Motherboard grounding questions

All hardware or software issues that do not have a specific forum go here.

Moderator: The Mod Squad

Motherboard grounding questions

Postby LakaWaka » Sat May 14, 2016 12:16 pm

Hello,

I was reading a few posts, including this one

viewtopic.php?p=427572

and I am curious about my grounding, and if it would be an issue.


I am currently setting up a really tiny form factor build, using an ITX mobo.

I had purchased 5mm standoffs, and the standoffs are tall enough to stop any solder points from contacting, but the issue I found is that the CPU Heatsink screws from my Noctua cooler is actually taller than the 5mm standoffs.

SInce I have very strict height requirements, anything taller might be an issue, as the case almost doesn't fit.

So my question is...

CAn I forgot the standoffs, and just have the cpu Heatsink screws contact the metal case without issue?

I have read that the grounding points via standoffs could be important as a secondary ground, but I hear that your main ground is through the PSU grounding to the Mobo 24-pin connector.


Here is a picture of my motherboard

http://www.dont.ru/Asus-P8H61-I-H61-B3- ... t9ugq2hhj1

I am using the ASUS P8H61-I R2.0 I believe.

I heard that the grounding via the standoffs is different per motherboard, so I wanted to show mine.

It seems that the issue is making sure it doesn't short itself out by touching solder points and frying.

Since the CPI Heat Sink screws shouldn't have any current in them (I would assume), would there be an issue?

Someone did mention that there looks to be a circuit path near the screws, but I checked and there is a decent amount of space. so I don't think there would be an issue (even so it would just go to the case and ground, as it's not touching anything else connected to my mobo). The issue the other person mentioned was if the screws became loose over time and connected.


I also was thinking of velcroing the screws down to the case, so it doesn't move, and I don't need to drtill holes for the standoffs this way.


So I'm curious if anyone has any issues with me not using the standoffs since they are too tall? Or should I look into bigger standoffs and go that way, even though the case will probably be too tall?


I also hear that the IO Sheild is grounded, and since I wont be using that (leaving it open for now, but want to get a wooden cover for this case), I'm curious if there is any issue leaving it open?

I will take some pictures to show, in case my wording is confusing.

Any technical advice is appreciated, last site I asked had no real info, but claimed it would work, but when doing research there seems to be a ton of info to consider, to make sure that you wont hurt yourself or your equipment.
LakaWaka
Initiate
Initiate
 
Posts: 77
Joined: Sat May 14, 2016 11:41 am

Re: Motherboard grounding questions

Postby Karlsweldt » Sat May 14, 2016 3:33 pm

All power grounding and returns are through the main ATX power connector. Other grounding means for a motherboard are only for static suppression of stray RF and EMF, in or out.
Standard height of the stand-offs is 1/4 inch, or about 6.5mm. Should be minimal about 1/8 inch between the metal case and any board connection.
Very few motherboard models use the solder-ring mounting holes as a ground connection. The user's manual would indicate if so. Metal or plastic stand-offs can be used depending on what type of case mount provision. Plastic stand-offs with screw threads are weaker than metal.. and may become fatigued and fail. Only the "collar button" type are more durable. I would not recommend using Velcro as a means of mounting the board!
Only the solder-ringed holes in the motherboard are intended for case mounting. There is no circuit ground connection to them, only the holes being reinforced with solder to protect the board material.
There should be no electrical connection to the heat sink mount from the board. A fiber plate between the bottom mount and the motherboard is required. If there are height restrictions when using longer standoffs, then perhaps a different CPU cooler would be in order. Those Noctua coolers are a lot taller than others. Chance to look at the Dynatron® K987 or similar. Copper core, aluminum radial fins, and quiet.
F@H.. to solve mankind's maladies.. in our lifetimes!
Karlsweldt
Mobo-fu Master
Mobo-fu Master
 
Posts: 20659
Joined: Wed Nov 12, 2003 11:57 am
Location: 07438

Re: Motherboard grounding questions

Postby LakaWaka » Sun May 15, 2016 7:29 pm

>I would not recommend using Velcro as a means of mounting the board!

The case is horizontal, and can be okay without any mountining, the Velcro was just an extra measure of "metal isn't touching metal."

