General question about transformers.

[THCynical]

I have a few general questions about transformers from a couple of basic experiments I have conducted.
I can find very specific and detailed information online, but not a general rough explanation of what going on.
 I’ll try to keep things as clear as possible.

I have recently been mucking about with transformers and I’m just wondering;
 Why do larger Microwave oven transformers, when powered from a 12V 100W supply, have a higher output drop off than smaller transformers when loaded?

For example, MOT 1 produced 227V from 12V, but when I tried to power a small 18W 120VAC MO fan It barley rotated
output down to just 38V, and with MOT 2  107V output down to 21V and the fan didn't move.

But when I use a midsized centre taped step down (wired in reverse) A 190V output drops only to 90V and the fan rotates with vigour, drawing 3A from the 12V supply.
I use this transformer and fan when soldering to blow away the fumes, the transformer gets quite warm after extended use.
The diagram also includes the coil resistances.

What factors are at play? And what information is needed to estimate this effect?

But from these very basic experiments I’m also wondering why doesn’t wiring transformers secondary into the primary of another and so on produce stable very high voltage?
CTSD transformer, stepping up into MOT1, doesn’t work, MOT1 has a similar/same output as if it’s connected directly to 12V, upon investigation the first transformer has it output dropped right down to the 12V on the input.
Only when I use the two MOTs one stepping up into the next to I get around 2kV put it drops sharply with any type of loading.

Finally, and this is the one that has me most baffled, when using a small wall plug adapter 12VA AC wired in reverse as a step up 12V input to 207V unloaded. Loading with the fan gives 70V output, this is enough to power the fan but there is almost 3 times the rated current through the transformer, it gets hot and begins to fail.
It’s the nature of the failure I don’t understand. The fans speed after a small amount of time beings to fluctuate eventually slowing right down, But voltage remains constant!  The 12V in and the 70V output slowly drops to 64V, but not necessarily at the same time as the fluctuations. I’m assuming it has something to do with the increase in temperature but I have no idea how this is happening and would not know how to begin to ask google it.   
Both the fan and the power supply seem robust.

If this type of effect has been discussed before point me it that direction, if there are external links that are relevant include them.
My first question is the main one. If you have any opinion any way related feel free to chime in!
Thank you for your time and wisdom.

[johansen]

heat the microwave oven transformers to 100C (you can boil them in water for heating) and hammer out the magnetic shunts.
what frequency are you driving the transformers from 12 volts?
if you want to make a 60 or 50 hz 220volt inverter, i recommend rewinding the microwave oven transformer, but the details of that depend on the load.
microwave oven transformers are designed to deliver 2200vac so the inpedance of the windings is very high.

for a 300 watt 12v to 220vac inverter, 300 turns for the 220vac winding would be more appropriate. the turns count for the 12v source depends on how your inverter works, something like 10-16 turns.

[THCynical]

50hz i think, havent measured it, would it be ok to put a Scope across mains?

I'm going to use one to make a power supply, replacing the secondary with (+18 0 -18)V and (+7 0 -7)V coils. 
rectify and regulate to a +/-15V +/-5V DC

from looking into it, Im guessing removing the shunts is not going to affect the loaded voltage drop from the original question and is just for when i'm putting new coils on?

Thanks for your reply!

[mimmus78]

>> 50hz i think, havent measured it, would it be ok to put a Scope across mains?

If you "like the smell of magic blue smoke in the morning" it can be a good idea.

or better to check this video:

http://www.eevblog.com/2012/05/18/eevblog-279-how-not-to-blow-up-your-oscilloscope/

[johansen]

from looking into it, Im guessing removing the shunts is not going to affect the loaded voltage drop from the original question and is just for when i'm putting new coils on?

unfortunatly, there isn't much information on the internet about how they work.

basically the magnetic shunt keeps the flux flowing through the second half of the transformer just above the knee of the flux density/mmf curve, so that the output current and voltage is approximatly regulated.
the method by which this is achieved is the first half of the transformer is run well above the saturation "knee" and the difference is shunted away by the iron.
you'll notice the shunts are about 1/3rd to 1/5th the area of the transformer core.
typical unloaded flux is 1.7 to 1.9T, but the flux in the second half of the core is only 1.5T or something.
also, the air gap between the magnetic shunt and the core is part of the design.

the reason this is done, is so there isn't a huge difference in power output between 110vac, and 130vac, or 220 to 250 mains for 50hz countries. if there were no shunts, you would be looking at a burned out magnetron and/or transformer  and/or a 50% change in power levels

[THCynical]

My scope is in the post so i probably should read up how to use it before it arrives, cheers!

Thanks J that helps quite a bit.

R.

[THCynical]

Just regarding putting the scope accross mains.

"If you "like the smell of magic blue smoke in the morning" it can be a good idea."

Other than the ground issue and safety procedures around mains voltage, the scope is 300Vrms rated so why can't/shouldn't i use it on mains?
I have a dimmer switch here and i really want to see what it looks like on the scope.   

[IanB]

Other than the ground issue and safety procedures around mains voltage, the scope is 300Vrms rated so why can't/shouldn't i use it on mains?

Because the scope is a delicate and expensive piece of test equipment, and if you make any mistake, or if something happens to go wrong (such as spikes on the mains), you might damage the scope. Unlike a handheld multimeter, which is designed to be connected to the mains and has all kinds of input protections, the scope is not so designed and does not have such input protection. So it's best not to risk it.

(The problem with the mains is not only the voltage, but also that it is a "high energy" supply. It can supply very large fault currents, easily enough to completely vaporize delicate components.)

[johansen]

generally it is the policy of most forums that any discussion regarding connecting test equipment to mains is off limits.

basically, there's three problems, severity is in reverse order.

1:neutral isn't actually neutral.
2:metal parts on the scope might be live.
3:vaporizing the oscope ground lead...

most American homes have a neutral that is close to ground but this isn't good enough for low noise measurements.
as for myself i connect neutral to ground all the time and can work with the scope grounded to either end of the 120mains.

I have no idea if most 220 50hz lands have the mid point, or one side of mains grounded.

you should be able to safely play around with mains, provided you make sure NOT to clip the ground wire of the O'scope lead to the hot side of the mains.
this is the biggest problem.

i have one of these newfangled plastic scopes, i made an adapter that lifts the ground and lets me know if something is live.

[IanB]

I have no idea if most 220 50hz lands have the mid point, or one side of mains grounded.

Where the UK is concerned there is a single supply at 240 V 50 Hz. One side is neutral and grounded, the other is live. There is no mid-point.

[ejeffrey]

generally it is the policy of most forums that any discussion regarding connecting test equipment to mains is off limits.

Only when you make bad and dangerous suggestions like you do here.  I don't mind advising someone who knows what they are doing and has the proper equipment about mains measurement.  But forums are filled with people who did something one way one time and didn't die and think they know what they are doing.  Those people are like little typhoid marys running around giving advice on electrical safety.

as for myself i connect neutral to ground all the time and can work with the scope grounded to either end of the 120mains.

Please don't do that and don't advocate that people do it.  This is illegal and dangerous, and it will set off any RCD if the outlet is protected by this.

If you want to connect a scope to mains, do it: in order of preference:

1) Get a scope designed to be used floating, like the tek TPS2000 series.  The channels are individually isolated, and designed for this sort of application.  They are often battery powered so they can be completely floating.

2) Use a high voltage probe with a regular scope.  Either connect the ground clip to ground (not neutral) or don't use it.  Ideally, use a high voltage differential probe.

3) Disconnect the ground clip and use an ordinary 10x scope probe.  Ignore the fact that the neutral is not equal to ground, and that you have a high inductance signal path.  You won't have good frequency response, but what the hell: this is mains wiring.  Some probes are rated CATII, which is actually rated for probing mains outlets, but you are unlikely to get hurt by any 300 V rated probe measuring a 120 VAC line.

4) Use two ordinary 10x probes, and configure the scope in A-B mode.  You get a psuedo-differential probe, which will have acceptable CMRR at low frequency.  Again, it will be bad at high frequency, but you aren't measuring high frequency.

i have one of these newfangled plastic scopes, i made an adapter that lifts the ground and lets me know if something is live.

Just because the scope body is plastic doesn't mean that it is safe to lift the ground.  The shells of the BNCs are all connected to the 'chassis' ground, as are the shrouds on any IO ports (serial, GPIB, USB, VGA).  Furthermore, there may be non-obvious safety hazards: the buttons, knobs, case screws, and screen are not designed to be safety isolated  They don't necessarily satisfy any safety clearances, .  If you clip the ground lead to live, your scope becomes a deadly weapon.  The indicator is at least a nice touch, as long as your response to that light going on is to unplug everything at the wall.

[johansen]

since you addressed my oscope ground lift via neon lightbulb separately, please clarify which of the other suggestions are bad and dangerous, if i read my post correctly, i didn't make any suggestions, other than just go ahead and float the scope.

if you can't handle a hot Oscope, you have no business playing with mains equipment anyway, God forbid you shock yourself with an MOT, which is what the OP was trying to use to make a 220 vac inverter.

if you don't want to connect neutral and ground, simply float all of your equipment on the neutral, and you won't have noise from other loads showing up on your ground referanced test equipment. (which should be less than 100mV anyway)

[johansen]

I have no idea if most 220 50hz lands have the mid point, or one side of mains grounded.

Where the UK is concerned there is a single supply at 240 V 50 Hz. One side is neutral and grounded, the other is live. There is no mid-point.

good to know. that simplifies things.

[Kremmen]

I have no idea if most 220 50hz lands have the mid point, or one side of mains grounded.

Where the UK is concerned there is a single supply at 240 V 50 Hz. One side is neutral and grounded, the other is live. There is no mid-point.

good to know. that simplifies things.

Generally (but not always) the distribution network is oriented to deliver ungrounded 3 phase power end-to-end. Latest the final distribution transformer (Dyn11; 20 kV to 400V) has a neutral wire from the secondary star point. The premises entry to the main panel is 400VAC50Hz 3 phase+N. The N is grounded at premises entry thus creating the PE grounding point in the distribution panel. The single phase wiring inside premises is one live phase against neutral (producing 230VAC phase against neutral), with separately wired PE prongs for class I outlets. The load is (more or less) equally divided between the 3 live phases. All outlets in one room should be connected to the same phase. For high power appliances (typical is the electric sauna stove in Finland  ) a 3 phase connection is used, utilizing all of the phases in various star and delta configurations. These are most often semi-permanent i.e. not pluggable connections, although a large variety of 3 phase sockets/plugs is of course available for professional use.

[Jeroen3]

would it be ok to put a Scope across mains?

Not if you have to ask.

[THCynical]

Thanks for the replys, I was interested in viewing the waveform from a Lighting Dimmer and the mains when the kettle is turned on or off and stuff like that. I'm happy enough to keep off mains and just look on google.

I did know that neutral was tied to ground at the consumer unit, and when watching the EEVblog above on scopes it had occurred to me that i had better get it right, which in itself is enough to put me off.

 "Disconnect the ground clip and use an ordinary 10x scope probe."

I had noticed that the scope still reads a signal when the ground clip is removed, in college they just told us not to do that, no reason given, so i thought it might be bad for it or something.
but its more that the measurement is no longer referenced to anything?

 

[ganzuul]

There isn't just one wiring scheme for residential mains; there are several. They vary from country to country, year to year, and within countries as well. Your neighbor can very well have a different class of wiring than you do.
Usually when someone doesn't give you the reason for why you shouldn't do something it's because they just don't know. There is however a special class of troublesome people who know but don't actually tell you.

It is wise to question such a toxic atmosphere where potentially life-saving information like what you are asking for is withheld.

The numbers and units on the spec sheet of your scope should tell you if it is possible to hook it to something high-current, high-voltage like the mains without frying it. Properly educating yourself about the environment of the electrician is however the only way to mess with this stuff and not fry yourself. Scopes can be replaced.

To safely view what the power supply does under load however you should use a 'clamp probe' to view the waveform of the current. These probes are a bit expensive but they do provide galvanic isolation from the dangerous stuff. You don't even need to peel any wires to use them.

Eg. with one clamp probe on the live wire and one on the neutral the two waveforms should look identical. If they don't look identical then you have a big problem as some of the current is escaping from the system, possibly through a short to ground. There are automatic breakers which constantly check that the current in equals the current out. This dynamic should let you understand a great deal about the otherwise invisible wave mechanics at work here.

would it be ok to put a Scope across mains?

Not if you have to ask.

Ah, so it's better not to ask and just do it.

Or perhaps the issue is more complicated than that?

[ejeffrey]

I did know that neutral was tied to ground at the consumer unit, and when watching the EEVblog above on scopes it had occurred to me that i had better get it right, which in itself is enough to put me off.

What do you mean by consumer unit?  Ground is connected to neutral at the point where the electricity enters the customer premises at the main fuse/breaker panel.  In some jurisdictions it may be connected other specific places as well, but it should never be connected at the load.

[/quote]
 "Disconnect the ground clip and use an ordinary 10x scope probe."

I had noticed that the scope still reads a signal when the ground clip is removed, in college they just told us not to do that, no reason given, so i thought it might be bad for it or something.
but its more that the measurement is no longer referenced to anything?
[/quote]

The problem is that the current flowing into the scope probe has to return to the device under test somehow.  If you don't have a ground clip, that loop goes through the ground conductor in your electrical wiring.  This loop has high inductance which will mess up high frequency signals quite a bit.  For the low frequency signals you are likely to be looking at it works OK, and avoids the potential to clip your ground lead somewhere other than ground, which could lead to blowing out the ground lead and/or creating an unsafe condition.  It also can pick up more noise, but when you are measuring line voltage, a little noise is rarely a problem.

[THCynical]

"What do you mean by consumer unit?"

In Ireland and UK we mean the fuse board. yeah, probably worth clarifying that.

[IanB]

What do you mean by consumer unit?  Ground is connected to neutral at the point where the electricity enters the customer premises at the main fuse/breaker panel.

Yes, the consumer unit is the main fuse/breaker panel. It's where responsibility for the electrical wiring passes from the supply company to the consumer.

[David Hess]

I did know that neutral was tied to ground at the consumer unit, and when watching the EEVblog above on scopes it had occurred to me that i had better get it right, which in itself is enough to put me off.

What do you mean by consumer unit?  Ground is connected to neutral at the point where the electricity enters the customer premises at the main fuse/breaker panel.  In some jurisdictions it may be connected other specific places as well, but it should never be connected at the load.

This is where it gets tricky and why I always use an isolation transformer for safety whether using a differential probe or not.

Earth ground is not connected to the primary side neutral at the load but it is usually connected to the secondary side.  When making a primary side measurement using the oscilloscope, connecting the oscilloscope probe ground which is connected to earth ground through the oscilloscope to the neutral of whatever is being tested will create a ground loop.  My solution is to use the isolation transformer on whatever is being tested and connect the earth ground to the same earth ground that the oscilloscope is using at the same place that the probe ground is connected.