This is one of those cases where electrical practices on opposite sides of oceans differ significantly.
On Fri, Jan 1, 2016, at 12:29 PM, Bertho Stultiens wrote: > On 01/01/2016 05:47 PM, Dave Cole wrote: > > That's done all of the time. In fact it is part of the NEC (National > > Electric Code) that is followed (for the most part) in the US. > > Pretty much every house in the US is wired like that. (I'm not making > > this stuff up. :-) ) > > I agree with the connection at power entry in the house. That is quite a > different story. That is actually the same here in EU (most countries). > > I think that we should separate two things: > - house installation > - machine wiring > > House installation is pretty much standardized with a lot of rules and > reasons. Specifically to ensure referencing and protection. > > Machine wiring is different in that you can have scenarios where > references are moved, especially in a 2-phase system where you are not > using the neutral, which is the scenario we have here. 120V transformer secondaries (in the USA) need to be grounded on one side, even if they are part of a machine control panel. Such transformers are referred to as "control power transformers" and traditionally powered electro-mechanical control devices such as relays and contactors. Today they still power those things, and they also power AC-to-DC supplies that run the electronic parts of the control. The low voltage DC power distribution (typically 24V) is sometimes un-grounded, sometimes single-point grounded, for exactly the noise reasons Berthos mentions. But the 120V "control power" isn't used a the "reference" for anything - it is still "dirty" power, although not as dirty as the main power that might be going to a VFD or whatever. The main reason for grounding one side of the 120V control power circuit is for safety in the event of a primary-to-secondary transformer insulation breakdown. Machines in the US are powered by anything from 240V single phase to 480V three phase. An insulation failure in the transformer could put up to 480V on the control circuits if the secondary was ungrounded. With a grounded, secondary, it will simply blow the transformer primary fuse(s), which are already rated for the full primary voltage. Control circuit fuses, wiring, pushbuttons etc, etc, are typically rated 150, 250, or 300V. Not 480V. So all kinds of bad things could happen if 480V wound up on ungrounded control circuits. Another reason for grounding one side is that you have a distinct hot and neutral. Hot is protected by fuses or breakers, neutral is not. A typical industrial panel might have a half-dozen small single-pole breakers or fuses feeding multiple low power loads. If ungrounded, those would all have to be two-pole devices. > The reason for /not/ connecting the ground on a secondary winding is to > prevent a capacitively coupled ground path. Especially if you cannot > guarantee a 100% balanced primary-to-secondary side wrt. ground (*). > > What happens is that there will be a current in the ground connection > which causes an imbalanced current on the primary side (remember: > primary is 2-phase circuit without neutral reference). This is a > differential current discrepancy on the primary side. If you have an RCD > (which you should), then it can or will trip due to the current imbalance. RCD's are extremely uncommon in the USA for industrial stuff. In fact the term RCD itself is virtually unknown. "Ground fault" is the term used here. There are two kinds of ground fault protection. In residences, ground fault protected receptacles are required when near water (kitchen and bathroom sinks, outdoor receptacles, basements, and several other places). These GFCI's (Ground Fault Circuit Interrupter) are intended to protect people from shock, and have trip points around 6mA IIRC. In industry, ground fault relays are typically applied at the transformer, where the neutral is grounded. They are intended to protect against ground faults where the impedance might be high enough (and the fault current low enough) that normal overcurrent protection doesn't trip. But they don't protect against shock; the trip levels range from a few amps to a few tens of amps. They pretty much shrug off stray capacitive leakage. > > There are a lot of good reasons to tie one leg the transformer to ground > > besides to establish the safety ground and neutral as is common on the US. > > Intermittent faults to ground, with an ungrounded system, can cause the > > secondary of the transformer to fly way above absolute ground causing > > connected devices, or the transformer to suffer from insulation > > breakdowns. That's the extreme, but it can happen. > > Actually, the secondary should normally be floating. Most transformer > setups are to ensure galvanic separation and that means you may never > connect neutral to the secondary side. I don't think anyone is saying you would connect the primary side neutral to the secondary. We are saying that you pick one side of the secondary, declare that to be (secondary) neutral, and tie that side to safety ground. > Connecting ground on a secondary is a different scenario, where you most > often see the 0V (DC) potential connected to protective ground. This is > often done in the PC's power supply. The 0V (DC) to protective ground > connection prevents the scenario to which you refer to. > > The 0V (DC) to protective ground connection is repeated in many devices, > which have both power and signal connections. This gives rise to > ground-loops, which are to be prevented at all times and it is the cause > of the whole thread. DC supplies (usually 24V, almost always under 50V) are a completely different story. You are correct that single point grounding or floating DC supplies are better from a noise perspective. > > One of way too many references on the web. > > http://ecmweb.com/bonding-amp-grounding/basics-bonding-and-grounding-transformers > > It looks like the link is about a 3-phase system, where you would > connect the center point in a star-configured secondary to ground. That > is a different scenario. > Three phase or single phase doesn't matter. When a transformer delivers power at 120VAC one side is grounded. There are a few very limited exceptions in the code, such as: A control circuit derived from a transformer with a primary not over 1000 volts need not be grounded if: (1) only qualified persons will service the installation, (2) continuity of control power is required, and (3) ground detectors are installed on the control circuit. [250-21] Some more detail at: http://www.ecmag.com/section/codes-standards/motor-control-circuits-ground-or-not-ground -- John Kasunich jmkasun...@fastmail.fm ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
