 has.

I'm one of the guys who (if i could afford an athlon 64) would take that heatspreader off for a couple degrees celsius (overclocking). But I see where you're coming from. would there be a way to actually put rubber between the screw and the little hole the screw goes through in the fan when you mount it? That's what I would try to do. Your ways of testing to double check sound great so you should know very well if it works alright.

-------------------------
-The Desktop: Abit NF7-S V2 w/ the old AXP-M 2500+ (OC'd) and TT Volcano 12, 1 gig RAM, 37GB WD Raptor, 80Gb Hitachi, XFX NVidia GEForce FX 5700LE (OC'd), Enermax Aluminum Case and XClio 450watt psu, DL DVD-R/RW, DVD-ROM (+ floppy)
SB Audigy 2 ZS
-Everyday:HP Laptop 15.4WS w/Turion64x2 TL-56, 2gig, 160gb, Lightscribe DVD-R/RW, NVidia graphics, Vista
Statistics
112018 users are registered to the AMD Processors forum.
There are currently 1 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.



Contact AMD Terms and Conditions ©2007 Advanced Micro Devices, Inc. Privacy Trademark information