Wellbore Genius

Pad-scale verification suite

Eleven published reference stages solved through the same whole-pad kernel as /pad-parallel, each graded against the measured field result with a declared tolerance band. Every check states the band that produced its verdict, and pairings that measure a different physical quantity are graded as a ratio window instead of accuracy.

Public benchmark Compare your own well

Field-calibrated physics

Verifying with the constants fitted to HFTS-1 and MSEEL.

Fitted defaults

Tune the constants Side-by-side numbers

Constants from a measured core

No saved core yet, so this page is solving with the constants fitted to the published HFTS-1 cores. Import and save your own core report and it will take their place here.

Import a core

Constants from a measured log and pressure test

No saved well log or DFIT yet, so this page is solving with the constants fitted to the published field data. Import a log and a shut-in pressure test and their permeability, closure stress and proppant reach take their place here.

Import a log or pressure test

Well path and rate history

No saved deviation survey or production history yet, so this page uses the published pad's own depth and spacing. Import a survey and its true vertical depth, inclination and azimuth set the closure stress and stage spacing solved here, while rate history is compared against the modelled propped reach.

Import a survey or rate history

Fluid model (black-oil PVT)

The pad-scale kernel solves with a single-phase fluid. These curves show where that assumption stops holding for this well's fluid, so a disagreement can be read rather than guessed.

Oil gravity (°API)

Gas gravity (air = 1)

Producing GOR (scf/STB)

Water cut (%)

Reservoir temp (°F)

Use my well's own data

Nothing imported yet measures a fluid property, so these curves come from the typed values above. Import a PVT lab report — plus a well log, a DFIT and a production history — and the oil and gas gravity, gas-oil ratio, bubble point, temperature and pressure will all come from your well instead.

Import my well data

Black-oil PVT curves at 210 °F

p_b 3,201 psia

Solution gas R_s (scf/STB) & B_o (rb/STB)

Gas z-factor & oil viscosity (cP)

Mixture density (ppg)

In-situ free gas (% of volume)

Standing R_s / B_o with Vasquez-Beggs undersaturated corrections, Dranchuk-Abou-Kassem z-factor and Lee-Gonzalez-Eakin gas viscosity. Below p_b the mixture density is a no-slip average: the true two-phase gradient needs a holdup model, so read it as an upper bound.

Compositional cross-check

Flash pressure

1,920 psia

Phase state

two-phase

Vapour (mol)

8.5 %

Vapour (mass)

2.8 %

Apparent GOR

101 scf/STB

Condensate yield

9,892.1 bbl/MMscf

Liquid gravity

77.3 °API

Vapour gravity

0.873

The two fluid models differ by -79%. Treat the black-oil gas figures as indicative and confirm with a measured PVT report.

Component	z	x (liquid)	y (vapour)	K
N2	0.0040	0.0022	0.0229	10.181
CO2	0.0120	0.0112	0.0207	1.850
C1	0.3400	0.2559	1.2476	4.875
C2	0.0750	0.0746	0.0792	1.061
C3	0.0550	0.0583	0.0194	0.333
iC4	0.0110	0.0119	0.0018	0.149
nC4	0.0260	0.0281	0.0032	0.115
iC5	0.0120	0.0130	0.0007	0.053
nC5	0.0140	0.0152	0.0007	0.044
C6	0.0210	0.0229	0.0004	0.018
C7+	0.4300	0.4698	0.0004	0.001

Wilson K-values with a Rachford-Rice solve — a screening split, not a converged fugacity balance. Source: Typical Wolfcamp B separator recombination.

Want the fluid on its own? Open my well's fluid page for the curves, bubble point, phase table and free-gas flags at a glance, or the fluid audit to see your PVT lab report set against the correlations pressure step by pressure step.

Also solve the instant path and show the difference

Every stage is solved twice — the fast path behind the live sliders and the field-calibrated path — and each result lists what the fast answer would have told you instead.

Height containment window: 0.35–1×

Published HFTS-1 / MSEEL band. Narrow it to your own fibre or microseismic data.

Published window

Adjust the window

Where the stage actually goes, cluster by cluster

Even enough

Every cluster shares one pressure in the casing; what differs is what it has to beat first — friction on the way to the toe, and the squeeze from its neighbours. Clusters nearer the heel take more, and the shielded middle takes least.

Clusters

Stage rate (bpm)

Shots per cluster

Hole size (in)

Cluster spacing (ft)

Pipe friction (psi/ft)

Neighbour squeeze (psi)

Spread across clusters

3.3 %

Perforation friction

1,272 psi

Heel vs toe

0.6 pts

Under-taking clusters

Cluster	From the heel (ft)	Rate (bpm)	Share
1	0	14.0	17.5 %
2	30	13.4	16.8 %
3	60	12.8	16.0 %
4	90	12.8	16.0 %
5	120	13.3	16.6 %
6	150	13.7	17.2 %

What the split does to the fracture. Each cluster is solved on the share it actually took, instead of an equal slice of the stage.

Import a fibre run on the field data page and the measured share of each cluster appears beside the modelled one.

A warn is a disagreement, not a defect

Accuracy bands come from the measurement uncertainty in the source papers, so a warn means the model sits outside the reported read accuracy — worth explaining, not necessarily wrong. Invariant failures (non-finite geometry, a mass balance that does not close, a non-converged pad) are solver defects and should block a run.