earth grounding -- potential difference?

[longano]

greets all.
new here.. i apologize if i'm in the wrong
forum. but i was hoping someone could help.

in trying to isolate a little spurious noise from a data acquisition setup, i ended taking my voltmeter right to my wall socket. (the things we do at 3am)..

i'd like to know if what i've found is "normal":

standard 3prong wall socket..
measured 220v/50hz between 'hot' and 'neutral' (i am in Europe)
then i measured 220v/50hz between 'hot' and 'ground'

between 'neutral' and 'earthground' i read about 2 VAC

is all of this normal?

thanks in advance.
-tony

 

[RajT]

Your readings are all normal.

 

[Kiribanda]

Hello longano,
This is a typical “NOISE” problem.

Basically the potential difference should be 0 Volts, so that your Neutral and Earth are at the same zero potential. This 2 Volts is called the “Common mode Noise “ which appears between Earth and the Neutral terminal. One of the reasons being that your Earth terminal which is connected to the wall socket is not properly grounded. So if you connect an oscilloscope you may be able to find out how this 2 Volts is “dancing?

Therefore it induces noise on most of the microprocessor based equipment like modems etc. That may be the reason why you are complaining a “spurious noise from a data aq. setup”.

So the solutions are
1)Check your earth terminal connected with the socket and ensure that it is grounded properly.

2)Use a “Clean earth “ for all your data aq. equipment.

3)Make sure that very low earth fault currents are not flowing in any of the protective earth conductors (PE as per IEC) while not tripping the associated Earth Leakage (RCCD) devices.

Regards!

 

[longano]

my first step tomorrow will be to bring in an earthground from outside. hopefully this will clear things up.

question: the main component in my DAQ setup is a vibrator/shaker to drive a vibration table. the vibrator is 3Ph/380V. it is delta-wired.

today i measured 38VAC between the motor ground (which is also connect to my "system" common ground) and any of the 3 powerfeeds.
ie.. between U, V, or W and ground, i am measuring 38VAC. (trueRMS multimeter, not a scope)

is this typical?
shouldn't my 3ph motor have FIVE terminals? one for each phase, one for neutral, and one for ground?

thanks again
-longano

 

[Kiribanda]

Hello longano,

1) Yes, you are correct. Tomorrow the difference may 10 Volts. So it shows that your “Ground” is not a true ground as you assume. The ground wire connected to your motor may be broken somewhere and it is no longer an earth wire, but it is now become a floating wire. Also there may be an earth fault current flowing in one or more your protective conductor system (earthing system) which raises the potential to some value. Normally, it should not happen. Always when there is an earth fault current, the associated earth fault protection should operate in time thereby isolating the fault.

It is not typical. It should not to be. The difference should be ZERO.

2) Windings inside the motors can be either STAR or DELTA connected. That means – take the base case- if you imagine 3-coils inside a motor, then you have got 6 terminals coming out. They can be either DELTA connected from which you get again 3-terminals out or STAR connected from which you get 3-terminals plus another common terminal (3+1). This exercise is done inside the terminal box. Normally the 3-phase input supply is given to the 3-main terminals and the common point is LEFT OPEN. You are not allowed to connect it to ground unless it is a special case. But a separate ground terminal is provided in the terminal box for you to connect your safety ground so that if somebody touches the motor body, he will not get electrocuted since the body is at almost Zero potential. So to make it zero, we should design and maintain a good effective grounding system. Therefore you have got only 3-phase conductors plus an earth(PE) wire. But for a static load it is not so.

In your case this safety ground is floating. If you measure the same voltages at the main supply panel, you may get a good voltage profile with voltage between Neutral & Earth is almost zero.

Regards!

 

[DanDel]

longano, I was just curious where you were going to connect your earthground which you said you were going to bring in from outside, based on Kiribanda's opinion of the 2 volt differential between neutral and ground at your receptacle(which is probably perfectly normal).
You seem to understand just enough about power systems to be 'dangerous'. (Sorry, I don't mean to offend, but I don't believe that Kiribanda's advice is the best for your question and your response is likely to make things worse or dangerous).
I bet you will find that the receptacle is solidly grounded(which is the first thing to check, and this should probably be done by a licensed electrician). You can then check the difference between the neutral and ground at your service panel and see if there is 2 volts difference there.

As far as the industrial problem, are you really measuring '38VAC between the motor ground (which is also connect to my "system" common ground) and any of the 3 powerfeeds' as written or 380V?

A normal electrical system will always have some small value of voltage between ground and the neutral, depending on how far away from the neutral grounding point you are making the measurement. This can be due to induction or circulating ground current. Sometimes, you may have a broken ground, but this is not typical, and usually will result in a differential voltage of much higher value.

 

[alehman]

When DanDel said about bringing in a "clean" ground... This can create a very dangerous situation if used in place of the equipment grounding conductor.

The voltage you measure from neutral to ground at the receptacle may be due to voltage drop in the neutral conductor if it is carrying any current. (The ground would not have any voltage drop since it should not normally have any current). The voltage you measure is very normal for a circuit with load.

Most motors have no neutral connection brought out. So you should see three wires and a ground.

