Can an ac and a dc circuit share a ground point?

[FionaZuppa]

sorry, could not edit in under teh 15min allowed

I would not expect to see either the positive or negative side of the power supply to be connected internally to the AC EGC. The grounding of the DC system would occur on the load side of the power supply output.

don, the question is, can AC and DC share the same ground? what exactly is the AC ground? its the egc. on the DC side what is the ground, its the neg terminal. so if they share same ground "plane" then you have essentially tied DC gnd to AC egc. i think that is a bad idea. i would build in isolation. the only "shared" ground would be earth.

but if the argument is around shared gnd between DC chassis and AC egc, i would also place caution there too, all depends on if the DC gear (not power supply) has DC neg tied to chassis. many times DC gear/circuitry uses chassis to reduce wiring. if i sink amps via my mosfet i dont need to bring a wire from fet back to dc neg term, i tie the fet source terminal right to the chassis because my DC power has neg tied to chassis, this the chassis itself becomes a big CCC, etc.

Remember, at SOME POINT, the "ground" connection will have to be the same... earth.

AC egc and DC gnd tied into earth is not the same as DC neg tied to AC egc.

both connected into the earth is like throwing a dart blindly into a potentiometer, both land on same "device" but at different potentials. the earth is not zero charge everywhere, etc.

 

[iwire]

but that just means "on the DC side". the common "gnd" for the DC circuits is the Neg side of DC source.

Not so with the NEC. The NEC does not specify which side to ground.

There are PV systems that are negative ground and PV systems that are positive ground. In either case the grounded conductor must be white.

 

[iwire]

so if they share same ground "plane" then you have essentially tied DC gnd to AC egc. i think that is a bad idea. i would build in isolation. the only "shared" ground would be earth.

Why do you feel that is a bad idea?

but if the argument is around shared gnd between DC chassis and AC egc, i would also place caution there too, all depends on if the DC gear (not power supply) has DC neg tied to chassis. many times DC gear/circuitry uses chassis to reduce wiring. if i sink amps via my mosfet i dont need to bring a wire from fet back to dc neg term, i tie the fet source terminal right to the chassis because my DC power has neg tied to chassis, this the chassis itself becomes a big CCC, etc.

Any equipment such as a power supply that takes AC in and outputs grounded DC is going to have a shared 'ground plane' with the AC.

 

[FionaZuppa]

Not so with the NEC. The NEC does not specify which side to ground.

There are PV systems that are negative ground and PV systems that are positive ground. In either case the grounded conductor must be white.

i knew this was coming. yes, it can be either side as common gnd.

 

[FionaZuppa]

Why do you feel that is a bad idea?



Any equipment such as a power supply that takes AC in and outputs grounded DC is going to have a shared 'ground plane' with the AC.

whats electrically common between the AC mains on left, and the DC on the right?? point #2 is a common "ground" on the DC side. you could tie point #2 to the chassis right at the xfrmer, thus you would have no need to wire C1 and Rl back to xfrmer, you just tie the bottom side of C1 and Rl to chassis <-- the common "ground". if the chassis is DC common ground you see no issue taking that DC chassis back to the mains EGC ??

 

[iwire]

whats electrically common between the AC mains on left, and the DC on the right??

Nothing in that particular schematic however it has been brought to your attention that some DC system do have a grounded conductor.

Most of the energy engagement systems I work on bond one side of the power supply to the enclosure. I suspect this is for the same reason the we do this with any source. To provide a fault current path to open OCPDs that will isolate the problem.

 

[iwire]

i knew this was coming. yes, it can be either side as common gnd.

Yet you stated that was not the case, an unusual way to debate but hey to each their own.

 

[iwire]

AC egc and DC gnd tied into earth is not the same as DC neg tied to AC egc.

both connected into the earth is like throwing a dart blindly into a potentiometer, both land on same "device" but at different potentials. the earth is not zero charge everywhere, etc.

Well they will be the same as we can't use dirt as a conductor.

 

[FionaZuppa]

Well they will be the same as we can't use dirt as a conductor.

there's a diff between common ground electrode, and having one for each system?

Nothing in that particular schematic however it has been brought to your attention that some DC system do have a grounded conductor.

