---
title: "Thalmann E-L"
description: "What the exponential-linear model assumes about off-gassing, what selecting it does to your schedule, and why it is not a model to plan a real dive with."
---

> Documentation Index
> Fetch the complete documentation index at: https://docs.divelogic.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Thalmann E-L

The Thalmann exponential-linear model is a dissolved-gas model that lets gas
into your tissues the way Bühlmann does and refuses to let it out the same way.
Once a compartment holds more inert gas than the depth you are at will support,
its off-gassing is held to a fixed rate instead of accelerating with the size of
the gradient. That asymmetry is the whole of the difference between the two
models, and it is what "E-L" names.

## What the assumption costs you

Bühlmann credits a large pressure difference with driving faster washout.
Thalmann assumes it does not: beyond a point, a tissue clears at whatever rate
it can and no faster.

A profile Bühlmann clears can therefore take longer under E-L, and the extra
time does not fall evenly. E-L tends to produce brief stops deep and noticeably
longer ones at 9 m, 6 m and 3 m. If you are used to reading a Bühlmann schedule,
the shape is the first thing you will notice, before the runtime is.

## Gradient factors do nothing under Thalmann

Selecting **Thalmann E-L** makes the GF Low and GF High fields inert. The model
has no conservatism control of any kind: the ceiling it hands the planner is its
own limit, taken raw. The planner passes your pair in and the Thalmann path
discards it.

Conservatism under Thalmann means selecting a different parameter set. The
**Parameter set** field appears next to the algorithm whenever Thalmann, either
hybrid model, or GF Calibration is selected, and it is the only lever the model
answers to. Which set is selected changes the schedule more than any other
control on the page.

## The parameter sets are not verified

The numbers behind every set in this build were transcribed from second-hand
sources and have never been checked against the original research they cite.
Treat every set as an unvalidated reconstruction, and do not plan a dive you
intend to make on one. The default set has no sourced helium behaviour at all,
so running it on trimix is a guess wearing a citation. Most of the others were
fitted to constant-ppO₂ rebreather diving, so selecting one on an open-circuit
plan asks it to schedule diving it was never fitted to.

What the model is good for is comparison: running the same dive under Bühlmann
and under Thalmann, and reading the gap between the two runtimes, shows how much
of your schedule is an assumption rather than a measurement. If you want that
disagreement to inform a real plan rather than replace it, use
[Hybrid OC](/theory/hybrid), which holds you at whichever model is deeper at
every stop.

Like Bühlmann, E-L has no bubbles, no anatomy and no diver. The fixed
off-gassing rate is an assumption about washout chosen because it fitted trial
outcomes, not an observation of what tissues do.

## Next

- [Hybrid models](/theory/hybrid) run this model alongside Bühlmann.
- [GF calibration](/theory/gf-solver) expresses a Thalmann schedule as a
  gradient-factor pair.

Source: https://docs.divelogic.ai/theory/thalmann/index.mdx