[SOLVED] Does having traces to close to cooler mounting holes increase the risk of physical damage?

Banqu0

Distinguished
Mar 11, 2015
108
1
18,695
Just out of curiosity. I see traces on motherboards mere millimeters from mounting holes and am wondering if such a layout increases the risk of damge from forces imparted on the board from CPU coolers.
 
Solution
Once the cooler is mounted and vertical, the compression on the pcb changes the weight factor. There's very little actual downward force at the point of connection, it's mostly outward force because of the height of the cooler and the force vector applied by that height. The weight isn't on the board as such, it's hanging in the air.

If you hold 50lbs close to your body, you'll get the full 50lbs added to your weight, and your feet will feel it. You hold that 50lbs at full arms length, that 50lbs sees very little actual weight added to your feet.

I'd be more worried about tall coolers pulling the backing plate through the mobo horizontally, than whether the holes will stop downward pressure.

Karadjgne

Titan
Ambassador
You see the surface traces. What you don't see is the subsurface traces in the pcb layers. These generally will include a ground plate, hot plates for easier voltage transmission.

And no, motherboards are designed around push-pin cooler mount abuse, and cooler compression abuse, so mm of clearance is fine.
 

Banqu0

Distinguished
Mar 11, 2015
108
1
18,695
You see the surface traces. What you don't see is the subsurface traces in the pcb layers. These generally will include a ground plate, hot plates for easier voltage transmission.

And no, motherboards are designed around push-pin cooler mount abuse, and cooler compression abuse, so mm of clearance is fine.

What about a cooler with a back plate such as a Noctua NH-U9S? It weighs over a pound. Looking at the traces on an Asus B560M which is the board I just purchased, it's hard to imaging that not impacting the traces somehow. My older Asrock had what looked like some sort of reinforcement at the mount points for the cooler and some distance to the traces. The Asus just has holes.
 

Karadjgne

Titan
Ambassador
And the Deepcool Assassin III comes in at a whopping 1460g (3.22 lbs) and is fine on motherboards. They will all have backing plates, which spread the weight over the 4x mounts. The weight of the cooler is pulling outwards on the backing plate, that's what supports the cooler, not the motherboard. Just like putting a nail through drywall into a stud, no damage to the drywall unless the nail bends, and it takes a lot more than just over 3lbs to bend the mounting plate or screws.
 

Banqu0

Distinguished
Mar 11, 2015
108
1
18,695
And the Deepcool Assassin III comes in at a whopping 1460g (3.22 lbs) and is fine on motherboards. They will all have backing plates, which spread the weight over the 4x mounts. The weight of the cooler is pulling outwards on the backing plate, that's what supports the cooler, not the motherboard. Just like putting a nail through drywall into a stud, no damage to the drywall unless the nail bends, and it takes a lot more than just over 3lbs to bend the mounting plate or screws.

That analogy would make more sense to me if the cooler was anchored to the chassis rather than the backplate which is supported entirely by the board. I understand that the weight is spread out, but there's still a downward force exerted on each horizontal surface the board from the weight of the cooler hardware. In this case, it's the mounting holes. This isn't even factoring in leverage forces from the center of gravity being extended away from the anchor points. The U9S is only 4.9 inches tall, so this won't be as severe as the taller towers.
 

Karadjgne

Titan
Ambassador
Once the cooler is mounted and vertical, the compression on the pcb changes the weight factor. There's very little actual downward force at the point of connection, it's mostly outward force because of the height of the cooler and the force vector applied by that height. The weight isn't on the board as such, it's hanging in the air.

If you hold 50lbs close to your body, you'll get the full 50lbs added to your weight, and your feet will feel it. You hold that 50lbs at full arms length, that 50lbs sees very little actual weight added to your feet.

I'd be more worried about tall coolers pulling the backing plate through the mobo horizontally, than whether the holes will stop downward pressure.
 
