Joule Thief 101

[tinman]

Poynt has already explained this to you.  Do you really need for it to be explained to you a second time?  Playing the game of "missing" the first time something is stated and forcing it to be reposted again is just plain silly.

yes. There needs to be a series resistor added in the circuit,just like they do when designing circuits.
The sim will not solve the unsolvable.
Like you,the sim dose not understand ideal.


Brad

[MileHigh]

yes. There needs to be a series resistor added in the circuit,just like they do when designing circuits.
The sim will not solve the unsolvable.
Like you,the sim dose not understand ideal.

Brad

The sim could solve for the ideal coil easily.  But the software designers have limits in how high the floating-point numbers can go.  If there is a resistance then the final current through the coil is finite as time goes to infinity.  If there is no resistance then the current through the coil goes to infinity as the time goes to infinity.

The software designers of the sim don't want the sim breaking all the time with out-of-bounds errors for the computational variables.  They know it would make the program seem buggy and unreliable.  So to prevent things like that from happening they force reasonable constraints on what you can enter as a schematic to prevent the software variables from going out of bounds.

If you temporarily remove these intentional restrictions the sim would happily calculate the response of an ideal inductor from t=0 to t=fifty years.

It's you that does not understand "ideal" right now, and like usual you have erected a wall because you are so sure of yourself.

[MileHigh]

Sure indeed whatever,,

The sim chokes because it is trying to solve for an unsolvable calculation.

You are totally our of your element and some Dunning-Kruger is manifesting in you.  See my previous posting.

[poynt99]

Brad,

You may want to refer to Partzman's last sim. He used a very tiny value for his R, and it still gave him the correct answer of 2.4A.

[partzman]

yes. There needs to be a series resistor added in the circuit,just like they do when designing circuits.
The sim will not solve the unsolvable.
Like you,the sim dose not understand ideal.


Brad

Tinman,

Have you used or are you familiar with LtSpice? Are you aware there are many various parameters that can be set prior to running a simulation not including the circuit parameters? I can simulate the 5h inductor with a dcr of 1 fohm (1e-15) without crashing so obviously Poynt and I are using different operating parameters. For example, in this particular simulation I am using modified trap for the integration method with the other options being trapezoidal or gear.

Does this possibly change your opinion about simulating this ideal inductor?

Poynt,

I would like to know what parameters you are using to replicate your crash condition.

partzman