Isolation Transformers

[chala]

it look to me that according to the drawing at post 27, the transformer is made of two 115V coil in the input and two in the output that you can connect in serie or parrallel, if the input is 240 and protected by CB and the output 110 then if you feed 110V supply into the labelled output you have an output of 240V of the labelled input protected by CB so you dont have to change anything. Now every machine should a have stamp of approuval for the country of use. ask the manufacturer what stamp the machine carry and for which country. To explain VA, PF and W could take a lot of writting

[DeepFrz]

If you have 240 volts the input looks like

L1 ----breaker-----
N ------------------
L2 ----breaker-----

with either 120 or 240 volts output.

If you have 120 volts in you have

L1 ----breaker----
N -----------------
L1 ----breaker----
effectively the L1 and L2 inputs are shorted together.
with either 120 or 240 volt output.

[chala]

possible configuration
230-115
115-230
115-115
240-240

[MidLandOne]

Originally Posted by Extemporaneous

...Maybe only one of the 16amp breakers is used when the transformer is receiving 230v, but through some BLACK MAGIC it uses both 16amp breakers to make a 32amp breaker when receiving 120v????...

...I've never heard of paralleling Breakers, but perhaps it can be done. But I'm not sure and I don't like things I don't understand (sucks to be me). But I try.
Also note, I did try to check with the maker.

If they have paralleled the equal sized 16A toggle circuit breakers for 120v (as it seems they likely have) I do not see that as a problem as long as they are not intended for and not accessible as the means to isolate the 120v/230v downstream distribution.

If they were so accessible an obvious safety hazard would exist as both TCB's would have to be opened for isolation. But as far as I know they are not easily accessible nor intended as being the main breakers their being behind the front panel so there would seem to be no problem - as the manufacturer has said, you have everything you need.

[Extemporaneous]

Originally Posted by chala

possible configuration
230-115
115-230
115-115
240-240

chala,
This is obviously a strange setup, way beyond me. Perhaps you can derive more from this.
http://www.victronenergy.com/upload/...%20DE%20ES.pdf

I hope, you (or anyone else) can shed some light on this, I'd appreciate it.

I guess (with my limited understanding) that the two 16amp breakers in series, it would still trip (one then the other) at 16amps. Line voltage would just hang in there until the strongest of the two breakers tripped. The weakest would have already tripped (make any sense?). With the 115v would the breakers not have to be is parallel (which I or anyone I've talked to has not heard of) to become 32amp? Is it somehow possible?
Would make sense that it is somehow being done like this, that would be why they don't connect directly to the breakers. They have have to do it with the jumpers. I hate being so ignorant.
If anyone can come up with the truth, type the explanation very slowly and use small words so that maybe I'll understand.

Thanks,
Extemp.

[s/v Jedi]

Where do you see the breakers in parallel?? They are not! each breaker protects it's own primary winding. Each winding is rated for 16A maximum. The breakers don't act on voltage and thus not on wattage, just the current.

In 230V input indeed, the breakers are in series. Double secure, no problem either.

There is only 1 phase (hot) and 1 neutral. Not L1 + L2 hot inputs like US 240V. But when you connect it to US 240, you just connect L1 to hot and L2 to neutral inputs and put the primary windings in series. With a transformer, you don't really have a hot and neutral until you connect one of them to ground (it then becomes the neutral). This is why it is so safe, you can swap hot and neutral on inputs!

I would be more worried about it's isolation feature: check if input-ground stops right there and has infinite resistance with output ground.

cheers,
Nick.

[MidLandOne]

Originally Posted by Extemporaneous

If anyone can come up with the truth, type the explanation very slowly and use small words so that maybe I'll understand.

I'll expand out my last post which you seem to have ignored the possibility of its being correct (I sometimes wonder if the effort is worth it ).

As pointed out by others earlier in the thread for the rating of the transformer at 120v input the current required (with wriggle room) is 32A, while at 230v the current required is 16A (with wriggle room). So in one configuration we have to protect for 16A and in the other at 32A.

