# How do I explain it? Potential difference in the system discussed is independent

How do I explain it?? Potential difference in the system discussed is independent

Hi folks, - please view the attached image -
my colleague and I had a major disagreement in the office.
We both mutually put a lid on the discussion as we both felt that we needed more information in order to make the other understand if we were right or needed more information to understand and accept that we were wrong.
I have attached the image for your perusal.
In the image we have discussed an arrangement where there are two primary inputs,
The voltage from a sinusoidal AC source, the voltage from a sinusoidal Inverter, sourced through a battery.
I'd like to name the AC mains voltage as Va and the inverter Voltage as Vi
The frequency of Va is 49 Hz, freq. of Vi is 50 Hz.
Hence there is a continual phasor difference between Va and Vi is continually varying.
Vi is held at ~230 VAC rms, while the voltage Va may vary from 0 - 250 VAC rms.
One phase of the AC mains is connected to the Earth point and so is one phase of the Inverter mains. Making them both the neutral phases of their respective voltage systems.
In no case are the live phases of the two systems brought into contact of the other.

In this setup, I believe that the Voltage Va and its phasor relationship do not affect the Voltage Vi and its phasor relationship or power factor.
I tried to explain to my colleague that the two systems are independent as only one line is shared and neither form a complete circuit with the other system.

Somehow this is not sufficient to make him understand my stance.

My colleague believes that the diffenece in voltage and phasor relationship would be detrimental to the loads L1 and L2, and also the malfunction or loading of one phase would trigger the loading of the other. i.e., if the voltage Va from the mains fluctuates, the fluctuations would affect Vi as the neutral is common.

I am not able to accept his viewpoint on that.

• if I am right I request you to kindly help me make my colleague understand the same.
• if I am wrong I request you to kindly help me understand were I am going wrong.

Please I know that this post may sound simple and childish to you,
but the disagreement between us two colleagues is stalling a couple of our implementations in office.

Thank you for reading the post.
Take care...

#### Attachments

• Reference.JPG
21.7 KB · Views: 387
Last edited:

It is hard to read just the schematic to tell the real life connection. If you two separate system and only connect one wire from each system like your schematic to a common point that is earth ground, I cannot see there is any interaction between the two system. Then I have to agree with you. Put it another way, if the two system do not share any common return path, then there is no interaction between the two. Look at drawing A.

But if you wire in the way that one of the wire is carrying both current, then it is not ok. One will affect the other. YOu might not be able to see it on the schematic. This is what we call ground loop. YOu really have to have the wiring assembly diagram to confirm this. I show in drawing B and the red section of wire carry both current.

Looking at yungman's sketches I'd say he has wired things up before, both correctly and incorrectly, and learned from that.

Averagesupernova
Gold Member

I may be misunderstanding yungman, but I disagree with him and here is why: In the USA we have split-phase 120/240 VAC service in our residences. The neutral is the center tap on the transformer and the wire that comes from the transformer to the service panel in the house carries current from BOTH ends of the secondary winding, or both 'legs'. As long as conductor sizes are adequate for the current carried, no matter what kind of load (inductive, capacitive, resistive) we put on one leg, the other leg is not affected.
-
So vish, you are correct, your colleague is not.

But vishal210 stated the supply to be 250 volts 49Hz.
This is nearer European standard, not US.

Further the second supply is generated by the battery, not the mains, so any return current from this supply is injected into the already balanced mains neutral return as an addition.

Note it is very poor practice, in the European system, for the subscriber to earth the neutral as shown.
However the US split phase supply is usually subscriber earthed at the centre tap.

edit
Looking again at the original information I find it hard to countenance the level of interference proposed with the types of loads described. Further is there not a huge difference between the capacity of the battery generated supply and the mains?

Last edited:
Averagesupernova
Gold Member

Frequency and voltage make no difference. Nor does the source in this case. The simple fact is that wires are 'shared' in this manner all the time with no ill effects. Keep in mind that I am not saying that this practice is always acceptable. But I would say in the OPs example it will work fine.

Frequency and voltage make no difference. Nor does the source in this case. The simple fact is that wires are 'shared' in this manner all the time with no ill effects. Keep in mind that I am not saying that this practice is always acceptable. But I would say in the OPs example it will work fine.

He is giving an example, not a house hold wiring.

I know about the house hold AC, they take two of the phase as 230V, and each with the neutral to get 110V. We have no choice on this. But they do end at the main breaker box.

It is not good even for house hold wiring inside the house. I can tell you the "arc breaker" will not work if you share wire. Every circuit should have their own independent black and white wire back to the breaker box. If not, you are going to run into trouble sooner or later. I know because I put in new wiring when I renovate my house, I put in arc breakers and it popped!!! The contractor made mistake be sharing the white wire and I had to went in and made sure all the wires are individual pairs by adding back the white wires when needed.

