Metal or Plastic Standoffs ?

[sturoc]

I am trying to find a definitive answer to this debate: Plastic or Metal standoffs for MB mounting to back plate ?
Seen a bunch of forum posters elseware swear to use metal and other experienced builders say plastic is fine.

The main argument for 'team metal' is having multiple grounds for the MB to use to chassis. i.e if your MB back plate is thoroughly grounded to case.

But 'team plastic' says 'hey the MB already has multiple grounds thru the PSU connection, the PCI cards jack plate, etc.'
and how good is the back plate ground anyways thru all those little screws to the case ?

So which is it and your reasons for that guys ?

[parsec]

Ah yes, the definitive answer, does one exist for a complex situation?

I don't have an answer for you, as I can see merit in both arguments. Let's look at the reality of what the real world gives us, and see if that helps.

1. I've never seen threaded plastic standoffs provided with a PC case or mother board, or even sold individually, but metal ones are commonly available. Metal is not needed for strength in this application, and plastic would be strong enough for this application.

2. Some cases I've used provide screws for the mother board, that are painted black, including the threads. I've also seen painted standoffs. Paint is an insulator, and if we needed an electrical connection through those screws, a bad choice. I use both plain and painted screws in PCs I use, with no difference I can see.

3. In electrical theory, the flow of current in a circuit, when multiple paths are available, will always be through the shortest path, given equivalent resistance, or the one of least resistance. Does that not imply that all available circuits/paths (grounds) will not have a current flowing through them, or be "used"?

4. I have never seen a warning or instruction stating that you must insure a ground between a mother board and the PC's case.

5. Electrical power to a mother board only comes from the PC's power supply. Obviously there must be ground connections maintained by the mother board between it and the power supply. Given that, why do we need more?

6. In audio circuits, it is believed that multiple, independent ground paths can create noise (hum.) That is not only for audio signal carrying circuits, but power source circuitry. In this case the use of a "star ground", where many if not all ground circuits/wires are connected together at one point (creating the "star"), and from there on one wire/conductor to the final ground/earth connection, is considered superior.

7. I think I have noticed in every mother board I've owned, that around the screw holes there is at least some solder or other metallic material on the top of the board, that would contact a screw. There is nothing like that on the bottom of the board. But do we really know if they are connected to the boards ground?

IMO, each of these "facts" point to either Yes or No need for grounding through mother board standoffs. IMO, the tally looks like this:

YES: 1, 7

NO: 2, 3, 4, 5, and maybe 6

Otherwise, the best I can find and makes sense is this: There are two kinds of ground, one to complete any circuit connection, and the second is the protective earth/ground. The first is essential, but is not related to connecting the motherboard in general to ground. The second is an added safety feature, given (assuming) that you can electrically connect the board to the case with standoffs and screws.

So, not essential and at best a safety feature.

[sturoc]

Good thoughts Parsec, here's my take on them:

1&2:
there are plastic standoffs available but of a different design, a push thru the mounting plate style, then the MB gets 'inserted' onto them / or screwed to those. Some MB case are painted on the interior so unless the 'ground ring' on the MB's pcb ground trace's are in contact with the screw head then the point is moot. And like you said a ground path just using the screw threads as contact points is minimal at best.

3, 4, 5,:
I just went thru this scenario with an Acoustic Bass amp vintage 1970. After a full once over & servicing by a reputable tech. I had him replace the two prong power cord with a grounded one. I got the amp back, plugged it in to an ext cord. With the amp still off, I was holding the instrument cord plug in one hand and picked up a splitter box that went to the soundboard with the hand ZAAPP ! I made a nice ground happen. Fortunately I have good heart ! The extension cord used had a bad ( read as no ) ground connection. The mixer and recording gear was a mis-mosh of grounded and ungrounded power cords, UGH .
Moral is: Yes least resistance and shortest route.

So why would the MB want to be grounded thru screws, a plate, then the case ? Especially when the PSU is grounded to the MB thru it's multi pin MB connector and itself already grounded to the case ; aka.the shortest path ?

6, What is commonly known as the Ground Loop ( multiple ground sources ) is the Audio world's bane. Causing many the time honored search for the Hum/ Shock source in grounded & ungrounded band gear, PA,etc Why in the world would a PC have so many paths then ? Could it be due to a PSU failure that in some way causes a ground lift to the MB ? I don't see it happening, the way PSU's are wired.

7, One can just follow those MB traces from the screw holes and see where they go.
I wil ck my ASRock Z77 today and I bet they all go to the ground connection at the PSU connector.

IF a ASRock engineer could chime to address our points here, that would be interesting and much appreciated !

[Nux Vomica]

Currant will follow all available paths and not just the one of "Least Resistance". A 1 ohm resistor in parallel with a 1 Megaohm resistor, the total resistance is less than 1 ohm.

[wazza300]

metal/brass for me,the motherboard can have a trillion grounding points but its better to be 100% safe using metal/brass standoffs rather than cheap plastic

[parsec]

sturoc, I see what you mean about not being able to find a definitive answer, I have found ~50 - 50 claiming each idea is correct, and that from good sources.

I decided to test the seemingly conductive points around the screw holes of a mother board. I have a P67 board I'm not using now, stored in its box. Fired up my volt/ohm meter, unboxed the board, and had at it.

Short answer, the conductive points around the screw holes are connected to ground. I noticed that this board has small conductive dots around the screw holes on the bottom of the board, that would contact the standoffs. Extra effort for no good reason?

The details, the screw hole conductive points are connected to: All the other screw hole conductive points. The ground pins on the 24 pin power connector, and the eight pin CPU power connector. The ground pin of the fan headers. The ground pins of the SATA data ports/connectors. Etc.

So the mother board screw hole contacts can and do serve as ground points, and are attached to the common ground of the board.

Must a board be grounded to the case via these contacts? Thoughts...

I could not find a requirement for a mother board to chassis ground in the ATX spec.

While a board has that capability readily available, painted screws and cases seem to ignore that. IMO, the board manufacture knows better.

You can buy insulating washers to separate the screws and contacts. Why do that? Weird issues where isolating the board seems to solve the problem. IMO, that is a work around for an issue with something else, and should not be necessary.

IMO, grounding the board to the case via standoffs is not required for the board/PC to function. It is not a bad thing in the vast majority of situations, does not damage anything, and may do more good than we realize. I would not defeat it on purpose. I would rather have it, than not.

You can always use plastic standoffs, and create your own ground wire for the board, connected between the screw hole contacts and screw, and then to the PC case.

[synack]

The reason there isn't a definitive answer lies in the fact there are many variables. Let's look at and clear up a few points

Audio equipment deals with the large amplification of alternating current (the audio itself is alternating) and has a whole bunch of rules on its own that mostly does not apply here.

Path of least resistance means *more* current will flow in the path with least resistance, not all the current. It flows in all paths as long as all paths have resistance. (ie a path with zero resistance in theory will take all current)
Again probably not entirely relevant here either. Any single solid connection to "earth" ground is fine. I say earth ground as a battery for instance has a "ground" in its circuit that is not attached to earth ground.

As to the answer of metal vs plastic: plastic is fine as long as the power supply gives good earth ground.
This assumes of course nothing goes wrong with the power supply...
I would never want to ground my system through an add-in card... Yikes to those who suggest that!

The SAFEST way is to use metal stand-offs to ground your board. In fact every device should have its own connection to ground and not rely on the device its connected to.

Whomever posted the shock example demonstrated it best: when one or more "devices" have bad ground the whole system can become a death trap.
So my answer is: metal (conducting) standoffs are safest. Plastic (insulated) are fine if everything else is working properly.

[sturoc]

Parsec: "You can always use plastic standoffs, and create your own ground wire for the board, connected between the screw hole contacts and screw, and then to the PC case."

An option i've considered.

Synack it was me the 'op' who mentioned bout getting shocked and I should have clarified by saying that was in an AC power set up.
You points are well taken. Funny how there is no definitive guideline, at least none that can be researched online.
The only info seen seems to be in forums and other posts. None by the manufacturers and then only in the way of 'we recommend using standoffs ' for the obvious reason of preventing shorting the board out.
Kinda surprised that not many others here in this forum have not chimed in.

[parsec]

I understand the basic differences between audio components and PCs, but a good point of course. I mentioned that because we do have audio signals in a PC, that may be affected by the grounds. Yes, low level signals, no power amps on mother boards, but audio noise issues are common in PCs, most boards I have are terrible regarding analog audio noise. Noise added at any point of an audio signal will only get passed on and heard, except when using balanced/differential connections, which are not found in PC audio.

I've noticed that electronic devices that have AC voltage connected to a circuit board do not have multiple ground connections to the chassis, but only to the third wire safety ground on the AC source. That makes perfect sense and is the point of the third wire ground, to prevent the chassis from having AC voltage applied to it due to a fault. PC mother boards only have DC voltages connected to them (or should!) so the grounding situation is different.

My point about path of least resistance was oversimplified, and those that stated otherwise are correct. I was referring more to the situation where large amounts of current would find the lowest resistance path, and become potentially dangerous.

sturoc, not many opinions? IMO, most people don't have anything new to add. Also, this is really not a big deal, no offense. Grounding of a mobo is not an issue for 99.999% of PCs. It's not something we need to deal with, it's taken for granted. Have you ever seen an instruction anywhere that states anything about grounding a board? I haven't. IMO, there is nothing more to add to what we know now.

[sturoc]

thanks parsec and no offense taken at all.

Seems we've beat this into the 'ground'.
After my build is finished I'll post any details that may seem relevant.

thanks guys for the info and replies.