The Victron unit has two 16A toggle circuit breakers. Two toggle circuit breakers of equal rating wired in parallel will load share between them and so give a combined rating of 32A. I don't know if this is how Victron have designed the unit but it seems the likely way so that when one sets up the jumpers for 120v input the tcb's are connected in parallel in the 120v line in giving 32A protection while when set up for 230v in there is only one giving 16A protection in the 230v line in from the 230v supply (that 230v supply being line (or phase), neutral and earth).

Now if one protects a device by paralleled equally rated tcb's it is important that they not be accessible for nor be intended to be the isolation breaker for the unit and the distribution downstream from it. If they were to be improperly used or be accessible so they could be used as the main isolation then the risk exists that someone could break one of them and not the other and assume that they now have an isolated system - but the other breaker will still be supplying the system so a shock danger exists. As far as I know the Victron tcb's are inside the cabinet so this requirement is met.

It seems to me that two tcb's of equal current rating can happily load share if they are wired in parallel because neither will trip until their combined current rating is reached - the current is shared equally between them (well as near as damn to equally). One may throw slightly before the other would due to manufacturing tolerances but the moment it does the other will be faced with twice its current rating and throw too. If you think about this you will see that for it to work both tcb's, I believe, have to have the same current rating.

Again, I am not saying this is how Victron have done it as I am not privey to their design and I am sure someone will disagree with me, but it would seem to me to be the likely route they have taken and stacks up with Victron's response that you have everything you need for 230v operation.

[chala]

Extemp.
I had a look.
First: the input at all times remains the input. Disregard my drawing, this because there is no mention of 16A CB but in fact there is mention of a more sophisticated circuitry with: overload and short circuit protection, variable fan cooling and over heating protection, and inrush current limiter. If the protections are of good quality you should not have to make any modification to it. If I was you I would pay an electrician to set it up and take all the responsibility.

The bad part.
In regard to the first disclaimer I would not buy.
I understand the second disclaimer.
There is plenty warning and here in OZ you would need to be a licensed electrician to set-up the transformer.
There is only one compliance mentioned and I do not know what that compliance stands for.
You still need an RCD and it makes sense. An isolation transformer which supplies more than one outlet is as dangerous as the shore supply. A true isolation transformer will prevent electrocution if it has only one outlet and you don’t get in contact with both wires.
It looks like it will produce a fair amount of heat

The good part.
It has a toroid transformer, they do not come cheap.
It has an alu box.
The instructions are clear enough.
The specification looks good enough if true.
The configurations are as in my previous post. If it is that 240-240 should read 230-230

[hellosailor]

Chala-
In the US, a licensed electrician is sometimes needed for home wiring, but in many cases an owner can do their own work, and in some areas a contractor does not have to be licensed or used at all. Remember, we are 50 sovereign states and the laws aren't even uniform within the states. That's intentional.
Boatowners? There are no codes forcing civilian boat owners to hire licensed electricians, that's a total non-issue here. Disclaimers? Should we start with how US and Ozzie federal and state codes make some of them meaningless anyway?<G>

I can't help laughing at the acronynm "RCD" and the concept of a residual current device. In American YnGlitch, 'residual current' is what you have left over flowing in something like a superconductor coil, after the primary current has been shut down. Or what we call the "back EMF" surge coming out of a coil after thecircuit powering it has been shut down. Residual here means "left over", not stray or unintended. There's nothing I can see that's residual about an unintended and dangeous stray current flow, maybe I don't understand how an RCD is different from a GFI device.

Jumpers, instead of expensive and bulky and less reliable switches, seem like a perfectly good way to reconfigure a device on installation. MANY AC electrical devices here, including military generator sets, are set for 110-220-440 volts by simply changing the configuration of the coil wiring, although that's usually done directly because it is a one-time operation and jumpers would just be "more stuff" to get in the way.

[s/v Jedi]

Hmmm... my post must have gone by unnoticed or incomprehensible. I'll try a bit more elaborate as I detect much confusion in the thread (and yes, this used to be my profession):

With transformers, one normally doesn't specify input/output voltages; one uses ratios, like 1:2 plus a VA rating, which is the maximum continuous amount of power that the transformer can pass. Sometimes, more than one VA rating is specified, like for continuous and maximum power. The maximum value in that case should specify the time-constraints for that value (like: maximum 2kVA for 30 seconds). There will also be a frequency rating, like 50Hz or 60Hz or a range, like 45-65Hz. The last rating is the maximum voltage allowed. This is sometimes called the insulation-rating of the coils (windings).

