# Problem with simulation

1. Apr 9, 2014

### rf74

Hi,

I have a problem with my simulation (Multisim). I don't find the error. Voltage and current increase very high in a short time. It's only 3 selfs with coefficient at 0.9 for L1/L2 and for L2/L3. K is at 0 for L1/L3. I give image for the circuit and the result of simulation. The problem come from K ?

K1 LL1 LL2 0.9
K2 LL2 LL3 0.9

thanks

#### Attached Files:

File size:
8 KB
Views:
71
• ###### tr23.png
File size:
219.8 KB
Views:
104
Last edited: Apr 9, 2014
2. Apr 11, 2014

### rf74

Hi,

Someone can help me ? I don't find my error. If I'm looking at energy input/output, the sum is no 0. I simulated with R1 and R2 at 1 MOhm but the circuit give high voltage. If K1 and K2 are under 0.5 it's ok, but with K1=K2=0.9 and K3=0, voltages at R1/R2 increase very high. I tested with another software and it's the same result.

K1 LL1 LL2 0.9
K2 LL2 LL3 0.9
K3 LL1 LL3 0.0001 or 0

#### Attached Files:

• ###### tr29.png
File size:
184.9 KB
Views:
63
3. Apr 16, 2014

### rf74

Hi,

I found the problem, it's because I let K3=0, I can't do that, the minimum must be at K1K2.

I find a circuit where I can have a single sign current from a sinusoidal source, and the sum of energy is not zero. But is it possible in a self to have a positive current ?

I give you the circuit, I tested with Multisim and LTSpice, it's the same result.

#### Attached Files:

File size:
51.8 KB
Views:
65
File size:
29.5 KB
Views:
61
• ###### dc6.png
File size:
64.1 KB
Views:
64
4. Apr 17, 2014

### rf74

Hi,

I simplify the circuit, it can be like:

vV3 14 9 dc 0 ac 1 0
+ sin(0 {90*1.414213562} 60 0 0 60)
vV2 0 6 dc 0 ac 1 0
+ sin(0 {90*1.414213562} 60 0 0 60)
rR4 11 4 1
K2 LL2 LL3 0.9
rR2 10 9 1
lL3 13 0 1
lL2 5 0 1

There is a positive current in transformer, the energy from the source is +V*I -V*I in a period, like I is always positive, the energy from source is 0. Like that it's logical, but energy must be conserved so it is a problem from Spice, the current at start is not like that in reality ? I think it could like that at start with an external DC source no ? I tested with Multisim and LTSpice. The difference of energy is 1200 J in a second. This is only in the transitive time. After current is at 0, no energy at output and no energy at input. If K=1 the energy is conserved, it's only when I let K=0.9 or 0.5. Angle phase must be at 60° for example, don't let it at 0°. Someone can test for verify my study ? It is the energy inside 1/2LI² ? At start I=110A so the energy inside selfs is LI²=121kJ.

If I let:

vV3 14 9 dc 0 ac 1 0
+ sin(0 {90*1.414213562} 60 0 0 60)

*$Multisim Component V2$*
vV2 0 6 dc 0 ac 1 0
+ sin(0 {90*1.414213562} 60 0 0 240)

The energy at outputisnot the same.

Thanks for your help

#### Attached Files:

File size:
28.9 KB
Views:
67
• ###### dc8.png
File size:
10.2 KB
Views:
71
Last edited: Apr 17, 2014
5. Apr 20, 2014

### rf74

Hi,

I understood, it's because the energy is inside inductances ! Software give start values for current and there is energy inside inductances. All is fine !

bye

6. Apr 20, 2014

### sophiecentaur

Well done rf74. You constructed a whole, satisfactory thread with question, discussion and answer, all by your self!

In the old days, before the Internet, this was how we all used to solve problems. lol

7. Apr 20, 2014

### adjacent

But it's 11 days of hard work . lol

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?