6502.org Forum  Projects  Code  Documents  Tools  Forum
It is currently Sat Nov 23, 2024 5:25 am

All times are UTC




Post new topic Reply to topic  [ 9 posts ] 
Author Message
PostPosted: Mon Oct 01, 2018 5:31 pm 
Offline
User avatar

Joined: Wed Aug 17, 2005 12:07 am
Posts: 1250
Location: Soddy-Daisy, TN USA
I've read that "ground is not ground". I still don't totally understand that but that's never stopped me....lol

So, in my limited understanding....I have a power brick (wall-wart) that is either a switching PSU or a linear one (transformer). Both take AC and step it down to convert to DC. Most of the ones I have are not grounded that I see (two prongs).

Then you have the normal ground rail/plane in your circuit that is essentially isolated from the "real earth" ground. Meaning, it's not connected to a cold water pipe, earthing rod, etc.

Then you have "case ground". Which, again as I understand it, ties the metal casing of the enclosure (computer case, project box, etc.) to the circuit ground above.

That's a long winded paragraph to simply ask....should I tie my circuit ground to the mounting holes in my board? So that when screwed down, the case (or enclosure box) becomes grounded along with the circuit? So, should it be left isolated?

Thanks!

PS, I plan on eventually installing a panel mount DC barrel jack on the case so that I don't have to go directly to the board. I also plan on installing serial/parallel jacks. So I assume they would also be grounded to the case and internal board.

_________________
Cat; the other white meat.


Top
 Profile  
Reply with quote  
PostPosted: Mon Oct 01, 2018 6:34 pm 
Offline
User avatar

Joined: Mon Apr 23, 2012 12:28 am
Posts: 760
Location: Huntsville, AL
I am sure there will be others on the forum that can supply a better definition.

Ground is truly one of the worst concepts ever conceived for electrical/electronics. Over the years, I've come to realize that ground is better thought of as the negative reference plane of an electrical circuit. Ground generally refers to the fact that in most buildings and similar electricity-based facilities, the local voltage of the earth, or ground, is used as the reference against which the other voltages are measured. In a localized environment, the local ground will appear to be 0V.

This is not strictly true, and is one reason for power transmission systems to rely on balanced AC circuits and transformers to isolate local ground potential from the power source. Many years ago, a two phase pump was installed at the cistern at the bottom of the hill. The outside two wires from the transformer at the house was run down from the house to the pump house where the two "hots" were connected to the pump. An awful racket was the result upon initial application of power to the motor.

A few measurements with a 90V neon AC tester indicated that one "hot" relative to the local ground connection at the pump was way low in voltage relative to the same measurement for the other "hot". The problem was that there was approximately 40 V of ground potential between the house at the top of the hill and the pump house at the bottom. The solution to this problem was to run the "ground" from the house down to the pump. A local protection ground, i.e. "earth", could be provided at the pump house for protection of the local inhabitants.

In your case, the neutral (wide plug) from the wall is supposed to be 0V, or ground, for your standard 115V ac outlet. That connector and the "earth"/"ground" connector are essentially wired to the same point at the entry to your house. Thus, if no AC current were to be "flowing" in the neutral, the neutral and the ground should be at the same potential, i.e. a 0V difference. Because there is AC current flowing in the neutral from the devices plugged into the "hot"/"neutral" AC wire pairs, and these wires have resistance, when there is an AC load there should be a small difference between the neutral and the ground of 1-4 VAC (under normal usage).

If your device is fully isolated, then I would wire the mounting holes together, but not in a full loop, and tie that trace back to the common reference of the power supply feeding your board. If those mounting holes connect to a metallic case, there could be a substantial difference between that reference and the AC neutral. Therefore, to dissipate the potential difference between your supply common and the neutral, I would connect a large value resistor (DC) and capacitor (AC) to "equalize" these two references.

That being said, the fact that your supply plugs into your AC outlet using only two prongs, means that the neutral can be at a fairly high potential above the local "ground" if a fault occurs. Fully isolated supplies are great in many situations, but the equipment must also be constructed in such a manner as to prevent contact with it when a fault occurs. I personal prefer to have a three prong AC-DC supply, and to connect my equipment case and ESD components' conduction path to the third connector, i.e. "earth/ground".

There should be a very small amount of current on that circuit except under fault conditions. Hopefully, the fault conditions are such that the circuit breakers interrupt it before any significant harm comes to anyone or anything.

Summary: a circuit ground planes are just another voltage reference. They are really the circuit return current path. They can be at thousands of volts relative to the local ground.

Hope this helps, and I apologize if my explanation is not on target or not clear enough.