> A fiber plate between the bottom mount and the motherboard is required.

Are you saying that I'm supposed to buy one, because there is already a backplate that has been installed on the mobo.

> If there are height restrictions when using longer standoffs, then perhaps a different CPU cooler would be in order. Those Noctua coolers are a lot taller than others.


The noctua cooler is hte nh9i or w/e, and is a total of 37mm in height, and is the smallest cooledr I've seen to date.


So, I'm not sure of your answer, but it made it seem like the main ground is through the ATX connector, and that additional grounding is made for extra interference "just in case."

But you then talk about me getting extra standoffs.


initially I wanted to drill holes, but since the standoffs aren't useful, then I wouldn't really need to use them, or I could get standoffs taht are similar in size to the cpu screws, or a little bigger, so that I can still mount it properly(but that's for the future imo).


So, is it possible for me to get away without having to use the standoffs? From your answer it seems I could.


What about the IO area? Am I supposed to protect that, or other areas?


Thanks for any advice[
Last edited by LakaWaka on Wed Sep 07, 2016 1:00 am, edited 1 time in total.
LakaWaka
Initiate
Initiate
 
Posts: 77
Joined: Sat May 14, 2016 11:41 am

Re: Motherboard grounding questions

Postby Karlsweldt » Mon May 16, 2016 6:34 am

Any type of electronics product that uses a circuit board requires it to be secured to the case. Any movement that may cause a short-circuit could cause a fire or property damage, otherwise. Part of the Underwriters Lab rules for approval of any electrical-related device.
Use of Velcro may secure the board in place, but is not resistant to punctures by component lead ends. A hard plastic or fiber sheet would be required between the Velcro and case. And a full pad under the board may cause overheating of the board and components. Even if mounted flat, there should be free air space under the board.
A fiber or plastic "pad" is part of the CPU cooling mount plate under the motherboard. If part of the installation package, acceptable. No additional insulator needed. The traces on the board bottom have only about the thinness of a hair of protective coating, easily damaged.
The Noctua coolers I looked at all had a working head that was higher than standard coolers.. with vertical fans and radiator on an evaporator heat sink plate. Perhaps there are other smaller, more conventional models, but did not note them.
The type of CPU cooling design used for laptops may be a better plan.
There should be a good amount of space over a horizontal CPU cooler, for free air movement.. or the CPU could overheat quickly. Case venting is also critical, to expel unwanted heat properly.
Should be no issue if the heat sink screws contact the case, but the sensors may be connected to it for static shielding via their return leads. Could create a 'ground-loop' electrical problem.
The rear of the case where the I/O port shield mounts, with the motherboard ports, has to align exactly. Off by even a tiny bit may cause problems. The shield should be pressed into the opening securely. There is no need for current (load) grounding, only for RF and EMF static suppression. Part of the FCC rules state that a device must not emit excessive RF or other type of electrical or magnetic radiation that could interfere with use of other devices.
Your choice as to use of stand offs or not, but monitor the system for excessive temperature limits.
F@H.. to solve mankind's maladies.. in our lifetimes!
Karlsweldt
Mobo-fu Master
Mobo-fu Master
 
Posts: 20659
Joined: Wed Nov 12, 2003 11:57 am
Location: 07438

Re: Motherboard grounding questions

Postby LakaWaka » Mon May 23, 2016 1:42 pm

Any type of electronics product that uses a circuit board requires it to be secured to the case. Any movement that may cause a short-circuit could cause a fire or property damage, otherwise. Part of the Underwriters Lab rules for approval of any electrical-related device.


Interesting, so it's to protect the entire circuit, and not just the part that's connecting..... So securing the screws would not work?

Use of Velcro may secure the board in place, but is not resistant to punctures by component lead ends. A hard plastic or fiber sheet would be required between the Velcro and case. And a full pad under the board may cause overheating of the board and components. Even if mounted flat, there should be free air space under the board.



Well, to me, the "punctures" would exist if there was enough space to do so. Since there is quote a large space between the leads and the case, I should be fine... Plus, the velcro will be tiny sheets that will go directly under the screws, maybe I didn't make much sense on that. The velcro would be between the metal case, and the screws ,and no leads should be touching.