People do it don't mean it is right. That's why I told people to install ground fault and arc fault detector to be sure, any problem usually show up with those put in. Then you can go in and fix it......You'll have a better chance of having the wiring done right.....Still no guarante.

I did a lot of system grounding before. You cannot all the problems cause by improper return and grounding.

Last edited:
Averagesupernova
Gold Member

Yungman, you're preaching to the choir....
-
So you are telling me that the wires from the transformer to the service panel, what you call 'household wiring', doesn't follow the same rules other electricity does?

Yungman, you're preaching to the choir....
-
So you are telling me that the wires from the transformer to the service panel, what you call 'household wiring', doesn't follow the same rules other electricity does?

I don't think they follow the rule at all!!! Electricians try to put the same amount of wires from each of the two phase, but still that is not a guarante the house is going to plug things in evenly. I bet the time when they set the standard, they never dreamed of sofisticated electronics plugging into the power lines.

Back to the grounding, I worked in a company that operated at least 10 of the CAMECA SIMS mass spectrometers. They are the same machine but each and every single one behave differently!!! They have been fighting these problems for a long time. One time it got so bad they came and asked me to take a look. I took out all the groundings of the system(each system have a few racks of electronics and the spectrometer pot and needed to be grounded). Redo the whole grounding using the single point grounding method and isolate each connecting coax with feride toroid and put it back together. You should look at the face of those scientist with their jaws dropped!!! Scared them to death when I ripped the grounding off!!! After I put the new grounding back, everything worked!!! It really become a specialty on these kind of grounding and signal integrity issue.

Averagesupernova
Gold Member

I really don't wish to argue with you any longer. I've watched some of your posts in other threads and there is no winning with you. I learned a while ago to never argue with an audiophile.

to all The ckt being discussed is not per say an actual housing wiring ckt, but yes the same applies. Also the reference to Earth is made only as standard safety process(not discussing the Earth here) and if needed you may as well consider a system without the Earth connected to the neutral, but please limit the discussion to transients, faults etc as would be observed across Li and L2.

If you two separate system and only connect one wire from each system like your schematic to a common point that is earth ground, I cannot see there is any interaction between the two system. Then I have to agree with you. Put it another way, if the two system do not share any common return path, then there is no interaction between the two. Look at drawing A.

But if you wire in the way that one of the wire is carrying both current, then it is not ok. One will affect the other. YOu might not be able to see it on the schematic. This is what we call ground loop. YOu really have to have the wiring assembly diagram to confirm this. I show in drawing B and the red section of wire carry both current.

The situation under consideration is Case A, Common node, not common conductor. Aren't case A and Case B that you have mentioned the one and the same. Just that a common conductor is included and not just a common node. So if the red wire is sufficiently rated for the sum of the currents to pass through there should be no problem.
But the question is not just about the safety of the ckt, but if harmonics induced in one ckt will affect the next ckt, or whether the systems will be immune to the harmonics of the other, considering CaseA for simplicity.
For convenience sake the Inverter is rated to 500VA, and it will shut off once the power rating is exceeded. Also the inverter is equipped with both Over Voltage and Under Voltage cut-off protections and also the Overcurrent is monitored.
Under such situation, (either Case A (common node) or Case B (common conductor)) would a fault in AC mains trigger a pseudo fault detection in Inverter ckt. Also is there a possibility of the fault actually transgressing from Va to Vi ckt. I do not find that a possibility, but I may be wrong about it.
In case B would the voltage levels of Va and Vi, and their respective ckt power factor relationships affect each other across L1 and L2. I do not think there would be an influence @ L1 and L2.

@Studiot, the issue is not just about the ckt being Ok. But I'd like to know how can I explain that the ckt is ok, to my colleague. I tried explaining that the ckts are still separate. He still contends that since the supply Va and Vi are AC, there would be harmonics induced by one ckt into the other, which I strongly disagree with him there.

I also conducted a small experiment by replacing the ckt with the secondary of two transformers, each connected to a separate phase with 120 degree phasor relationship.
The voltage waveforms across the transformers were stable and undistorted in the equivalent CaseA and also Case B connection, compared to the actual waveform on each phase, No earth was used on the secondary but one phase of each transformer secondary alone was connected to each other. - Please see attached image -
But unfortunately, that was not sufficient demonstration to remove the ambiguity. I would like to know how I may explain to him that the voltages and currents of each ckt are still independent, though not isolated maybe.
@StudiotI think the frequency of the system and the voltage levels are not points of influence.

@AverageSupernova But how do I explain the same to my colleague. What technical, physical concepts and may be fundamentals can I use to illustrate the same.

@yungman The Inverter has a fault protection of its own, the wiring and the fuse breakers on the Mains Ac, Va have a rating akin to the load requirements of phase requirements of the housing. I guess so the protection is taken care of here, I may be wrong, so please do correct me. But that is another issue.

