# LTspice: obtaining weird readings with my circuit

#### frustratedgrape20

Joined Mar 3, 2020
23
Hi, I am building a pulse forming network (image of equivalent circuit as attached).

The repeating units of capacitors and inductors form an artificial transmission line. Each capacitor is initially charged up to 360 V, with a capacitance of 0.003 F and ESR of 20 mΩ. Each inductor has an inductance of 0.416991 µH with an ESR of 1.34 mΩ. I would like to simulate this from t=0 (closing of the switch) to t=0.0032 s.

The characteristic impedance of the line is 11.79 mΩ.

The load (represented by the series connection of 'R' and 'L', has a resistance of 4.147178 mΩ and parasitic inductance of 2.97 µH. Due to an impedance mismatch between the load and the artificial transmission line, I am going to use a coaxial, air core pulse transformer to minimize the mismatch.

The objective is to simulate the current through inductor (Ls) as a function of time.

However, the simulation has returned strange results and the results are as described:
1. the current flowing through the primary winding Lp is 51 kA and is much higher than an initial calculated value of 16 kA
2. despite a coupling coefficient of 0.9 between the transformer windings, the current through the secondary (Ls) is less than 1 mA.

May I know if there is anything that can be done to fix this problem?

I have also attached the circuit schematic in this post.

#### Attachments

• 8.4 KB Views: 5

#### Alec_t

Joined Sep 17, 2013
13,793
The coils are all shorted out by the top rail.
For future reference, please avoid using long strings of zeroes in parameter values. Not only does it make reading the values difficult, it is also prone to error. Spice is quite happy with abbreviations u, m, k, Meg etc.
Also, it is unrealistic having parameter values quoted to more than about three significant figures. Usually two is enough, considerng typical component tolerances.

Last edited:

#### frustratedgrape20

Joined Mar 3, 2020
23
Thanks for the assist. I have fixed the circuit diagram and I am able to obtain far more reasonable values for the current through the primary winding.

#### atferrari

Joined Jan 6, 2004
4,701
Thanks for the assist. I have fixed the circuit diagram and I am able to obtain far more reasonable values for the current through the primary winding.
Could you post the result? Sharing is part of the game.

#### Bordodynov

Joined May 20, 2015
3,091
See

#### Attachments

• 2 KB Views: 8
• 55.9 KB Views: 22

#### Alec_t

Joined Sep 17, 2013
13,793
Interesting. That use of arrays is new to me.

#### Bordodynov

Joined May 20, 2015
3,091
See

#### frustratedgrape20

Joined Mar 3, 2020
23
I have amended my circuit with component parasitics as shown below:

A diode is placed near the right end of the circuit to ensure that negative current does not flow into the polarized capacitors & thyristor (not shown).

The coupled inductors (L19, Ls) represent the primary & secondary windings of a coaxial cable type pulse transformer of the type as shown:

This transformer is supposed to match the impedance of the 'E' type pulse forming network (ladder LC network) to that of the load, represented by a series connection of 'R' and 'L'.

Due to the nature of the pulse transformer's construction, a very high coupling coefficient greater than 0.975 is possible. This circuit was designed with the following criteria in mind:
♤ discharge time should be about 1.6 ms, driving an average current of about 12.2 kA through the primary winding L19
♤ pulse transformer's turn ratio of 7:2 should amplify the primary current by about 3.5× → 42.7 kA

Instead, when the numbers were crunched, the following results were obtained:
◊ primary current (through L19) of 600 A for a duration of <0.05 ms
◊ secondary current of 13 A

On the contrary, I am supposed to get a pulse shape like this on the secondary winding:

I have attached a copy of the problematic schematic for everyone's perusal.

Once again, thanks for looking into it.

#### Attachments

• 8.1 KB Views: 1
Last edited: