# error in applyExp when the time step is

**URL:** <https://itensor.discourse.group/t/error-in-applyexp-when-the-time-step-is/526>\
**Category:** ITensor Julia Questions\
**Tags:** julia\
**Created:** [November 22, 2022, 4:18pm UTC](https://itensor.discourse.group/t/error-in-applyexp-when-the-time-step-is/526 "2022-11-22T16:18:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![asing](https://avatars.discourse-cdn.com/v4/letter/a/c68b51/32.png) [@asing](https://itensor.discourse.group/u/asing)\
**Post date:** [November 22, 2022, 4:18pm UTC](https://itensor.discourse.group/t/error-in-applyexp-when-the-time-step-is/526/1 "2022-11-22T16:18:08Z")

</div>

Hi!

I am currently using TDVP algorithm to perform calculations of real time dynamics of certain chain system.  
I require really long time scales of the order of 10^4, to do that I wanted to run the calculations with larger time step. I was able to complete the calculations with dt=0.05 without any error.  
Since I wanted to see the dynamics at really large time scales, I tried to increase the time step to say 1 or 2 with non uniform time steps.  
But I am getting the error that applyexp is not converged in 30 steps. Is it possible to remove this error just by increasing maxiter. If so, can you tell me how to increase maxiter with non-uniform step tdvp.

---

<div class="post-metadata">

**Author:** ![miles](https://yyz2.discourse-cdn.com/free1/user_avatar/itensor.discourse.group/miles/32/6_2.png) [@miles](https://itensor.discourse.group/u/miles)\
**Post date:** [November 22, 2022, 4:53pm UTC](https://itensor.discourse.group/t/error-in-applyexp-when-the-time-step-is/526/2 "2022-11-22T16:53:50Z")

</div>

The two keyword options you can pass are `solver_krylovdim=30` and `solver_maxiter=100` where those numbers (30,100) are the defaults. So please try passing a `solver_krylovdim` value greater than 30 and let us know if that helps.