Most of the energy engagement systems I work on bond one side of the power supply to the enclosure. I suspect this is for the same reason the we do this with any source. To provide a fault current path to open OCPDs that will isolate the problem.

one side of the power supply? which part, the DC part or the AC part? if you use chassis as common gnd for the DC devices that use DC power, there is no fault protection. if you use a 3 wire system and use the chassis common ground for fault protection then you need to wire everything like we do for AC. many DC systems are not that way, they are simply fused in some way. modern automobiles kinda went the other way, there is no real common ground any longer. they use common pos which is not tied to chassis, and sink the current via fet or bjt inside the ECU, which means they homerun wire back to ECU where the "common ground" point lives.

this is a good read, note the use of isolation transformer.
http://emersonprocess.com/systems/support/documentation/provox/docvue/System/Planning/pn003.pdf

 

[gar]

160805-1708 EDT

I can take a DC voltage source and connect it to a load with no defined relationship to earth, and for most self-contained circuits it will do its job without relation to earth. Quite possibly that circuit can be raised to 1,000,000 V + or - relative to earth and work just fine.

I might want to put a battery powered oscilloscope on the hot side of a 300,000 V transmission line to measure line current.

Now consider most computers and CNC machines. The electronics in these machines are seldom isolated from the machine chassis, they are solidly bonded. This means the DC supply of the electronics is connected to the machine EGC and ultimately to earth. This creates a big problem for RS232 communication, and to truely get reliable communication, and prevent electronic component damage there needs to be electrical isolation in the RS232 communication path.

One can easilly get 80 V or more difference between the CNC and computer EGC potentials under fault conditions.

Almost all non-battery power oscilloscopes have probe common (ground) connected to the scope EGC. A big problem if you do not know what you are doing.

.

 

[FionaZuppa]

160805-1708 EDT

I can take a DC voltage source and connect it to a load with no defined relationship to earth, and for most self-contained circuits it will do its job without relation to earth. Quite possibly that circuit can be raised to 1,000,000 V + or - relative to earth and work just fine.

I might want to put a battery powered oscilloscope on the hot side of a 300,000 V transmission line to measure line current.

Now consider most computers and CNC machines. The electronics in these machines are seldom isolated from the machine chassis, they are solidly bonded. This means the DC supply of the electronics is connected to the machine EGC and ultimately to earth. This creates a big problem for RS232 communication, and to truely get reliable communication, and prevent electronic component damage there needs to be electrical isolation in the RS232 communication path.

One can easilly get 80 V or more difference between the CNC and computer EGC potentials under fault conditions.

Almost all non-battery power oscilloscopes have probe common (ground) connected to the scope EGC. A big problem if you do not know what you are doing.

.

good point. and yep, my scope's probe gnd is tied to AC egc. yikes. my scope is both AC and batt powered. i use batt power as often as i can
have a read http://electronics.stackexchange.com/questions/27066/why-on-earth-are-o-scopes-earth-referenced

 

[rlundsrud]

This may be off topic or not apply to this case, but there are many BAS devices like differential pressure transmitters that have a 24 volt AC input and a 0-10 volt DC output that use the same common wire. That is to say the negative wire on the DC side is the same as the 24 volt AC grounded conductor making it a 3 wire system.

 

[FionaZuppa]

This may be off topic or not apply to this case, but there are many BAS devices like differential pressure transmitters that have a 24 volt AC input and a 0-10 volt DC output that use the same common wire. That is to say the negative wire on the DC side is the same as the 24 volt AC grounded conductor making it a 3 wire system.

does the common CCC (aka "common gnd") tie directly back to the mains EGC ?? my guess is nope.

if i take a battery and i drape one end of it across the N (N or EGC, same thing) wire of house mains (outlet, etc) and continue to create a DC circuit with that batt, the current will flow flow from batt + through circuit to batt neg. but since the DC circuit is not isolated from the other AC power source do you see a hazard there ??? AC N decides to lift or become higher ohms, hmmmmm, sounds bad to me. let alone the noise you now coupled directly into the DC system(s).

remember, household AC is 1ph 240v CT, and the CT is the ground reference. where's the CT in a isolation xfrmer ?? no CT so you have just two wires with one side now the reference ground.

so yes, you can have AC and DC share a common ground plane, i just would do that by tying a DC common ground to the EGC of a std residential service.

