Find TX turns ratio and 'class' questions?

[tim9000]

Hi, Just made a small thought experiment on the train, haven't really thought about it in depth. I was just wondering, if I created the above scenario (open circuit secondary and a voltmeter), would I get a voltage (assuming there is a difference in turns for primary and secondary)? Neglecting like capacitive or inductive coupling.
I have a feeling not, because both sides are isolated. (?)

And if so, could this be used to tell me something about the turns ratio or it's error?
Presumably if there was a voltage, both sides of the TX would be in-phase. But, please remind me, if the sides were slightly out of phase, that would be from the resistance in the primary and secondary coils?
So say the phase of the secondary was a bit out from that of the input, what would that mean to the 'class' of the transformer?
Also, would you then need like a bridge to feed in the primary and secondary to see the phase error of the TX?

Or what if the turns ratio is like N decimal point something? (not exactly just N) what does that mean for the TX 'class'?

Thanks

 

[Merlin3189]

Neglecting like capacitive or inductive coupling.

I think you are right, that you should get no reading in rhe absence of capacitative coupling. (If it's supposed to be a transformer, I think there must be some inductive coupling.)
And in reality I think you would get a reading, because there is likely to be some capacitative connection. (Left hand diagram)
I would simply make the connections in the right hand diagram. If you don't, any reading you get will be affected by the unknown capacitances as well as the turns ratio.
Depending on which way round the windings are, you should get the sum or difference of the voltages, which are proportional to the turns.
You could simply connect your voltmeter across each winding in turn, then get the actual voltages and be able to calculate the turns ratio directly.

Edit: Sorry I missed the last bits. IMO there should never be any phase error if the same flux links both coils. This is the aim. Since in reality the flux linkage will not be perfect, there certainly can be a magnitude error. I struggle to see how flux is changing direction (magnitude and sense are changing) in a transformer like this, but it must be possible, since it happens in a polyphase transformer. If parts of the core are more lossy than others, I suppose you could get a shaded pole effect.
I don't know how you could see this phase error without looking at the voltages on an oscilloscope or using some sort of phase sensitive meter.
I don't know what bridge you have in mind and I don't know a standard method for doing this, but it should be possible to compare two signals with varying attenuation and varying phase shift network to calculate the phase difference.

 

[Baluncore]

Hi, Just made a small thought experiment on the train, haven't really thought about it in depth.

Why are you doing this, what is the aim of the experiment ?

 

[tim9000]

Sorry it's taken me so long to get back here. Thanks for the replies.

Why are you doing this, what is the aim of the experiment ?

I'm trying to understand the meaning of 'class' of a TX specifically, among other things. Please See below.

I think you are right, that you should get no reading in rhe absence of capacitative coupling. (If it's supposed to be a transformer, I think there must be some inductive coupling.)
And in reality I think you would get a reading, because there is likely to be some capacitative connection. (Left hand diagram)
I would simply make the connections in the right hand diagram. If you don't, any reading you get will be affected by the unknown capacitances as well as the turns ratio.
Depending on which way round the windings are, you should get the sum or difference of the voltages, which are proportional to the turns.
You could simply connect your voltmeter across each winding in turn, then get the actual voltages and be able to calculate the turns ratio directly.
View attachment 101304

Edit: Sorry I missed the last bits. IMO there should never be any phase error if the same flux links both coils. This is the aim. Since in reality the flux linkage will not be perfect, there certainly can be a magnitude error. I struggle to see how flux is changing direction (magnitude and sense are changing) in a transformer like this, but it must be possible, since it happens in a polyphase transformer. If parts of the core are more lossy than others, I suppose you could get a shaded pole effect.
I don't know how you could see this phase error without looking at the voltages on an oscilloscope or using some sort of phase sensitive meter.
I don't know what bridge you have in mind and I don't know a standard method for doing this, but it should be possible to compare two signals with varying attenuation and varying phase shift network to calculate the phase difference.

Okay, so no voltage across them except from inductive coupling. Could you please remind me, what's a 'shaded pole effect'?