_________________
Michael A.


Top
 Profile  
Reply with quote  
PostPosted: Mon Oct 01, 2018 7:45 pm 
Offline
User avatar

Joined: Wed Mar 01, 2017 8:54 pm
Posts: 660
Location: North-Germany
The reason for the 3rd plug (the wide one) is in its name - protective ground. It is a provison (and mandatory for metal cases) that if the "hot" wire gets off its terminal - accidentally - it connects to the protective-grounded case and trips a fuse or (in modern electric wiring) a "fault current detector" which cuts both ac rails immediately.

This protective-grounded case should not be directly (hard-wired) connected to your “GND”. It is a common practice to have a resistor (which has a working voltage greater than line voltage, e.g. 250V) of 10 MOhm between “GND” and case. Parallel to that resistor an Y-capacitor (>= 250 Vac) of 1 to 2.2 nF is placed. That is to keep both potentials nearly same in case of fast voltage transients (“ESD”). The construction of a Y-capacitor is considered to be fail-save against internal shorts even under over-voltage conditions.


Regards,
Arne


Top
 Profile  
Reply with quote  
PostPosted: Mon Oct 01, 2018 10:56 pm 
Offline
User avatar

Joined: Fri Aug 30, 2002 1:09 am
Posts: 8544
Location: Southern California
In my work, I'm usually addressing this from an analog standpoint, and noise getting into audio circuits. The noise could be from the light-aircraft alternator, gyros, strobes, or from the computer in our unit that controls the audio circuits, or from a switching power supply.

Here's a common scenario: An installer puts one of our units in a small airplane, and mounts the headset jacks with their bushings connected to the airframe at that point, assuming ground is ground is ground, instead of making the only ground connection back to the intercom the one in the cabling we supplied with the unit. For the worst case, suppose the aircraft has the biggest engine option, requiring the battery to be in the tail of the aircraft to counterbalance it. The capacitor on the alternator to smooth its output voltage basically is the battery; but now all that noise current also runs through the airframe, and the airframe's impedance, although low, is not zero, and the current humps, multiplied by that impedance, develop a voltage. Now suddenly the headset microphones' signal is not referenced against what the intercom sees as ground, but rather against something else that has all this noise on it. Even if you were to short across the mic, there's a strong noise signal getting into the intercom's mic inputs.

Same with a music input. I had a customer who put a CD changer in the back of the plane, with a remote control, and brought the cables to the front and into our intercom (which has very nice performance for stereo music entertainment as well), but failed to observe this. This one was close enough to me that I could drive down and see his plane, a Beechcraft Bonanza, 6-place IIRC. Right away I saw what I suspected. Although the CD changer was mounted on the carpet on the back deck, the mounting hardware was metal, grounding it to the sheet metal underneath. I told him that would have to change. He took the screws out, and sure enough, the noise completely disappeared. So then he re-mounted it with insulating hardware. Happy customer.

Our built-in intercoms' metal cases do not go directly to the circuit ground, but go through a 1M or 2.2M resistor instead. (No capacitor. I don't want AC noise going through there.) That way you can discharge static when handling it when it's out of the plane; but the main ground connections to the radios and entertainment sources go through the connector and cabling.

If you have D/A and A/D converters onboard, you will need to join the analog and digital grounds only at one place, and have a power-supply filter capacitor right there and run the signal lines to and from the converter preferably right near that point, so you're not referencing analog signals against a supposed ground that has digital noise on it. It's possible to do this even with wire-wrap. This goes for our audio circuits powered by switching regulators too, even on units with no A/D or D/A converters or microcontrollers.

The other posts above address the matter of safety. I would just join the grounds where the power enters the board, not at all corners. Depending on what you're doing though, some of what I've said about noise may matter enough to worry about it. Electrical noise is certainly important for analog work, and maybe including video; but I know many of the hobby builders here aren't doing those.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?


Top
 Profile  
Reply with quote  
PostPosted: Tue Oct 02, 2018 3:04 am 
Offline
User avatar

Joined: Wed Mar 01, 2017 8:54 pm
Posts: 660
Location: North-Germany
We have to differentiate between two "different grounds".

The one I talked about is sometimes called "protective" ground and is related to the local "earth" potential.

The one Garth talked about is usually referred to as "signal" ground. This should be the reference potential for some other signal. If one forgets to consider the resistance of a wire that carries some current, the "connection" to ground is actually a voltage source with an alien signal. In cases like Garth mentioned only a fraction of a millivolt may have a severe impact. One way to reduce problems like that could be working with differential signalling.