@everyone Thank you everyone for having taken the time to go through the
thread. and thank you all who have contributed.. Yes, this thread is still open....

take care..

#### Attachments

• Experiment.jpg
37.3 KB · Views: 382
Last edited:
Averagesupernova
Gold Member

Vish, in the real world, there will most certainly be a response of some sort in one circuit based on behavior of the other circuit. This is because no conductor is perfect, there is an inductance in the conductors that will show up, etc. when sharing a conductor. When sharing just a node, it is much less likely that there will be cross-talk. Just how far do we want to take this? My example of the 3-wire system in use in the U.S.A. is an example that works quite well. It was not necessarily EVER intended to keep cross-talk (man I can't believe I'm using that term in a power system) between legs to the level required in the internals of sensitive lab equipment. So the 3-wire system used here in the U.S.A. is adequate for what it is used for. However, when dealing with microvolt levels from microphones and things of this nature it is much more suitable to maintain as much isolation as possible. So what is the context here? From the way you described the setup in your first post, I assumed that it was so similar to the 3-wire split-phase system that I described that you could safely say you are correct.
-
How to convince your opponent? I would show them this thread, but from what you've already posted I doubt it will help.

The situation under consideration is Case A, Common node, not common conductor. Aren't case A and Case B that you have mentioned the one and the same. Just that a common conductor is included and not just a common node. So if the red wire is sufficiently rated for the sum of the currents to pass through there should be no problem.
But the question is not just about the safety of the ckt, but if harmonics induced in one ckt will affect the next ckt, or whether the systems will be immune to the harmonics of the other, considering CaseA for simplicity.
For convenience sake the Inverter is rated to 500VA, and it will shut off once the power rating is exceeded. Also the inverter is equipped with both Over Voltage and Under Voltage cut-off protections and also the Overcurrent is monitored.
Under such situation, (either Case A (common node) or Case B (common conductor)) would a fault in AC mains trigger a pseudo fault detection in Inverter ckt. Also is there a possibility of the fault actually transgressing from Va to Vi ckt. I do not find that a possibility, but I may be wrong about it.
In case B would the voltage levels of Va and Vi, and their respective ckt power factor relationships affect each other across L1 and L2. I do not think there would be an influence @ L1 and L2.

Even if the red section of wire is big and is super conductor, there would be no DC voltage drop and therefore no DC coupling between the two circuits. BUT you are running AC and a wire is an inductor at AC. So you do have impedance along the wire and will have AC voltages of the two signal on it. Therefore the signal start to cross couple between the two. That is where most of the problems of grounding in the systems now a days.

Yes AverageSupernova, the system is going to share a node and then part of conductor on the mains ckt, not the inverter ckt. As put in the first post
The frequency of Va is 49 Hz, freq. of Vi is 50 Hz.
Hence there is a continual phasor difference between Va and Vi is continually varying.
Vi is held at ~230 VAC rms, while the voltage Va may vary from 0 - 250 VAC rms.
'Crosstalk', well I too am surprised at your using the term, but I guess there is no other way of describing it.
So would there be transients observed at L1 an L2. (I am only referring to the transients developed internally in one of the ckts affecting the system, external EMI, ESD is another issue altogether). Transients at the node or the common conductor is another issue but would it be observed at L1 and L2 and other nodes?

;) Well he is just a colleague, not an opponent, and since I need to convince him and not the other way around, I guess u know that he is my senior as well as my good friend. Wldn't wanna show him a thread dedicated to target him. Convincing may be the last thing on his mind after he see this thread....;)
I would like to know what exact terminologies and laws would help me convince him. I can relate immediately to the ckt if it was completely a DC network. Since the system is AC I am not in an exact position to dissect transients and noise behaviour, not my forte. Need some help here.

In the experiment both the frequencies were similar, so the continuous phasor difference did not apply there. Hence the experiment was inadequate

... you are running AC and a wire is an inductor at AC. So you do have impedance along the wire and will have AC voltages of the two signal on it. Therefore the signal start to cross couple between the two. That is where most of the problems of grounding in the systems now a days.
Well the system being implemented is similar to case A.
And you are right. Well my knowledge about AC systems is limited and hence my earlier statement. Is there any way or document that could help me understand the transients on the conductor. The problem is that all my correlations are similar to DC and hence I do not understand the nuances in AC :(

In the experiment mentioned above, the voltage waveforms of Vi and Va (should have used other notations) were same, regardless of them being viewed as separate ckts(the common node was removed) or with one common conductor. They also maintained their phasor relations and voltage levels. Since the transformers in the experiment were not loaded I guess there was no loading effect. Also as mentioned earlier in this experiment, the frequency of both the phase voltages was the same.

Last edited:

Well the system being implemented is similar to case A.
And you are right. Well my knowledge about AC systems is limited and hence my earlier statement. Is there any way or document that could help me understand the transients on the conductor. The problem is that all my correlations are similar to DC and hence I do not understand the nuances in AC :(

In the experiment mentioned above, the voltage waveforms of Vi and Va (should have used other notations) were same, regardless of them being viewed as separate ckts(the common node was removed) or with one common conductor. They also maintained their phasor relations and voltage levels. Since the transformers in the experiment were not loaded I guess there was no loading effect. Also as mentioned earlier in this experiment, the frequency of both the phase voltages was the same.

Here is the inductance calculator of a straight wire:

http://www.consultrsr.com/resources/eis/induct5.htm

Impedance of inductance is $Z=j2\pi f L$ where L is the inductance and f is frequency. With this you can calculate the voltage developed on the wire if you know the load current. V=I X Z.

Last edited by a moderator:

Here is the inductance calculator of a straight wire:

http://www.consultrsr.com/resources/eis/induct5.htm

Impedance of inductance is $Z=j2\pi f L$ where L is the inductance and f is frequency. With this you can calculate the voltage developed on the wire if you know the load current. V=I X Z.
But in case A, how is there a current flow possible between the two ckts., What is the voltage (Potential Difference) on the conductor connecting the the nodes of the two ckt.
I assume the potential difference is very close to 0.
Am I missing something here??
Your link is very helpful, I will at least formulate the inductance of the wire.... So that if I am able to measure any instananeous voltage between the two nodes across the conductor, then the current transients can be calculated form that. Thank you very much for that.
But at present I am not sure if any current would flow in that node.

Last edited by a moderator:

But in case A, how is there a current flow possible between the two ckts., What is the voltage (Potential Difference) on the conductor connecting the the nodes of the two ckt.
I assume the potential difference is very close to 0.
Am I missing something here??
Your link is very helpful, I will at least formulate the inductance of the wire.... So that if I am able to measure any instananeous voltage between the two nodes across the conductor, then the current transients can be calculated form that. Thank you very much for that.
But at present I am not sure if any current would flow in that node.

In case A, there is NO current flowing in the blue wire. That is the whole point of separating the circuit. Both case A and B have the same schematic like yours, it is all about the implementation of the wiring. There can not be any current if there is only one point connection between two circuits if it is done right.

Schematically, you are right about your argument, there is no if and buts about it. There is no correlation between the two circuits. Problem only arise when people start hooking up circuit use the same wire for both circuits like in case B.

Load L1 is driven from the supply and have no connection with inverter. So Load L1 can not be driven from inverter. The reason is that the inverter neutral cable is connected to ground.
Load L2 is driven from the battery and inverter and it can not be driven from supply as supply and return point can not be same.
This shows that both the loads are independently driven from their supplies.
For further questions you can freely ask.
Thanks

I dont have questions per say, only how to help someone else understand without embarrassing them, hope you understand.

And well yungman, I agree with you.. and well I guess as the implementation is only of case A type, there are no exact worries.

But I have a question guys, about a common wire condition in DC. I guess there is no problem with the implementation, but only with my understanding. I shall post the image soon for discussion.

I dont have questions per say, only how to help someone else understand without embarrassing them, hope you understand.

And well yungman, I agree with you.. and well I guess as the implementation is only of case A type, there are no exact worries.

But I have a question guys, about a common wire condition in DC. I guess there is no problem with the implementation, but only with my understanding. I shall post the image soon for discussion.

Do you mean you don't think there is problem of Case B if it is DC? That is not true. No conductor wire is absolute zero ohm. There is always some resistance and voltage will develope on the common wire in red. Also, in real life, you never see only DC current, it is always AC and Case B will rig havoc everytime.

AND also even in DC, the red wire will carry both load and can be over loaded. It is not an "Oh well" issue, it is actually very important regarding to cross talk, safety.

Last edited:

No no... there is a separate condition.. I shall post the image soon, it will be more clear if we discuss on that once I post the pic, as I have specific questions on that.

I dont have questions per say, only how to help someone else understand without embarrassing them, hope you understand.

You can invite him/her to join in here. We don't know him and we can say it as it is......and you still smell like roses!!!! :rofl: :rofl: :rofl:

Hi yungman, this is with reference to the DC ckt I was asking about.
Well actually, by the time I finished drawing the ckt, my doubts were cleared... actually no doubts, just laziness that led to a few assumptions.
I love DC, one can eventually resolve the ckt without too much fuss...

Anyways.... the image is as attached.. Had a few doubts about I [DB] and DB itself. But once I finished drawing the ckt, the doubts vanished.. ;)

And even here the ckts are independent, just the I[DB] and I[WZ] are dependent on the attached ckts....

#### Attachments

• Experiment 2.jpg
54.5 KB · Views: 354