Presumably if there was a voltage, both sides of the TX would be in-phase. But, please remind me, if the sides were slightly out of phase, that would be from the resistance in the primary and secondary coils?
So say the phase of the secondary was a bit out from that of the input, what would that mean to the 'class' of the transformer?
Also, would you then need like a bridge to feed in the primary and secondary to see the phase error of the TX?

So say if one coil was loosely wound, and so had more leakage flux, what magnitude will that effect, what magnitude error are you talking about? Like power factor? Or the input and output voltages being out of phase?

I would assume that having the input and output voltages being in phase would be pretty much in phase would be an important parameter of a TX?
Is that what you mean by Phase Error? I was thinking of like a Wheatstone Bridge, is that what you'd have to use?
I finally got a chance to go through my notes (which aren't very detailed) and I saw (exactly like your LHS drawing, Merlin) a scribbled drawing of something like this:

Now I'm not sure if it was an isolation TX or one with a turns ratio, but I think I was trying to find the ratio error of the turns ratio. Do you think this would tell you the ratio error of an isolation TX? (Or if you had a step-up VT with a turns ratio of 5 and you put 100V in, the voltmeter should read 400V? But in that case it would be easier to just measure input and output and divide them to calculate the ratio error? *as you said Merlin*)

Thanks

I guess a post script to this would be, so core and copper losses, what impact to they have on the 'class' or important parameters that you would look for in either a VT, isolation, or CT?

 

[Merlin3189]

I know nothing about classes of transformer. I looked on Google and found Electrical Engineering Portal, which describes 12 classes - all bar 1 and 2 are completely outside my experience! Some classes go up to GW powers (or rather GVA.) They also mention class 2 is broken down into 3 other classes I, II and III , which appear to do with safety.(Class I earthed, maybe with earthed screen between windings, II double insulated, and III safe extra-low voltage transformers.)
Since the other classes, 3 up, are 'real power' transformers, I imagine losses are much more important there. But I don't know anything at all about them.

Shaded pole is used in a type of single phase AC motor. Rather than have a separate starting winding powered via a capacitor, to provide a rotating field, there is a shorted turn of copper round half of each pole. The induced current in this coil changes the phase of this part of the field and the result is that the field oscillates from side to side rather than staying straight between the poles. This is like a rotating field which helps the motor start.
I was thinking that maybe eddy currents or something like that (is there anything else 'like' that?!) could cause a similar change in phase in one winding compared to another in a similar way. But I was grasping at straws. I think in normal transformers they try to ensure the flux is the same in both primary and secondary windings.

If you were testing a 1:1 isolation TX as you suggest, I'd expect you to get either double the voltage or zero depending which way round you connected them. So perhaps this would be a sensitive test of matching, when they were corrected in the null sense.

 

[tim9000]

Great reply Merlin, thanks. So there's much more to the 'class system' than I expected, and the important defining aspects would depend on what sort of transformer.

If I could touch back on to the Bridge for a second. Is it possible to not have the input and output voltages being out of phase? Because it's the same flux, traveling at the speed of light, so I'd have thought not...But what did you mean by "phase error"? (I'm writing this on the train without internet...)
The other reason I ask is that I remember using a Wheatstone Bridge, to measure something important (in centi-radians, what's that, like 0.00somethin rad?) on like a transformer that was used for Bucking. But I can't remember what it would have been to testing (that's why I thought it must have been like a phase difference between input and output), did you have any ideas? ...actually could it have been something to do with the resistance of the coils being different, making a difference in voltage magnitude due to one coil impedance being more resistive than the other coil? (Could that be the Phase Error you mentioned?...Though I'm still not sure how that would be an important measurement)

I did machines, and I thought shaded pole did sound familiar, but I don't remember anything about oscillating fields to start a single phase IM. Did you have a link to anything resembling your explanation?

Side question, something I really should know, if I excited the primary AND secondary, so that the fluxes were fighting each other (like if in the picture I showed in post #4 if the voltmeter was a short), they would oppose such that to completely de-saturate the core. So the permeability would be quite high and they would both be acting as an inductor, to not draw that much current from the source?

Thanks again!