A normal isolation transformer is 1:1. The voltage applied on the input is equal to the voltage found on it's output. If you feed it 118.75V you get 118.75V on it's output. In other words, there is no stabilization or, the only way to get 120.0V on the output is to feed it that exact same voltage.

A ratio of 1:1 also means that the primary coil (input) has the same number of turns as the secondary coil (output). Different ratio's are the result of matching the number of turns to the desired ratio. This also means that if you short a couple of turns on the output, you lower the output voltage. If you do that on the input, you increase the voltage output. Some isolation transformers have a tap on the primary coil. This enables one to boost the voltage by connecting one of the input wires to the tap instead of to the closest end of the coil. This is good when a 110V outlet on the dock only supplies 80 or 90V.

Another kind of transformer is a step-up or step-down transformer. This is mostly a 1:2 or 2:1 transformer when used for converting mains power for different parts in the world. A 1:2 transformer makes 240V when you feed it 120V. When you connect it the other way around it becomes a 2:1 transformer, it makes 115V when you feed it 230V or 60V when fed 120V.

The galvanic isolation part is natural behavior of a transformer with insulated input and output coils. The power fed to it is converted into an electro-magnetic field. The output-coil is positioned in that field and converts it back to AC electricity. There is no electrical connection between input and output.

Now we get to the tricky part: the smartest isolation transformers for boats have two input coils and two output coils, all with the same amount of turns. It's just that, 4 coil total and has nothing to do with how many "hot wires" to connect to it, 2-phase or 3-phase power or any of that. It's just coils. But one can interconnect these coils in three different ways: series, parallel or not interconnected at all. (That last option is more advanced than needed here but could be used to connect to two shorepower outlets to add their capacity together). One can connect both primary and secondary coils in the same way (like both in series) and the transformer will act like a standard isolation transformer. But one can also connect the input coils in series and the output coils in parallel. This makes a 2:1 transformer -> step down. It allows you to adapt to whatever voltage is offered on the dock.

Now we come to the breakers. The transformer should be protected against overload. Fuses (any form) are used for that. Their only function is to protect the transformer, NOT to switch it off or protect wiring in your boat. You need one (1) fuse for each input coil. The fuse protects that coil. So, when you have 2 input coils, you should have two fuses. The rating of the fuse is the same as the maximum amperage rating of that coil. Just like for the breakers in your switchpanel: they should be matched to the wire they protect.

Now the Victron transformer. This one has 2 input and 2 output coils. (I am Dutch and we refer to windings; I think it's called windings in English too instead of coils). The maximum current that can run through each coil is 16A so each has a 16A breaker connected in series to the coil. When we connect the two input-coil in parallel, we actually create a network of series and parallel mixed, just like in many 12V house battery banks when using 6V or 2V batteries. Each breaker is still in series with the coil. The breakers ARE NOT parallel. Only one side of the breakers is connected to the other breaker; the other side is only connected to the coil it protects. The internal resistance of the breakers is much lower than that of the coil so the input-current splits over the breaker equally as the resistance of the coils determine the ratio (and they are precise). Minor differences in internal resistance of the breakers does NOT lead to differences in load. When the breakers would be in parallel, that minor difference would be a factor but because each breaker is in series with a coil, the sum of resistance of breaker+coil for each path and the ratio between the 2 paths formed, results in the amount of current through each path. OMG I hope this makes sense for non-electricians ;-)

cheers,
Nick.

[MidLandOne]

Originally Posted by s/v Jedi

...Now the Victron transformer. This one has 2 input and 2 output coils. (I am Dutch and we refer to windings; I think it's called windings in English too instead of coils). The maximum current that can run through each coil is 16A so each has a 16A breaker connected in series to the coil. When we connect the two input-coil in parallel, we actually create a network of series and parallel mixed, just like in many 12V house battery banks when using 6V or 2V batteries. Each breaker is still in series with the coil. The breakers ARE NOT parallel...

