Power Transformer Model For LTspice

Thread Starter

newbie2019

Joined Apr 5, 2019
95
Hi:

I am trying to create a power transformer model for LTspice.
Primary = 120 vac
Secondary = 12 vac

Then I decided to add a 1/2 wave rectifier circuit.

I can't get a dc reading on the vdc output.

Does anyone have any suggestions?

Thanks
 

Attachments

eetech00

Joined Jun 8, 2013
3,859
Hi

In LTspice, a transformer is basically two coupled inductors. The value of each inductor is the “square of the turns ratio”. For your transformer, the turns ratio is 10:1, so the primary inductor is 100 times larger (10^2) than the secondary inductor. The voltage source should also be set to ~169v (120 VRMS) and its serial resistance to 100 milliohms. Also set a reasonable amount of series resistance in the inductors (use the ratio to estimate). CTL-RHT-CLK the label in the waveform viewer to view the RMS value.

Unless you really want to model a real transformer, in which case, it can get really complicated.:(

eT
 
Last edited:

ericgibbs

Joined Jan 29, 2010
18,766
hi 2019,
Checking your asc file, you have 120Vac Voltage source driving a 100uH inductor/transformer secondary of 1uH.

Run a .tran analysis and you should see a secondary voltage.

The inductances are much too low a value. [Use the values shown in post #3]

Note also the Vsource is Vpeak , not Vrms, so Vs should be 170Vpk

E

EDIT: This transformer PDF is a good guide.
 

Attachments

Last edited:

Thread Starter

newbie2019

Joined Apr 5, 2019
95
This info has been great. Two more questions
1) How would I go about setting up the attached power transformer
as a component?
2) The DC voltage does not show up on the dc analysis. Why is that?

Thanks
 
Last edited:

Bordodynov

Joined May 20, 2015
3,177
1. You can draw a transformer with a separate diagram. Add the required parameters. Read about hierarchical schemes in Help.
2. With DC analysis, there is no time. Actually Time = 0. And accordingly there is no sinusoidal voltage. In addition, the transformer does not transmit a constant voltage.
 

Thread Starter

newbie2019

Joined Apr 5, 2019
95
1. You can draw a transformer with a separate diagram. Add the required parameters. Read about hierarchical schemes in Help.
2. With DC analysis, there is no time. Actually Time = 0. And accordingly there is no sinusoidal voltage. In addition, the transformer does not transmit a constant voltage.

Thanks Bordodynov!
 

crutschow

Joined Mar 14, 2008
34,285
The transformer primary inductance must be high enough to keep its magnetizing current below the core saturation point.
Typically this is a small fraction of an amp for small power transformers.
 

ebeowulf17

Joined Aug 12, 2014
3,307
Sorry for resurrecting such an old thread, but I was trying to apply your transformer demo circuit here and got unexpected results.

I assume the 50mA figure in the calculations refers to the maximum primary current you need to handle. I was trying to run the numbers for a transformer that could handle more current, specifically 400mA, and I was surprised to see that this formula delivers a lower inductance value for higher current values.

Am I misunderstanding what the 50mA in this formula represents? Or is this right, and a transformer that delivers more power can get by with a smaller inductance? Intuition tells me this is wrong - among other things, a transformer that's capable of handling more power would still also be expected to work at lower power levels too... so surely I must be misunderstanding the function of the 50mA value in this formula. Could you help me better understand how to apply this to different values than those in the original post?
 

kubeek

Joined Sep 20, 2005
5,794
Yep, the 50mA represents the magnetizing current when the secondary is open circuit, so the primary inductance is chosen to achieve that 50mA at 120V 60Hz.
 

ebeowulf17

Joined Aug 12, 2014
3,307
Yep, the 50mA represents the magnetizing current when the secondary is open circuit, so the primary inductance is chosen to achieve that 50mA at 120V 60Hz.
Ah, ok. So the 50mA doesn't directly related to how much current the transformer can actually handle, but rather the minimum amount of primary current that's required in order to transfer power effectively to the secondary?

Presumably the magnetizing current relates to core properties, including its size, which in turn depends on the maximum power you want to be able to transfer? If so, how do you make an educated guess as to what the magnetizing current would be for a transformer of any given power rating? I know transformers involve some fairly complicated math, but are there ballpark estimates that can get you close?

If I want to simulate a transformer with a 12VA rating, how do I guess what the magnetizing current will be? If I later want to sim one with a 250VA rating, how do I guess what the magnetizing current for that one will be?
 

kubeek

Joined Sep 20, 2005
5,794
The magnetizing current is more of a result of the number of turns in the primary and the shape and material of the core, in power supply transformers it generally is not a thing that you aim for.
 