A fiber or plastic "pad" is part of the CPU cooling mount plate under the motherboard. If part of the installation package, acceptable. No additional insulator needed. The traces on the board bottom have only about the thinness of a hair of protective coating, easily damaged.


Yeah, there's a plastuc piece that was already installed. So that will insulate what exactly?

How would the traces get damaged, besides the screws not being secure, even though there is a good distance in case they moved. IT seems like the board was designed that way, and I'm not syure if more boards give a little space to the traces, just in case.


The Noctua coolers I looked at all had a working head that was higher than standard coolers.. with vertical fans and radiator on an evaporator heat sink plate. Perhaps there are other smaller, more conventional models, but did not note them.



http://noctua.at/en/nh-l9i

At a super-low profile of only 37mm, the NH-L9i is ideal for extremely slim cases

This seemed to be the smallest HSF design I could find.


There should be a good amount of space over a horizontal CPU cooler, for free air movement.. or the CPU could overheat quickly. Case venting is also critical, to expel unwanted heat properly.


Right now there is not much space between the top of the case, and the HSF. What I was planning on doing was to drill holes, or possible cut out the letters in "creative" to apply a direct airflow to my CPU, from the outside. Holes seemed like a better idea, due to less dust, but that also will limit airflow a bit.

Should be no issue if the heat sink screws contact the case, but the sensors may be connected to it for static shielding via their return leads. Could create a 'ground-loop' electrical problem.


Wuh..... Sensors to shield the heat sink or...? How would I know if these "sensors" exist, or I wouldn't....? I would assume they were visible....

The rear of the case where the I/O port shield mounts, with the motherboard ports, has to align exactly. Off by even a tiny bit may cause problems. The shield should be pressed into the opening securely. There is no need for current (load) grounding, only for RF and EMF static suppression. Part of the FCC rules state that a device must not emit excessive RF or other type of electrical or magnetic radiation that could interfere with use of other devices.

So the I/O shield is important? It doesn't really fit well with the back of the case, so I wanted to leave it open, or do a custom wood backing. It also doesn't really fit wit hthe I/O in the back, it's a really janky i/o shield.[....


So the shield is actually very important then? I read something that some you could plug into your mobo, and I think my old computer had that, but we are talking pre 2010, so it might not be needed anymore.

Your choice as to use of stand offs or not, but monitor the system for excessive temperature limits.


So the heating would happen due to the sensor issue on the HSF, or something else? If I use the standoffs, what would the difference be? Or are you talking about due to adding a plastic sheet or w./e for protection?



So to me, it seems that this case might be more of an issue... than not?


I don't mind drilling holes for the standoffs, but figured this would be best.

The HSF is very close to the top of the case, and hoping that outside air would cool it off. There are side vents to exhaust air. Originally I was thinking of intaking side out, and exhausting out of the top with the CPU HSF, as heat rises, but from what I read intaking air into the CPU, from the outside, would probably benefit more, and then I can exhaust air with my tiny fans I have bought.


I really want to use this case, but it seems there are a lot of issues, and I don't want to hurt myself, or the unit.


IT seems the best course of action would be to use the standoffs, get the I/O shield setup properly (not sure how I will do that), and make sure that everything is secured in place?


Or, from what I mentioned above, do you think I could get away with no standoffs?


Would you recommend me just buying a regular case?

I really do not want to give up on this, but I feel that it mighnt be very risky, but maybe not...


There are many poeple whop build custom cases, so it can be done...