I think you will find that if you give it a bit of consideration, when 2 windings are in parallel each with a breaker in series with the winding in the manner that you describe, that the breakers are indeed in parallel when looking into the windings from the source.

[s/v Jedi]

To be complete: the comparison with a galvanic isolator (I'll call it a GI): said to be as good or better than an isolation transformer (called IT from here on). That is dangerous talk IMO !!!

In my previous post I wrote that the inputs of an IT have no electric connection to the outputs. That is not so for a GI !!!!! With a GI, the neutral input is connected to the neutral output and the hot input is connected to the hot output. In fact, the device is nothing more than a jumper for these two wires. But is does break the ground wire. It does this (simplest form) with 2 diodes connected in anti-parallel. Only electronics geeks will understand that so here it comes:

A diode allows current to flow through it one way only. If you connect it the other way around, it breaks the circuit. One side is called the anode and the other side the cathode. We need these terms to distinguish them later on but I won't go into their workings further as that's not needed.

When a diode conducts (anode connected to positive feed), you loose some voltage just like a resistor or resistance in a wire that conducts. But that voltage is always the same, independent of the amount of current (0.6-0.7V) You can use that trick to lower the voltage but it is more often considered not so good like in a diode-bridge for charging two battery banks with one charge source.

When a diode is connected so that it will conduct, it will only start doing so for a minimum voltage. In other words, it needs a minimum voltage before it opens. This is the most important "feature" for use in a GI. A very nice diagram: (just look at the green background-part and the V-on point where current starts to pass:



Everyone knows what a parallel connection is. You can connect two diodes in parallel by connecting their anodes on one side and the cathodes on the other side. Anti-parallel means one is swapped around (reversed), so anode to cathode and cathode to anode. This combination will conduct both ways but will still need that minimum voltage to start conducting. If you want that increase that minimum voltage, you can add multiple of these in series.

AC is different to DC. The polarity reverses all the time, 50 or 60 times a second. But two diodes in anti-parallel will conduct it because one diode conducts for the positive half of each cycle and the other conducts for the negative half.

In a GI this network of diodes is inserted in the ground wire. Now, before we get galvanic corrosion, we need a voltage difference between ground ashore and boat-ground (mostly the seawater we float in) that is HIGHER than the value that makes the diodes conduct.

Now the reasons why this is a bad product:

- If the difference in ground potential is big enough (read really bad conditions) the device doesn't work at all. This is just when you need it most and an aluminium hull would be eaten away quickly. That is the reason no serious owner of an aluminium boat would select a GI; they all have an IT.

- One can not put in more diodes because any rise in voltage needed before opening means less protection against electrocution: the primary reason for having a ground wire! I find it amazing that these devices are allowed at all!

- Reverse polarity: no protection. In fact, there is no isolation at all for hot and neutral wires. This alone should be reason enough for an IT, regardless of galvanic protection.

- If a diode fails (and owners of diode-bridges, alternators etc. know this happens regularly) you loose the protection of the ground wire. You get electrocuted to death if you are unlucky enough to touch a live wire or part without rubber soles under your feet. The new GI's do have an alarm to warn you if a diode fails. You better be able to hear that and check the alarm circuitry daily!

- If you get a power-surge leading to a bigger ground-current than the rating of the diodes, they burn and you are left with the previous point.

- Diodes get hot too, touch the ones on your alternator to confirm that. The problem is that the diodes in a GI do not work during normal operation and thus do not get warm. Only some monitoring circuitry, if present, can tell you if they will work or not. That circuitry can fail too! If they do overheat, in an emergency situation like guiding a full 30A short back to ground ashore, they might burn up (not unlikely with diodes) and your protection is gone. When an IT overheats you loose power and thus get a 100% protection.

With an IT, ground is fully disconnected. The ground from shore is connected to an inner shield inside the IT, protecting you from the primary windings on the transformer. The ships ground is connected to an outer-shield (metal housing) which is isolated from shore ground / inner shield. When the shore ground becomes 120V (happens a lot due to human mistakes) there will still be no galvanic corrosion.
The ground protection inside your boat will be as good as the bonding system (dynaplate is perfect ground).

cheers,
Nick.