Ylli

Joined Nov 13, 2015
1,086
No body else has, so I'm going to ask....
ericgibbs' post #4 shows the transformer primary to have an inductance of 100 uH.
Bordodynov's post at #3 uses a formula of Lmin = (120V/50mA)/(2*pi*60) which works out to be 6.3662 H.

What am I missing?
 

ebeowulf17

Joined Aug 12, 2014
3,307
The magnetizing current is more of a result of the number of turns in the primary and the shape and material of the core, in power supply transformers it generally is not a thing that you aim for.
I'm not trying to design a transformer - I'm just trying to get a very rough idea of how to make a reasonable transformer simulation in LTspice when I don't know the actual characteristics (inductance in particular) of a real transformer of equivalent size and power.

I often find myself wanting to simulate a power transformer - for my needs it usually doesn't need to be a perfect simulation, but I've learned the hard way that if the inductance values are too low you get weird results that don't match any real-world transformer. I assume that setting the numbers unrealistically high is also a bad idea, but I have no idea how to estimate what they should be. So far, I just always overshoot until I stop seeing weird limiting effects, but I'd love to do a little better than my terrible guesses so far!

So, referencing Bordodynov's formula, if the 50mA represents magnetizing current in the primary, is that a fixed number that I should just leave the same no matter how large or small of a power transformer I'm simulating? Does it move proportionally with the power of the transformer? Does it change just a little?
 

Danko

Joined Nov 22, 2017
1,829
So, referencing Bordodynov's formula, if the 50mA represents magnetizing current in the primary, is that a fixed number that I should just leave the same no matter how large or small of a power transformer I'm simulating? Does it move proportionally with the power of the transformer? Does it change just a little?
Using magnetizing current value as 10% of primary full load current is good choice.

ADDED:
Transformers - The Basics (Part 1...Part 4)
 
Last edited:

kubeek

Joined Sep 20, 2005
5,794
No body else has, so I'm going to ask....
ericgibbs' post #4 shows the transformer primary to have an inductance of 100 uH.
Bordodynov's post at #3 uses a formula of Lmin = (120V/50mA)/(2*pi*60) which works out to be 6.3662 H.

What am I missing?
Bordodynov's is designed to have 50mA going through the primary without any load on secondary. Erics has there a few kiloamperes at 50Hz, so it really is not a realistic transformer for this frequency.
 

Bordodynov

Joined May 20, 2015
3,177
I'll explain where I got the number 50 mA from. I was holding several transformers (not very big ones) in my hands. They had a no-load current of 15mA to 60mA. For the user, the smaller the current, the better. It decreases as the inductance increases, and this requires an increase in the core and the number of coils. Therefore, a compromise is made on the magnitude of this current. There are methods and programs to calculate the transformer. I didn't accidentally point out that this is the minimum inductance. For a very powerful transformer, that current can be much higher. But in this case you have to do a more careful calculation of the transformer. For example, consider the hysteresis loop of the magnetic material, the number of coils and the geometric dimensions of the core. I did such calculations. Examples are on the Russian website:

http://kazus.ru/forums/showthread.php?t=113392&page=7
http://kazus.ru/forums/showthread.php?t=113392&page=46
http://kazus.ru/forums/showthread.php?t=113392&page=69
http://kazus.ru/forums/showthread.php?t=113392&page=70
http://kazus.ru/forums/showthread.php?t=113392&page=71
http://kazus.ru/forums/showthread.php?t=113392&page=72
...
 

MrAl

Joined Jun 17, 2014
11,396
Playing around with coupled inductor transformers can be a little problematic as you are probably finding out.
If you dont need to model things like excitation current and saturation and inductance and are mainly interested in seeing how the rest of the circuit after the transformer works, you can construct a dependent source transformer that actually works with DC also as long as the transformer is unidirectional.

The way it goes is you use a voltage controlled voltage source to get the secondary voltage from the primary voltage. If you are not that interested in anything else that works fine. If you need to model the primary input current then you use a current controlled current source that senses current on the secondary side and injects the primary current on the primary side.
If the ratio of primary to secondary voltage is A, then the ratio of secondary current to primary current is 1/A. Those two are used to set the gains of the two sources.
So to recap, the VCVS senses primary voltage and produces a secondary voltage, and a CCCS senses secondary current and injects it in one of the primary leads (with respect to proper polarity), and the gains are A and 1/A respectively.

Because they are simple dependent sources, this setup also works with DC. So you can do some testing using a static DC simulation. If you are working with AC inputs though you still need to use a sine wave.

Of course you can also add primary inductance if you feel like it :)
 
Top