Thanks for the help.
LakaWaka
Initiate
Initiate
 
Posts: 77
Joined: Sat May 14, 2016 11:41 am

Re: Motherboard grounding questions

Postby Karlsweldt » Mon May 23, 2016 4:14 pm

Noted that the NH-L9i CPU cooler is rated at 65W TDP (thermal design power). That equates to how much excess heat can be taken from the device, and maintain its normal temperature limit. A higher wattage CPU could cause failure.
If a CPU cooling fan has only two wires, then just the motor is connected, and is isolated from other physical points. If a 3-wire or 4-wire fan type, one lead is common to all features such as the fan's RPM sensor and the PWM control. Still isolated from any other physical point or circuit, but the metal assembly may acquire a static charge from the constantly moving air over its fins. So there may be a high-resistance connection (around 1 meg or so) from the assembly to the negative or return lead.
Even if two smooth surfaces contact each other and move about, there still would be an abrasive effect.. wearing the lesser coating away.
The coating on motherboards is a type of varnish or lacquer that is fairly durable, and intended only to protect the exposed traces and provide a minimal moisture protection for the board "sandwich".
Almost all types of electronic circuit boards will give off excess heat, and air circulation is designed into mounting same in a case or enclosure. Even when a motherboard is mounted horizontally in a case, the 1/4 inch of free space provides minimal passive circulation of air.
You could use an end-clipping pliers to cut all protruding leads flush, but do so without twisting the pliers.. or the connection may be damaged.
Good thought about a vent hole or grille over the CPU fan. But a filter of some kind would be needed to keep dust and debris out. Those thin "scrubbies" for dishes are suitable, and similar are used as air filter material for a lot of other needs.
F@H.. to solve mankind's maladies.. in our lifetimes!
Karlsweldt
Mobo-fu Master
Mobo-fu Master
 
Posts: 20659
Joined: Wed Nov 12, 2003 11:57 am
Location: 07438

Re: Motherboard grounding questions

Postby Twisty » Fri May 27, 2016 2:02 am

Hope I'm not getting the wrong end of the stick here.

Heatsink screws just touching the case unlikely to cause you any problem. Would be a problem if it is pushing up the motherboard against tension of mobo fixing screws and bowing the motherboard though.
Motherboard not screwed down, using velcro etc. Might work but likely to encounter problems, you really do not want the motherboard flopping around, it creates faff and there is risk of solder connections on underside shorting on case which could be really bad news.

I think there are quite a few options though
    drill holes in case/motherboard tray in same position as the heatsink screws - screws no longer push against metal and you can now access heatsink screws without removing whole motherboard :D
    Use taller standoffs
    Put washers between underside of motherboard and standoffs to lift it up a bit
    Use different screw for heatsink
    grind/chop screw for heatsink
I have now left the Building :tongue8:
Twisty
Black Belt 1st Degree
Black Belt 1st Degree
 
Posts: 1970
Joined: Fri Mar 02, 2001 1:01 am
Location: UK / Singapore

Re: Motherboard grounding questions

Postby Twisty » Fri May 27, 2016 2:24 am

Wait a sec...
Are you trying to put a PC inside the case for a Creative x-Fi?
If so I think that is awesome, but there are a few things you are going to need to think about.
Are you running an external PSU?
You may need to chop a hole in the top of the case, above the CPU cooler so it can suck in cool air (this will also help move air around the whole case), you can fill the hole with a fan grille. There are other things that are doable, such as adapting an old graphics card cooler to sit on top of the CPU, some are very low profile and could pull air from the side of the case rather than needing to cut holes in the top.
I have now left the Building :tongue8:
Twisty
Black Belt 1st Degree
Black Belt 1st Degree
 
Posts: 1970
Joined: Fri Mar 02, 2001 1:01 am
Location: UK / Singapore

Re: Motherboard grounding questions

Postby LakaWaka » Sun May 29, 2016 2:00 pm

Karlsweldt wrote:Noted that the NH-L9i CPU cooler is rated at 65W TDP (thermal design power). That equates to how much excess heat can be taken from the device, and maintain its normal temperature limit. A higher wattage CPU could cause failure.
If a CPU cooling fan has only two wires, then just the motor is connected, and is isolated from other physical points. If a 3-wire or 4-wire fan type, one lead is common to all features such as the fan's RPM sensor and the PWM control. Still isolated from any other physical point or circuit, but the metal assembly may acquire a static charge from the constantly moving air over its fins. So there may be a high-resistance connection (around 1 meg or so) from the assembly to the negative or return lead.
Even if two smooth surfaces contact each other and move about, there still would be an abrasive effect.. wearing the lesser coating away.
The coating on motherboards is a type of varnish or lacquer that is fairly durable, and intended only to protect the exposed traces and provide a minimal moisture protection for the board "sandwich".
Almost all types of electronic circuit boards will give off excess heat, and air circulation is designed into mounting same in a case or enclosure. Even when a motherboard is mounted horizontally in a case, the 1/4 inch of free space provides minimal passive circulation of air.
You could use an end-clipping pliers to cut all protruding leads flush, but do so without twisting the pliers.. or the connection may be damaged.
Good thought about a vent hole or grille over the CPU fan. But a filter of some kind would be needed to keep dust and debris out. Those thin "scrubbies" for dishes are suitable, and similar are used as air filter material for a lot of other needs.


