Anonymised case study
Client names, project references, and proprietary technical data have been removed. The methodology, timescales, and client response are real. The full case study PDF is available on request — contact us to discuss your project first.
A $1.25M concept package. One engineering question. Two hours to answer.
A European EPC was reviewing a $1.25M gas-to-power trial package proposed by a process equipment supplier to one of their energy major clients. The concept centred on conditioning natural gas from a wellhead through pressure letdown stages before feeding a power generation unit.
The EPC's engineer had a specific technical concern: "What is the JT effect on pressure drop and resulting temperature decrease? Can you produce a table profile? Or do we need simulation software?"
What looked like a one-line question was actually a pre-FEED independent review request. The concept proposal contained a two-stage pressure letdown from three possible source conditions, two gas compositions, and a downstream trim-heater whose sizing rationale was entirely absent. Answering correctly required real thermodynamic computation — not a handbook estimate.
Real computation. 12 scenarios. Branded PDF in under 90 minutes.
Scenario matrix
We computed 12 JT scenarios across a full parameter matrix:
- 2 gas compositions — sweet natural gas + rich associated gas
- 3 source pressures — 14, 27, and 41 barg
- 2 cooler outlet temperatures — 25°C and 40°C ambient cases
Tools used
- CoolProp with Peng-Robinson EOS — μ_JT computed at each letdown stage
- Python / ReportLab — automated tabular output + phase envelope plots
- Matplotlib — P-T phase envelope with two-stage letdown trajectories
- KU Automation branded PDF template — assumptions appendix + comment register
| Source P (barg) | Stage 1 ΔT (°C) | Stage 2 ΔT (°C) | Total ΔT (°C) | μ_JT avg (°C/bar) | Heater Required? |
|---|---|---|---|---|---|
| 14 | −3.9 | −3.1 | −7.0 | 0.46 | Likely no |
| 27 | −7.5 | −4.0 | −11.5 | 0.49 | Assess vs dew point |
| 41 | −12.4 | −6.1 | −18.5 | 0.52 | Yes — dew point crossed |
Values anonymised. Actual figures computed per CoolProp/PR EOS at specific compositions provided in the brief. ±10% scoping accuracy.
Key technical finding
The richer associated gas composition cooled more than sweet gas at higher source pressures — counterintuitive at first glance, but consistent with CoolProp/PR EOS. The latent-heat moderation effect (which reduces JT cooling for rich gas) only kicks in very close to the phase boundary; at our operating conditions, the heavier C3+ components simply increased μ_JT. This distinction mattered for the EH-203 trim-heater sizing question: the GATTI proposal listed the heater as "optional if required," which we identified as an inadequately justified position for the 41 barg case.
A 4-page branded PDF. Not a Slack message.
The deliverable was a 4-page branded PDF — not an email, not a back-of-envelope note. This signalled to the client from first touch: this is how we work.
Full scenario table
12-scenario JT profile matrix with per-stage ΔT, μ_JT, and trim-heater assessment for each case
Phase envelope plot
Matplotlib P-T phase envelope for both compositions with two-stage letdown trajectories overlaid
Full assumptions appendix
Every input stated explicitly — compositions, source pressures, EOS choice, accuracy basis, data gaps flagged
6 critical concept comments
Professionally framed comments on the supplier's concept — areas benefiting from closer scrutiny before acceptance
Three things that made the difference.
Real computation, not hand-waving
CoolProp with Peng-Robinson EOS. μ_JT values computed at the actual mixture composition and operating conditions. Numbers defensible to ±10% — appropriate for a scoping study. No fabricated decimals, no "approximately 0.5°C/bar."
Branded PDF, not a chat message
A professional document dropped into the conversation instead of an informal reply. Signalled "this is how we do work" from first touch. The PDF became a reference document the client could forward internally.
Turned a technical answer into a scoping prompt
Finished the response by anchoring three specific asks — actual separator discharge pressure, gauge vs. absolute basis, PCV intermediate setpoint. Gave the client a reason to come back to us with data, rather than a dead-end "thanks." Positions us for DWSim-grade detailed design if the package is awarded.
"We turn vague client requirements into actionable spec.
We rerun the numbers when yours change.
We don't fake certainty."
— Kingsley Uzowulu, CEng MIMechE · KU Automation
The scoping study was the opening move.
The JT scoping study positions KU Automation for detailed design engagement if the package is awarded. The next step is a DWSim simulation once three data points are confirmed from the client's side:
- Actual separator discharge pressure — normal, minimum, and maximum operating range
- Gauge vs. absolute basis confirmed for all pressure data in the original proposal
- PCV intermediate setpoint firmed up — determines whether EH-203 trim-heater is required or optional
Once those are confirmed, the JT profile resolves to a single operating point and the EH-203 sizing decision becomes deterministic.
Full case study PDF available on request.
We share the detailed PDF with engineering teams who are evaluating a similar scope. Tell us your project context — we'll tell you if it's a useful reference.