The first thing that comes to mind is static electricity.
If you work at any place that makes this equipment you will see ground straps on the wrists of people that handle the components or another way is through the use of conductive paint on the floor with the employees warring a ankle strap that has a lead to under their shoe.
On the way in the door there is even a device you stand on and touch with you hand to check the continuity of this setup.
At your home or office you will not be taking these precautions, so having a grounded case around the equipment keeps a static shock from frying sensitive parts inside.
Bob Badger Construction & Maintenance Electrician Massachusetts
A great deal of modern equipment is double-insulated and runs with no ground on the case. Take a look at most VCRs, for example, and you'll see that they have a two-wire cord, and the wrap-around metal case isn't connected electrically to anything.
Wrist-straps are commonly used when working with CMOS or other static-sensitive devices to prevent a static potential appearing between pins of the device. Once fitted into a circuit, the typical CMOS IC is protected from such differential-mode static discharge by the other components.
By the way, the grounded wrist-straps incorporate a high-value resistance somewhere in the connection lead for safety, typically 1 megohm or more.
Having a grounded metal case on a PC probably helps meet RFI/EMI emission requirements.
Not sure about the UK, but the metal case on every VCR I have looked at is connected to "circuit common". Most units have a 2-wire cord, with the neutral side of the line connected to circuit common through a typical 3.3 megohm resistance (for static discharge). The shells of the I/O connectors are also connected to the common bus, and to earth ground through the shield on the coaxial antenna input cable.
I had an experience with a residential computer system that was not grounded. It had a continuing string of unusual problems, seemingly totally unrelated. More as a last ditch effort than anything, we installed an independent grounded circuit back to the main panel for just the computer equipment. ALL the problems instantly disappeared. I was told by some engineer types that the switching power supplies common to computers need a ground reference to regulate properly.
Not sure about the UK, but the metal case on every VCR I have looked at is connected to "circuit common".
Oops! Sorry, I must have been in too much of a hurry last night and didn't quite say what I was thinking.
Yes, the case is generally linked to the circuit common here as well. What I meant to say, is that the case has no direct electrical connection to anything on the AC line side of the transformer, i.e. no equipment grounding conductor in the cord, and no metallic path to either the hot or neutral. High-value resistors connecting the circuit "ground" to the neutral are not normal practice in electronic equipment over here.
The grounded case on modern PCs certainly contributes greatly to RFI suppression, and the ground is also needed for the filters on the the power input to properly attenuate common-mode noise.
NEC article 647 is interesting in relating to a balanced supply. 647.4(D) specifies some quite close tolerances for voltage drop as well.
A few years ago a Silicon Valley start up company had an electronics lab with PCs, monitors, oscilloscopes and the usual whatnot. One bench was fed by an extension cord, 12/3, so someone at least avoided the 18 AWG hot wire problem. Another bench, by the wall, was powered more directly by a wiremold plug strip plugged into another wall socket. We were getting occasional intergrated circuit burnouts. We had some video and computer signal cables running between the two benches, and one day noticed I'd get a mild shock and a small amount of sparking when connecting or disconnecting these cables. traced it to an open ground on the 12/3 extension cord. What had happened was the PCs' and monitors' power supplies had small RFI bypass filter capacitors from both sides of the line to equipment grounds. Which mad a fairly high impedance 60VAC present itself on all the grounds of one bench. We replaced the bad extension (still likely an OSHA violation, but startups on shoestring budgets ... )
Although there will never be complete agreement on proper methods, “grounding” has been around for a long time. I think it’s safe to say it’s been around longer than “electric power” itself. Didn’t Ben Franklin ‘invent’ lightning protection using ‘grounding’ in 1752?
[War story — I remember the first time I had to “float” an oscilloscope on rubber-gasket sheeting to connect its ‘ground’ to a phase in a 480V/1500kVA unit-substation that served a switchboard-sized NEMA size-6YD magnetic starter.]