Electronics Test Equipment Grounding

[MedSpark]

What are some recommendations and "ground rules" (ha), for grounding everything (oscilloscopes, spectrum analyzers, power supplies, etc.) in an electronics lab?

My plan is to use a 1/4" x 2" x 36" copper bus bar drilled to accept a couple dozen Panel Mount Banana Connector Binding Posts (Cal Test Electronics CT2232-5) and mount it to our equipment rack. Should this also be connected to the building ground directly, or is there a risk of transients being introduced by doing so?

Thanks in advance for the help!

[nanofrog]

All you need is the building ground for instruments that require it (i.e. 3 prong IEC connector or fixed cable plug). Just make sure the outlets are actually grounded (construction date matters in this regard, as does ground wires having loosened enough to become open).

[MedSpark]

All you need is the building ground for instruments that require it (i.e. 3 prong IEC connector or fixed cable plug). Just make sure the outlets are actually grounded (construction date matters in this regard, as does ground wires having loosened enough to become open).

So there's no benefit to grounding everything to a master ground locally? Why do essentially all of our test devices have a separate ground terminal on the back?

[nanofrog]

The short answer is not all grounds are the same. I know, this is confusing.

An Introduction to Ground: Earth Ground, Common Ground, Analog Ground, and Digital Ground might be of some help. 

This can particularly come into play when not all of the equipment in use is earth grounded (inputs or outputs, as the case may still be grounded to earth). Think linear lab PSU's for example; the enclosure is usually earth grounded (and a safety earth ground terminal is located on the front panel), but the power output is isolated. The reason this may matter, is the local ground that the circuit is running on is greater than 0V.

Hope this helps. 

[Someone]

All you need is the building ground for instruments that require it (i.e. 3 prong IEC connector or fixed cable plug). Just make sure the outlets are actually grounded (construction date matters in this regard, as does ground wires having loosened enough to become open).

So there's no benefit to grounding everything to a master ground locally? Why do essentially all of our test devices have a separate ground terminal on the back?

Many ground points "round the back" are from historical requirements to support regions without protective earthing (3 pin plugs) and are now obsolete. There are modern devices with separate floating reference connections and protective earths but they are typically only found on SMUs for low current measurement.

A ground bar can be a useful addition to a laboratory measuring tiny currents or needing extra isolation from RF but it has to be added as part of the building supply to work properly which is a job for an electrician with knowledge of your local requirements/laws/codes/etc.

[beanflying]

If it is for Medical use or you are based in a medical facility in the US then there seems to be some regulatory hoops to jump through.

In some cases floating outputs (power supplies etc) or inputs are standard but the option to ground them to mains earth is sometimes provided.

In the case of anything in a metal case always ground the case to earth. If you have a scroll through the Feeltech 6600 (cheap chinese Signal Generator). The unit was unearthed in a plastic case as standard and the output virtual ground was being sprayed with 80-100 VAC because of a poor power supply 

Not that you are likely to have that sort of gear in your lab but worth a look as some of the gear you may be testing or working on may not like it.

Also look at proper earth leakage protection if you don't already have it fitted!

[MedSpark]

If it is for Medical use or you are based in a medical facility in the US then there seems to be some regulatory hoops to jump through.

In some cases floating outputs (power supplies etc) or inputs are standard but the option to ground them to mains earth is sometimes provided.

In the case of anything in a metal case always ground the case to earth. If you have a scroll through the Feeltech 6600 (cheap chinese Signal Generator). The unit was unearthed in a plastic case as standard and the output virtual ground was being sprayed with 80-100 VAC because of a poor power supply 

Not that you are likely to have that sort of gear in your lab but worth a look as some of the gear you may be testing or working on may not like it.

Also look at proper earth leakage protection if you don't already have it fitted!

Nope, I'm just setting up a new electronics development lab for our medical device development company and was wondering if I should put in some sort of local grounding provision such that all of the grounds on our test equipment were at a closer potential. And, no, I only plan on purchasing higher-quality equipment.

I know it's crazy, but I literally went through every post on the massive EEVBLOG forum "show your workspace" thread and I saw a couple pictures where there was a horizontal ground rail on the wall with the grounds from all of the test equipment attached to it...

Here's an example of where I'm wondering if it matters... We have a Picoscope oscilloscope that is plugged into a two-wire AC adapter and a USB data cable, but there's no dedicated ground hooked up right now. I assume the only thing grounding the scope is the shielding on the USB circuit that's connected to the chassis of the computer, that's connected to the computer's PSU, that's connected to the power cord, that plugs into the wall outlet... Seems like there's a good chance that the ground potential could be slightly off compared to other 3-prong plugs connected to the same circuit, no? 

So, is setting up a master ground plate that each of the devices are directly connected to not a worthwhile thing to do?

[cdev]

Once (on the Internet) I saw a "ground" wire that fed into a Styrofoam cup filled with a small amount of earth, that a very small potted plant was growing in.

"somebody's idea of a joke" 

I laughed pretty hard.

[beanflying]

Given the nature of what I do in my shack with bit hits to the single power run's voltage into it the earthing is the least of my test gears worries. I have just refurbished a UPS to run some of that gear and to keep it running during outages.

Have I considered running a localised or dedicated Earth Stake YES (not into a styrofoam cup  ) But I haven't seen a case yet where it would help. The mains ECLB has however saved me some embarrassment on a few occasions. 

[Fungus]

Nope, I'm just setting up a new electronics development lab for our medical device development company and was wondering if I should put in some sort of local grounding provision such that all of the grounds on our test equipment were at a closer potential.

Got to wonder about somebody called "MedSpark" who is "setting up a new electronics development lab for our medical device development company" asking questions like this.

Here's an example of where I'm wondering if it matters... We have a Picoscope oscilloscope that is plugged into a two-wire AC adapter and a USB data cable, but there's no dedicated ground hooked up right now. I assume the only thing grounding the scope is the shielding on the USB circuit that's connected to the chassis of the computer, that's connected to the computer's PSU, that's connected to the power cord, that plugs into the wall outlet...

You could always grab a multimeter and find out instead of just assuming.

Seems like there's a good chance that the ground potential could be slightly off compared to other 3-prong plugs connected to the same circuit, no? 

If there is and you wire them together you'll get current flowing down that wire, right? Not good.

Better to find out why/where there's a difference.

So, is setting up a master ground plate that each of the devices are directly connected to not a worthwhile thing to do?

Not necessarily. You're just as likely to make things worse.