[s/v Jedi]

Originally Posted by MidLandOne

I think you will find that if you give it a bit of consideration, when 2 windings are in parallel each with a breaker in series with the winding in the manner that you describe, that the breakers are indeed in parallel when looking into the windings from the source.

I am really sorry but they are not. For two devices to be parallel, both inputs and outputs must be connected. The inputs of the breakers are connected, but not the outputs. I think you look at the windings as just a piece of wire and thus a jumper. But they are not. If they were, you would have a full short when you connect the winding to the 120V shorepower, wouldn't you?

If you have a primary winding with 120V which is running a 16A current, the resistance of the winding is 120/16 is 7.5 Ohms. The resistance of the breaker will be 0.0 Ohms (it's too small for this scale, like 0.01 Ohms or less). This means that the windings can't be seen as a wire connecting the breakers, they are in fact the load of the circuit, just like a lamp, motor etc.

Compare to a DC switchpanel. All the breakers are connected on their input-side. But that doesn't mean they are parallel, they are not. On their outputs are loads, like VHF, lamp, etc. before they are connected to the common negative return. Between the outputs of 2 breakers are 2 loads, just as the 2 windings in the transformer.

The reason that the two breakers in the Victron each handle half the current, is that the two windings are the same. Just like when you connect two identical lamps to two different breakers on your DC panel. If the windings would be different, the currents would be different. But when you would put 2 equal parallel breakers in series with 2 parallel windings, the current through the breakers would stay the same even when you change the windings so that each winding carries a different current.

Lesson : parallel in electric or electronic meanings are exact and not an approximation of direction like one could use for two roads.

ciao!
Nick.

[MidLandOne]

Originally Posted by s/v Jedi

...I am really sorry but they are not. For two devices to be parallel, both inputs and outputs must be connected. The inputs of the breakers are connected, but not the outputs...

I am afraid that I can only put your response down to there being a different Dutch interpretation of "parallel" to the international one .

I am sure that anyone with some electrical knowledge who drew the situation out would using the generally accepted interpretation of parallel immediately see what I mean. The windings are in parallel you say, then in that case any device in series with one winding is also in parallel with any device in series with the other winding.

I'll leave it at that cos I can see this is going to go nowwhere .

[s/v Jedi]

Originally Posted by MidLandOne

I am afraid that I can only put your response down to there being a different Dutch interpretation of "parallel" to the international one .

sorry, it's International. The official term is a "series-parallel combination" circuit but it's commonly named a "series-parallel" circuit. Worldwide, no doubt. Also down-under. ;-)

I am sure that anyone with some electrical knowledge who drew the situation out would using the generally accepted interpretation of parallel immediately see what I mean. The windings are in parallel you say, then in that case any device in series with one winding is also in parallel with any device in series with the other winding.

I understand your confusion. Try to think of it in this way: during normal operation, the breakers are "invisible". They insert no resistance, capacitance nor reactance to the circuit. There could as well be just a connection, a short piece of wire or trace on the circuit board. Now, looking at the windings, we say that they are parallel; we "conveniently" forget the breakers because they don't matter. It is accepted to say the windings are in parallel because the breakers are only there for protection and have no influence on the workings of the circuit.

But talking about the circuit breakers, things change. Now we have to look at the protection part of the circuit which is different from the normal operating mode. If one would state they are parallel, outputs and inputs are directly connected. This is because it matters for the circuit; ie: the moment they activate. When connected in parallel, small differences between the breakers will decide which one flips first (the one with the least resistance). But that is not the case here because the windings dictate the current through each breaker.

I'll leave it at that cos I can see this is going to go nowwhere .

That's easy. But I can assure you I was well educated in this field, electric, electronic and micro-electronic (chip) and computer design and have extensive international experience in the field. I think I am the only cruiser with a full microcontroller development kit aboard.

A very good site explaining this incl. graphics is: Series-Parallel DC Circuits Analysis - Learning Activity
This is a US based site.

cheers,
Nick.