According to this http://noctua.at/en/nh_l9i_tdp_guidelines


95W OK
Recommended with good case ventilation only.

I figured with direct airflow from my room, to the CPU, I should be okay. What temps should I worry about? I heard recently that 70C is where it starts to throttle, but my Custom sits around 30-40C mostly, depending on if my room is cold or hot, as it fluctuates a lot.

The fan has 3 I believe, with the ability to swap down to a 2-wire.


So if the metal gets a charge, is there anything to worry about, and would the velcro carry the charge then? I believe when I looked up Velcro, it said it isn't a conductor, which is one other reason I wanted to put another material between the metal ends, and figured Velcro would work.

Thanks for the useful tip on this though! Any articles you can recommend to read more? I like it when people know what they are talking about.

Why exactly do I have to cut any leads? Leads meaning the ends of the solder points, right? If so, they wouldn't be an issue, unless something got in between the mobo and case, which hopefully would never happen, and Idk why it would lol (granted I did think about putting my ssd card under my mobo at one point, but maybe that would be a very bad idea) I'm not sure if my 850 evo is plastic or not.... But Idk what would happen in that case? Is that a bad idea?)


I'm a little confused about your paragraph talking about the issue with the lacur. So if it is removed then it could cause issue with high heat? But that is if there is movement? I just need to make sure it's on a stable surface?

How bad would dust getting in through the CPU be? I hear grills and dust covers would increase heat, due to less airflow, so if I used small drilling holes, would that work? I also hear that bigger holes would provide better airflow, but at the cost of dust.

I also have Nylon I was going to use, but hear that conducts static a lot, so I think that would be a very bad thing to have around a fan. Would the fan knock the charge (which would be electrons, right?), and send it into the case, possibly causing damage?


Some people claim static isn't a big deal, but comments on this site lead me to believe that static is super bad, and should be taken with caution, as some claimed there is microscopic damage from not using anti-static material.

So that means we should always use anti-static stuff?


Overall, still not sure what I'm going to do, I don't think the leads will ever touch the case, as there is about 3-5mm between them and the case.

I'm also curious if I get a custom wooden backplate if that would be good for securing everything, and if there is an issue with a wooden i/o shield?


Also, any more comments on the I/O shield itself, and why it has to be lined up? Are there conductive points around the ports as well that could short at the end of the case?


Right now there is no back, it;s open to the world, is that an issue that it's just open? I would assume not since there's nothing touching.


Alos, what does the shield do? It's plastic, so it doesn't conduct, or...?


Thanks.

Hope I'm not getting the wrong end of the stick here.

Heatsink screws just touching the case unlikely to cause you any problem. Would be a problem if it is pushing up the motherboard against tension of mobo fixing screws and bowing the motherboard though.
Motherboard not screwed down, using velcro etc. Might work but likely to encounter problems, you really do not want the motherboard flopping around, it creates faff and there is risk of solder connections on underside shorting on case which could be really bad news.

I think there are quite a few options though
drill holes in case/motherboard tray in same position as the heatsink screws - screws no longer push against metal and you can now access heatsink screws without removing whole motherboard :D
Use taller standoffs
Put washers between underside of motherboard and standoffs to lift it up a bit
Use different screw for heatsink
grind/chop screw for heatsink



You're correct.

I did have some bend in the MB due to the screws, but I loosened them to a point it was okay. I didn't hardcore screw them either, just "snug" but it just bent... So they are a little less snug but should be fine.


Well, the thing is, the solder points would not touch the bottom of the case, regardless, as it's all flat basically, and I removed any of the "disturbances."

The velcro was just to secure the mobo, since it's going to be flat. I could see if I bump it, but I don't think the velcro would budge at all.


I was thinking of drilling holes for the Heatsink screws, but then the mobo would be too low, and short the case.....

I might use taller standoffs, but then my space is probably an issue. Right now it;s a little taller than the case, but I can make it work, I hope :P.

Idk about the other stuff...


Wait a sec...
Are you trying to put a PC inside the case for a Creative x-Fi?
If so I think that is awesome, but there are a few things you are going to need to think about.
Are you running an external PSU?
You may need to chop a hole in the top of the case, above the CPU cooler so it can suck in cool air (this will also help move air around the whole case), you can fill the hole with a fan grille. There are other things that are doable, such as adapting an old graphics card cooler to sit on top of the CPU, some are very low profile and could pull air from the side of the case rather than needing to cut holes in the top.


Yes I am :).

The PSU I am using is a Seasonic U1 server unit that will fit inside the case. I also found something called a "Pico PSU" which is super awesome and tiny, but since I am using a 95w CPU, it might be too much overall draw for the 160, so I will look into that for another build.

As for flow I really wanted to cut out the letter for Corsair, hook up some leds and get nice flow, but the issue is dust, and Idk if that could be a big problem pulling in dust directly to the CPU. So I figured some tiny holes would be best, but I hears that bigger holes will allow better flow, but it seems a grill will make things hotter due to less air flow so Idk....

I do have Nylon that I was going to use on my other custom, since there is direct flow through the case, but I hear Nylon conducts big time, so I'm curious if you guys recommend me NOT messing with the nylon at all?











Thanks all for the help, much appreciated.






I should take some pictures to show everything, this might clear up some issues.
LakaWaka
Initiate
Initiate
 
Posts: 77
Joined: Sat May 14, 2016 11:41 am

Re: Motherboard grounding questions

Postby Karlsweldt » Mon May 30, 2016 8:17 am

Ambient temperatures are important with equipment cooling. Could be two or three degrees more for each degree of ambient temperature that the equipment has to disperse. And dryer air is more acceptable to absorbing waste heat than humid air.
As to thermal limits of any CPU, follow the manufacturer's spec sheet. At least 5°C lower than the design maximum for thermal throttling or alarm is recommended.. could be more with some equipment.
Any computing device, radio, TV or similar emits RF and EMF on a broad spectrum. Shielding limits this radiation, and a grounding path may be needed to be effective.. be it Earth ground or electrical ground.

Velcro and all other fabrics are considered non-conductive.. when dry! But if wetted, could be conductive. Depends on the mineral content of the liquid.

Any structure that is susceptible to static build-up should have a slightly resistive grounding path to keep the charge build-up minimal.
Some structures do have a conductive coating or additive that disperses charge build-up to other parts, keeping charge differential to a minimum.. and avoiding spark discharges. Dispersing static charges is critical with circuits and chips that cannot tolerate voltages over 10 volts, or they instantly die.
We all are (or should be) familiar with being about to touch a doorknob or other metal device, and get a spark of maybe 1/4 inch or more that shocks us! Could be more than 10,000 volts in that spark.. but very little current.

When circuit boards are manufactured, the process is almost totally automated. Very little human intervention. Component leads are pre-cut for needed design, but after assembly the board passes over a 'trimmer' that ensures no leads are longer than desired. Should not be more than 1/8 inch. For normal builds, no need to trim those leads. But for specific applications, the need of no exposed leads may be important.. such as surface mounting with an insulating sheet.
Open a PSU, and there is a semi-clear plastic insulator under the circuit board and on the sides to prevent unwanted contacts.
That coating on motherboards and other circuit boards is intended to protect the traces from corrosion, as copper exposed to air wants to revert to its primal stage.. an oxide state. Usually a greenish color. But the coating also protects circuits from accidental contact with foreign objects that could cause shorts.
SSD and SIM cards do have limits on heat. Or they may literally fry internally. They need air circulation to stay 'healthy'.
F@H.. to solve mankind's maladies.. in our lifetimes!
Karlsweldt
Mobo-fu Master
Mobo-fu Master
 
Posts: 20659
Joined: Wed Nov 12, 2003 11:57 am
Location: 07438

Next

Return to General

Who is online

Users browsing this forum: No registered users and 2 guests

cron