 

[don_resqcapt19]

sorry, could not edit in under teh 15min allowed


don, the question is, can AC and DC share the same ground? what exactly is the AC ground? its the egc. on the DC side what is the ground, its the neg terminal. so if they share same ground "plane" then you have essentially tied DC gnd to AC egc. i think that is a bad idea. i would build in isolation. the only "shared" ground would be earth.

but if the argument is around shared gnd between DC chassis and AC egc, i would also place caution there too, all depends on if the DC gear (not power supply) has DC neg tied to chassis. many times DC gear/circuitry uses chassis to reduce wiring. if i sink amps via my mosfet i dont need to bring a wire from fet back to dc neg term, i tie the fet source terminal right to the chassis because my DC power has neg tied to chassis, this the chassis itself becomes a big CCC, etc.

AC egc and DC gnd tied into earth is not the same as DC neg tied to AC egc.

both connected into the earth is like throwing a dart blindly into a potentiometer, both land on same "device" but at different potentials. the earth is not zero charge everywhere, etc.

If you ground one side of the DC power supply you are creating a grounded system and the code would require a connection to a grounding electrode. Also the conductor that is connected to the grounded side is a grounded conductor and must be identified just like any other grounded conductor.

In practice, for smaller DC power supplies the "ground" connection to the DC system is often the EGC from the AC supply to the DC conversion equipment.

I am not sure there is a provision that would permit you to use an isolated grounding electrode for the DC system. My understanding is that you would tie the DC grounding electrode to the AC grounding electrode or use the same grounding electrode for both. Either way there would be a direct physical connection between the AC EGC and the grounded DC conductor.

 

[don_resqcapt19]

A 2 wire DC system having a voltage greater 60 volts is required to be a grounded system per 250.162(A).

 

[FionaZuppa]

please clarify for me.

i have a box, its a "DC power supply", the input is 1ph 120vac 60hz, the output id 75vdc.

when you say the DC has to be a DC grounded per NEC, are you talking about the input side of the box ?? you do not need the AC EGC to provide fault protection on the DC side of the box.

i have a box (dc power supply), input is 120vac 60hz 3-wire with proper egc. the DC is 75vdc 2 wire. the DC is electrically isolated from the AC via xfrmer. ok, so now two big wires run from the box down a raceway into a cabinet full of telecomm gear. are you saying that the telecomm box and gear needs to be bonded to each other and to DC common ground (typically the neg terminal), and then that tied to the power supply's 120vac EGC?

 

[jim dungar]

....two big wires run from the box down a raceway into a cabinet full of telecomm gear. are you saying that the telecomm box and gear needs to be bonded to each other and to DC common ground (typically the neg terminal), and then that tied to the power supply's 120vac EGC?

Yes, absolutely, unless you are a utility which is effectively exempt from the NEC.

For a while, I assembled industrial control panels. Each panel contained a 24VDC output power supply. In every panel we grounded one of the DC terminals directly the the same ground bus as the incoming 480VAC power circuit. This grounded terminal was 'white with a blue stripe'.

 

[FionaZuppa]

Yes, absolutely, unless you are a utility which is effectively exempt from the NEC.

For a while, I assembled industrial control panels. Each panel contained a 24VDC output power supply. In every panel we grounded one of the DC terminals directly the the same ground bus as the incoming 480VAC power circuit. This grounded terminal was 'white with a blue stripe'.

hmmm, is the proper way to homerun this DC "egc" back to a earth ground that shares the AC point of earthing?

i keeping hearing noise when tying a DC system power line to AC system.

i do need to investigate further. thanks.

 

[jaggedben]

please clarify for me.

i have a box, its a "DC power supply", the input is 1ph 120vac 60hz, the output id 75vdc.

That would probably fall under the exception for a rectifier derived system, and not require grounding.

when you say the DC has to be a DC grounded per NEC, are you talking about the input side of the box ??

No, the DC output, (if it doesn't meet the exception).

i have a box (dc power supply), input is 120vac 60hz 3-wire with proper egc. the DC is 75vdc 2 wire. the DC is electrically isolated from the AC via xfrmer. ok, so now two big wires run from the box down a raceway into a cabinet full of telecomm gear. are you saying that the telecomm box and gear needs to be bonded to each other and to DC common ground (typically the neg terminal), and then that tied to the power supply's 120vac EGC?

Most likely any of these AC powered supplies meet the rectifier exception. However, if you have a supply that does not meet that exception (e.g. battery, PV panel, dynamo) then yes to all of that, except that all the bonded stuff needs to be tied to a grounding electrode, not an AC EGC. If there are any AC supplies in the structure that are also required to be grounded then they'll all have to be tied to the same grounding electrode system.

 

[dmxtothemax]

Of course they can !
Look at a modern power amplifier
Lots of dc and ac grounds
all working together at the same time !