In "usual" electronic circuits the "ground" is isolated from "earth" due to transformers (even in SMPSs). It is similar to running the circuit with a battery. The transformer causes the insulation. It is not a perfect insulation, but that isn't necessary. Using a SMPS you will have a higher hum voltage than a classic transformer (50/60 Hz) would produce - that may result in an uncomfortable tingling sensation when touching any signal (including ground). (I hope that we can omit special considerations when using "autotransformers"...)

If you use a metal case and inside of it there are wires running at mains voltages - then protective grounding the case is mandatory. Otherwise you could simply forget "earthing" as long as that hum does not matter.

The frequent hints other members had already given that you should consider a 4 layer pcb (having a true "ground" plane) or at least using very fat traces for ground (and V+) are of more importance.


Regards,
Arne


Top
 Profile  
Reply with quote  
PostPosted: Tue Oct 02, 2018 9:24 pm 
Offline
User avatar

Joined: Thu May 14, 2015 9:20 pm
Posts: 155
Location: UK
Ahh, Ground vs. GND vs. Earth vs. 0V

Back when the first electrical items ran from one wet cell (or a number of cells wired in series), there were two battery connections to the circuit, positive and negative.

But someone at some stage decided they needed a ‘centre tap’ connection. Now, what to call this third connection? Can’t call it positive, can’t call it negative, as we have those already.... :?

The second problem was that some people had the switching in the positive wiring. While others had the switching in the negative wiring.

Right, who said that? Was it you at the back that shouted “standard”? Don’t do such a silly thing again. Of course there were no standards, this was the “Wild West” where anything goes (or more often, didn’t go/work!).

Anyway, the end result is that with low voltage DC circuitry, whether based on the switching and control being in the positive or the negative circuit, the return path has become known as 0V or in some cases, as GND.

The ‘GND’ bit is misleading, as normally there is no connection to either mains ground/earth or to the soil/dirt outside and under your home or office.

But, in some devices/equipment, this 0V/‘GND’ is connected to either mains ground/earth or to the soil/dirt outside and under your home or office (using a ground rod/earth rod/stake).

If the device or equipment is mains powered and there is a risk of a single insulation failure resulting in the metalwork of the item becoming live at mains voltage, then as described above, the regulations in most countries requires that a safety electrical ‘earth’ or ‘ground’ connection is made to the metalwork of the case. This connects to the mains earth or ground connection.

I hope some of that helps.

Mark


Top
 Profile  
Reply with quote  
PostPosted: Wed Oct 03, 2018 9:35 pm 
Offline

Joined: Fri May 05, 2017 9:27 pm
Posts: 895
cbmeeks wrote:
That's a long winded paragraph to simply ask....should I tie my circuit ground to the mounting holes in my board? So that when screwed down, the case (or enclosure box) becomes grounded along with the circuit? So, should it be left isolated?

Thanks!

PS, I plan on eventually installing a panel mount DC barrel jack on the case so that I don't have to go directly to the board. I also plan on installing serial/parallel jacks. So I assume they would also be grounded to the case and internal board.

Would it be possible to use a plastic enclosure?


Top
 Profile  
Reply with quote  
PostPosted: Wed Oct 03, 2018 10:23 pm 
Offline
User avatar

Joined: Tue Oct 25, 2016 8:56 pm
Posts: 362
No, you should not attach the enclosure to your circuit ground. Unless there's a specific reason otherwise, you'd be best off using a plastic enclosure to avoid this issue entirely. However, if the case is metal, and there is any danger of high voltage (> 50V) coming into contact with said metal enclosure (i.e. if a wire came loose inside), you must attach the enclosure directly to Mains Earth for reasons of electrical safety.

_________________
Want to design a PCB for your project? I strongly recommend KiCad. Its free, its multiplatform, and its easy to learn!
Also, I maintain KiCad libraries of Retro Computing and Arduino components you might find useful.


Last edited by Alarm Siren on Thu Oct 04, 2018 6:52 pm, edited 1 time in total.

Top
 Profile  
Reply with quote  
PostPosted: Thu Oct 04, 2018 1:42 am 
Offline
User avatar

Joined: Thu May 14, 2015 9:20 pm
Posts: 155
Location: UK
Just to be clear. ANY exposed metalwork that may become live at a dangerous voltage should be earthed/grounded to mains earth/ground via a direct cable/wire/conductor. Regulations require the resistance between ANY part of ANY exposed metalwork to be a very low value to the mains plug earth/ground.

It is up to you if you decide to earth/ground the 0V/GND rail on your PCB/circuit board to the mains earth/ground. Either directly, or via a parallel resistor/capacitor. There are advantages and disadvantages. What else you connect (or want to connect) to it often determines the answer.

Mark


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 9 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 21 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: