Would an isolation transformer eliminate a large voltage on the neutral? 2

[jedijunk]

I work in Cambodia and have a question about isolation transformers and what one would do in my situation. Some people may also weigh in on local grounding, which past experience tells me could get entertaining. ;-)

The Setting
Here in Cambodia, we have an ungrounded power distribution system. A 210v hot wire and a neutral wire are provided to the building, and almost all buildings only use 2-wire runs internally, with no local ground or ground bonded to neutral.

So I decided to rewire my house with 3-wire and install a local grounding rod/spike so I could have a ground wire (for safety and because I use sound equipment that needs a ground). It seemed like a great idea in theory, but...

The Problem
When I measure from hot to neutral I get somewhere between 210 volts and 195 volts as I always have (depending on the time of day and the neighborhood load). That's not ideal but I can live with it. BUT now when I measure from the neutral to my new local ground I get anywhere from 10 volts up to 55 volts (yikes!). Measuring from the hot wire to the local ground shows a corresponding voltage reduction (so if neutral to ground is 40 volts, hot to ground is around 170 volts).

So basically the supply neutral wire is nowhere near neutral, and its voltage is in a different phase/direction than the hot wire. It's almost like the hot wire is a strong phase and the "neutral" wire is a weak phase, and the voltage distance between the two gives me my target voltage of around 210 volts. The fact that the distance between the two adds up to 210v makes me suspect that the power company knows about this and is controlling it somehow, but complaining about the supply voltage will get me nowhere and there's not much chance I could even get any info out of them about this!

The Question
So here's my question. Let's say I could find and install a 1:1 isolation transformer big enough to power my whole house (I don't think it matters to the question but I draw 16-20amps at 210v). Given an input voltage of 170v hot and 40v (in the opposite direction) on the neutral, would the isolation transformer's output be: a) 210v hot leg and a 0v neutral (this is what I'm hoping for), b) 170v hot (or whatever voltage the supply hot to local ground is) and 0v neutral, or c) would the voltage on neutral carry through the isolation transformer, giving me the same problem of 170v hot and 40v neutral? Things you purchase here in Cambodia are non-returnable, so I want to have an informed hope of success before I try to find and purchase a big isolation transformer.

Thanks in advance for any insight or advice you can offer!

 

[pwrtran]

Do you mean ungrounded system has a neutral wire brought out to customer location?
Ungrounded system is mostly for industrial 3-phase load customers not a very good practice for single phase load customers. A Delta-Wye transformer should be installed at the demarcation to create a new system ground, or you can call it an isolation transformer. Install a single phase isolation transformer between one phase and the drifted neutral will not work, the secondary voltage mimic the change on the primary.

 

[itsmoked]

It would allow you provide power to your place that has a neutral and a safety ground that work as you desire but as pwrtran points out, whatever jinking about the utility power does during the day will be reflected on your side of the isolation. That is, if the voltage drops from 210 to, say, 170 you will have the same drop.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Your thinking is basically OK. But, if Your primary goal is to be able to ground your equipment, that should be possible regardless of how the incoming power is arranged. The grounding applies only to the chassis and the cable shields. Not to the internal power circuitry. It is only the most primitive equipment (like old radios and TV sets) where there is no isolation between incoming power and internal circuitry and I doubt very much that your sound equipment is that primitive.

But, if you really want to have it your way, you are definitely on the right track. A 1:1 isolation transformer will do the job. Then, if you want to keep voltage constant during the day, you can install a resonant transformer, aka magnetic stabilizer for the part of your load that you want to use on a stable voltage.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

As Skoggs says, you may install a ground electrode and ground all of your equipment without grounding your neutral.
You may install an isolation transformer and ground one side of the secondary for Zero volts to ground on one leg and full voltage to ground on the other leg.
How expensive is power in Cambodia? Are there any subsidies for low consumption? The transformer will have losses in the single digit percentage referred to full load current. (At less than full load the losses will be less but the percentage may be higher.)
If the present voltage swings are acceptable then the output voltage of a properly sized transformer will be acceptable.
There is an option that may be attractive. Install a sub panel for your sound equipment. Install an isolation transformer sized to supply your sound equipment. Ground the secondary of your isolation transformer.
Leave the circuits for the rest of the house floating.
You will have grounded circuits for your sound equipment at the least cost for both initial purchase and monthly losses.

you can install a resonant transformer, aka magnetic stabilizer

This will work, but.
Be aware that these run quite hot. The initial cost is higher and the operating cost is higher.
For safety, Ground Fault Circuit Interrupters will work on an ungrounded supply.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jedijunk]

Scogsgura, my goal was to get the voltage differential between neutral and ground as close to zero as possible. As you suggest, I originally thought the voltage differential might be ok for the reasons you state, but then (and this is going to open a whole other can of worms)... the reason I'm suddenly looking for a solution to this is that last week I had a friend decide to unplug a laptop power brick from a live power strip that also had an expensive audio amplifier connected to it. There was a large spark from the laptop power supply's socket, and simultaneously a circuit board inside the amp fried. The charring on the circuit board suggested an electrical arc between a grounded metal heatsink on the board to the windings of a transformer near the heatsink on the board.

My best guess was that the large voltage differential between the neutral and the ground was to blame, possibly because he pulled the plug out at an angle, disconnecting one prong before the other and somehow sending a current (likely from the 170v "hot" side but possibly the 50v "neutral" side or maybe even the capacitor in the laptop power brick) into the ground pin, where it travelled into the amp and arced across. The obvious immediate lesson is not to unplug things from a hot powerstrip until I figure out exactly what happened. But my suspicion towards the voltage differential was strong enough to investigate what it would take to fix it.

So, aside from any insight you can provide on that expensive little fiasco , my questions/replies are:
1) If I installed an isolation transformer, and then locally-grounded the isolated neutral for zero voltage differential between neutral and ground as waross said, then I would be left with whatever voltage I currently measure from hot to local ground, right? That wouldn't do because I'd have as little as 170volts sometimes.
2) Measuring from the utility-provided hot to the utility-provided neutral, voltage generally stays between 205v and 215v. It's only when you start measuring those two to my local ground that you start seeing the big fluctuations in the hot leg. So the voltage in the neutral must be fluctuating in the opposite direction at the same time to keep the relative voltage near 210v. I haven't figured that out yet, and might be able to!
3) power is a significant expense here. I hadn't considered the efficiency loss of running the whole house through an isolation transformer. That's a factor I'll have to consider...

 

[pwrtran]

Isolation transformer will not fix your problem. The voltage difference between the hot and the neutral matters only, while the difference between the hot or neutral to a third reference point is relative. If your V(L-N) is applied to the isolation transformer varies, the secondary voltage follows. Call your local utility or hire an electrician to check your connection. Where there is a voltage difference between N-G, there is high impedance in between and there is a voltage drop. If the utility neutral is not drifting, then it could be a poor bonding at the service entry or poor connection between your neutral to the service entry neutral. If you really want to fix it your self you can spend money on AC-to-DC-to-AC, basically rectifier-DC battery-inverter.

 

[waross]

The voltage out of a transformer will be with in a few percent of the voltage in. Forget the existing voltage to ground. You will be connecting the transformer from line to line. That is; From the Hot to the Neutral.
Re the power strip: An ungrounded power strip with a surge suppressor will develop a voltage on the ground pins of about 50% voltage. I remember using laptops and other equipment on a power strip fed from an ungrounded inverter supply. The equipment which had grounded cases would deliver a shock of about 55 Volts if you touched the case while in bare feet,or while wearing damp footwear. One laptop had an internal power supply and was "too hot to handle" unless your footwear was dry.
Solution: Ground the ground pin of the supply cord of the power bar(s).
Possible cause of burnt board: The power brick may have sent back an "inductive kick" when it was disconnected. The heavy arc does not prove this but does suggest it as a possibility. This would be a high voltage transient, transferred by the arc to the supply line. It may not matter too much which line, hot or neutral, was disconnected first.
Grounding the incoming ground terminal of the power bar will allow the surge suppressor to work properly and hopefully prevent future burnouts.

Why is there a voltage on the neutral on an ungrounded circuit? Each wire has a high resistance to ground and a high capacitance to ground. The result is a voltage divider circuit. Ungrounded circuits that I have encountered in the past have generally had fairly equal impedances from both lines to ground and the voltage to ground from both lines was about 50% of the applied voltage.
Often, on physically small circuits, the impedance is quite high and the impedance of the VoltMeter will affect the reading. This was more of an issue with analogue meters which had much lower resistance than today's digital meters.
For some reason your supply has a lower impedance from neutral to ground that from line to ground. It is what it is. Don't worry about it.
Good luck.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Compositepro]

I am confused by the terminology being used in this discussion. Is not a neutral wire, by definition, a grounded conductor that is solidly bonded to ground at one point (at the service entrance) and therefore there should be zero volts to ground? In an ungrounded power system both wire would be "hot" and the voltage to ground would be floating.