Solution
....
If you hold 50lbs close to your body, you'll get the full 50lbs added to your weight, and your feet will feel it. You hold that 50lbs at full arms length, that 50lbs sees very little actual weight added to your feet.
....

mmm....so where does that mass (weight) go? it can't just disappear. In reality, when extended outward it's exerting a much greater torque against the base of the socket. To use your analogy: the 50 lb weight held close to your body (1 foot) exerts 50 foot/lbs of torque against the shoulder while extending it out to two feet it's now exerting 100 ft/lbs of torque against the shoulder: it increases with no change in mass. But if standing on a scale, there can be no change in weight since total mass (body + weight) didn't change.

@Banqu0 : there are circuit board design guidelines that the board designers will follow. They dictate things like circuit trace spacing and how close to an edge it can get. Both will vary depending on board material and the specific signal/frequency/voltage/current a trace path carries. CAD and layout programs keep track of those things and will not violate the guidelines unless explicitly over-ridden by the designer. Which would be highly unlikely and a sure way to get the boot once warranty returns pile up.
 

Banqu0

Distinguished
Mar 11, 2015
108
1
18,695
mmm....so where does that mass (weight) go? it can't just disappear. In reality, when extended outward it's exerting a much greater torque against the base of the socket. To use your analogy: the 50 lb weight held close to your body (1 foot) exerts 50 foot/lbs of torque against the shoulder while extending it out to two feet it's now exerting 100 ft/lbs of torque against the shoulder: it increases with no change in mass. But if standing on a scale, there can be no change in weight since total mass (body + weight) didn't change.

This is what I was thinking. Hold 50lbs of weight at arms length vs closer to the body is much harder on the muscles precisely because of this force. I can see this leverage action pushing the heatsink toward the board.
 
This is what I was thinking. Hold 50lbs of weight at arms length vs closer to the body is much harder on the muscles precisely because of this force. I can see this leverage action pushing the heatsink toward the board.
That used to be a problem in the early days of huge tower heatsinks. But modern designs are both lighter (less use of copper) and as well the large backplates used spread the forces over a large area. I would trust any Noctua design as they're well engineered and very reliable. An NH-U9S may seem large and heavy but it's actually quite small and lightweight compare to their NH-D15, with two 140MM fans mounted to two towers. Those heatsinks are regularly suspended.... cantilevered actually...on CPU sockets and have been for close to a decade now.
 
Last edited:

Zerk2012

Titan
Ambassador
What about a cooler with a back plate such as a Noctua NH-U9S? It weighs over a pound. Looking at the traces on an Asus B560M which is the board I just purchased, it's hard to imaging that not impacting the traces somehow. My older Asrock had what looked like some sort of reinforcement at the mount points for the cooler and some distance to the traces. The Asus just has holes.

http://www.cryorig.com/r1-universal_us.php

Almost 2.7 LBS for my cooler no problems.
 

Banqu0

Distinguished
Mar 11, 2015
108
1
18,695
Just to clarify what I'm talking about, this is a closeup of my motherboard. The plastic spacers for the Noctua U9S will definitely be covering those traces. Basically any cooler's mounting hardware is going to touch those traces and put pressure on them once tightened. The dead weight of the cooler is one thing, but pressure from the mount itself is another matter. This can't be a sensible design. Why risk force being applied to delicate structures like these? I've noticed this on a lot of boards.


oFcgnEn.png
 

qiuzman

Honorable
Aug 26, 2018
5
0
10,510
Just to clarify what I'm talking about, this is a closeup of my motherboard. The plastic spacers for the Noctua U9S will definitely be covering those traces. Basically any cooler's mounting hardware is going to touch those traces and put pressure on them once tightened. The dead weight of the cooler is one thing, but pressure from the mount itself is another matter. This can't be a sensible design. Why risk force being applied to delicate structures like these? I've noticed this on a lot of boards.


oFcgnEn.png
I am curious about this too. I have traces on the other side as well and screwing in kind of messed some of the traces up but the board still works.