I don't understand the 38 volts unless your ground is open.

 

[jghrist]

I agree with DanDel. The 2 volts between the wall socket neutral and ground does not in itself indicate any problem. Any single phase load will result in a voltage drop (or rise, depending on how you look at it) in the neutral. This will result in a voltage difference with ground. The neutral and ground are connected at the service but nowhere else.

There is something wrong with the 38 volts between the motor U,V,W and ground. I would assume that the source voltage is 220/380 or 230/400 grounded wye. You should get about 220 volts to ground. Even if you had an ungrounded source, if you had only 38 volts from ground to one phase, you would have more than 220 volts to the other phases. You do have about 380 volts between the phases don't you?

 

[longano]

my apologies.. caught an error.. the inverter is NOT outputting 380V/3Ph.. its output is 240V/3Ph.

that is, my inverter can only output 110% of line voltage (220V) ie.. 220x1.1=242V

inverter is supplying 3ph electric vibrator, DELTA connection (WYE connection limits my available RPM/Freq range.. according to the manufacturer) "For 9,000 rpm, DELTA CONNECTION"

so, in brief:
INPUT 220V/1Ph
MEASURED: 220V between Neutral+Hot
220V between Hot+EarthGround
2VAC between Neutral+EarthGround

from what i understand.. so far so good.

OUTPUT 242V/3Ph
MEASURED: 242V between U&V, V&W, U&W
38VAC between U, V, or W + EarthGround

thats the part i don't get.

the EarthGround terminals in my inverter
are on the same bus (ie, brass block).. so
my line 1P line voltage, and my 3Ph inverter output have a COMMON earthground. somewhere in the mix i lost a neutral wire.. which i have come to understand is normal.

my fear was that my inverter was putting 38VAC on my earthground. but this isn't the case since BEFORE the inverter (in the circuit) i am only measuring 2VAC between Neutral+EarthGround.

so where is my problem?
apart from the fact that maybe i should've been a cabinet-maker.

how do i get rid of the 38V? *SHOULD* i get rid of it?

if i tied my inverter's EarthGround to my computer's powersupply ground.. will the noise go away? (which is the goal of all of this)

wow.. sorry if this ended up a bit longwinded.

-longano

 

[busbar]

Are the data-acquisition input connections referenced to some “ground,” like an equipment frame? Are they capable of being operated in a differential mode?

There is a important difference. Does the 240V 3ø inverter have a neutral terminal with its output terminals? If not, then given data-acquisition equipment limitations, the acquisition channel may need to be reconnected ø-ø. Please though, understand the personnel and equipment safety implications of such a change.

 

[longano]

the DAQ board can operate in either single-ended or differential mode. I'm set up in differential mode, as this should typically eliminate (reduce) ground noise. the problem is that (i think) i'm bringing 38VAC *directly* into the channel. differential mode benefits are nill here.

the inverter does not have a neutral terminal on output, no.

 

[DanDel]

I would expect that your output has no reference to ground(probably by design) and the 38V is capacitive coupling.

 

[longano]

re: capacitive coupling

can i get rid of that?

 

[DanDel]

Not necessary; it is usually not a problem. If you had a low-impedance meter(like a Wiggys), I'd bet that the 38V would disappear.
I know I'll regret this advice, but you could try to connect a load across the 38V(like a 120V lightbulb) and then put your multimeter across the lightbulb terminals. This should also cause it to disappear.

 

[jghrist]

I think DanDel is right again. Your meter must have a low enough impedance to drop the voltage of the phase it's connected to to 38 volts. The other phases to ground would be higher, but when you move the meter to the other phase, it drops to 38 volts and the other phases rise with respect to ground.

I assume that your DAQ input boards are not connected to the output of the inverter, which is just there for the 3ø vibrator. If this is the case, the DAQ is probably pretty well isolated from any noise from the vibrator. The 38 volts would have no effect on the DAQ. The 38 volts is not on the ground, it is on the phases of the inverter output.

 

[longano]

i appreciate all the feedback, thanks again.

not to be difficult, but my accelerometer is definitely picking this noise up. as *soon* as the accelerometer body touches my vibration table, my signal goes crazy.

perhaps the accelerometer is akin to my high-impedence meter?

two things happen: simply turning my inverter on affects my sensor signal (RF)..

mounting the sensor makes everything worse.

mounting the sensor while the inverter if OFF, everything comes up roses.

-longano.

ps.. i put a 220V lightbulb in the circuit as recommended and i've burnt my house to the ground.

just kidding.

 

[jghrist]

The front-end section of your inverter may be creating high harmonic distortion in the input voltage. You might try powering your DAQ equipment from a different circuit.

 

[DanDel]

longano, har-de-har-har...
But seriously, what sensor signal is being affected by this noise?

 

[longano]

DanDel,
the sensor is an accelerometer

http://www.omega.com

part# ACC103
connected to a (battery-powered)
amplifier before entering the DAQ.

provides 0-10V output where
10mV = 1g acceleration.

at low frequencies, the noise comes
into the card as a steady 10-20mVAC
with lots of ugly spikes.