 

[Skogsgurra]

Thanks, Bill. I didn't have the energy to explain. You did it very well.

CP: There are so many grid configurations out there. Your assumption that "a neutral wire, by definition, a grounded conductor that is solidly bonded to ground at one point" isn't always true. That is more true for a Ground wire. Read about TN, TN-C, TN-S, IT and TT networks. There are also different transformer connections like Wye, Delta, Open Delta, Delta with center tap and probably a lot more that I haven't seen. Yet.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Neutral implies "Equal in Both Directions". The neutral may also provide a path for zero sequence current.
With a 120:240 Volt transformer, the center tap will be the neutral, 120 Volts to either line.
With a three phase wye connected transformer bank, the wye point is generally the neutral.
With a four wire delta, the center tap of the 120:240 Volt winding is taken as the neutral even though it is not the electrical neutral. (exception #1)

The grounded circuit conductor is the grounded circuit conductor. It is most often the neutral, and many codes specify that the neutral must be the grounded circuit conductor, if there is a neutral.
In a circuit without a neutral, such as a delta or open delta, one line conductor may be the grounded circuit conductor.

An interesting exception where the neutral is NOT the grounded circuit conductor.
By way of example, in a 120:208 Volt wye circuit the center point is the neutral and is the grounded circuit conductor. When a two wire circuit is derived from a either a 120:240 Volt center tapped transformer or from a 120:208 Volt supply, the grounded circuit conductor is still referred to as the neutral, even though a two wire circuit does not actually have a neutral.
In great Britain a common connection is a three phase wye with the wye point being both the neutral and the grounded circuit conductor, as would be expected. A single phase 220 or 230 Volt single phase circuit will still have one line grounded and it will be referred to as the neutral.
I have seen generator sets exported that were grounded in accordance with this standard, even though the sets provided 120:240 Volts. (Exception #2)
Line 1 was the grounded circuit conductor. The true neutral, that is the center tap was at 120 Volts above ground potential.
The voltage from ground to the other conductors was 120 Volts and 240 Volts.
The voltage from the neutral to the two other conductors was 120 Volts and 120 Volts.

Now I should explain that this applies to utilizations voltages and systems.
High voltage transmission and distribution folk often use the terms ground and neutral interchangeably. Distribution folk will "Ground" a wye point to provide a connection to the supply neutral to provide a path for zero sequence currents. Are they wrong. No they have agreed on a slightly different terminology is all. The difference can be confusing to the uninitiated, but that does not imply that it is wrong.
Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

The output of a 1:1 ratio, 1-coil transformer would be 210V between the terminals. By default, the transformer doesn't have a hot or neutral, so measuring the voltage on each terminal with respect to ground is rather pointless.

You create the neutral by grounding one of the secondary transformer terminals. You would then measure 0V between ground and that terminal. You would then also measure 210V between ground and the other terminal.

The reason you are measuring voltage between your local ground and the "neutral" of the power system could be due to local earth potential differences. The ground beneath your house isn't at the same voltage as the ground where the neutral coming to your house is grounded.

 

[MatthewDB]

Cambodia has "TT" grounding. The neutral is grounded at the LV transformer. The building is grounded to a ground rod. The two grounds are not wired together.

For the detailed reference see Schnieder Electric; System earthings in LV

For the quick explanation see Wkikipedia TT (Terra-Terra) earthing system

TT is fine so long as a RCD (residual current) breaker is installed. The problem is that there are often un-corrected persistent ground faults in many developing countries with TT grounding because they don't bother with the RCD. Without the RCD, the power is not cut on a ground fault. What will happen is the ground at the transformer will rise due to ground resistance. In the building with the ground fault, the neutral will approach line to line voltage and the hot will nearly reach ground voltage. In the remaining buildings, the voltage from neutral to groud will rise. No one will fix the ground fault until the ground step potential or leakage currents get too bad.

Putting a transformer in and grounding one end of the secondary will make a local TN-S system and solve the issue.