SPP's HILLGA Process Explained: How to Interconnect Generation for Your Data Center And Why the Withdrawal Study Can Make or Break Your Project

By Keentel Engineering — Power System Studies | EMT & Dynamic Modeling | Interconnection Support

The math of the AI era is brutal: data centers want hundreds of megawatts in 18–36 months, while the standard generator interconnection queue delivers new supply in five to seven years. The Southwest Power Pool's answer is HILLGA — the High Impact Large Load Generator Assessment — a 90-day study pathway, housed in Attachment BB of the SPP Tariff and approved unanimously by FERC in January 2026, that lets dedicated generation interconnect quickly for one purpose: serving a specific large load.

HILLGA is fast, but it is not simple. It is a conditional pathway with a sizing formula, siting geometry rules, an operating leash called LLRIS, a five-year expiration clock, and a study gauntlet that includes the same SCRCCT screening and EMT modeling expectations as the load-side HILL process. And beneath the process sits the single most valuable piece of pre-application engineering a developer can commission — the withdrawal study: an hour-by-hour determination of how much power your load can actually withdraw from the grid at your POI, which drives everything from BESS sizing to generation strategy to whether the application succeeds at all.

This guide walks through the complete HILLGA application process step by step, explains the models SPP expects (including CMLD and PERC1 dynamic load models and PSCAD EMT models), and makes the case for why the withdrawal study deserves the first line of your project budget.

Part 1: What HILLGA Is and the Leash That Comes With It

HILLGA runs in parallel with the HILL load registration process

While the HILL Delivery Point Study (HDPS) evaluates connecting your large load to the grid, HILLGA evaluates connecting the supporting generation — dedicated capacity, sited near the load — that will serve it. The concept: give large loads a bridge to fast, local supply while limiting impact on the wider grid, then transition that generation into SPP's standard planning processes over time.

The defining feature is the service type:

Load Limited Resource Interconnection Service (LLRIS)

A HILLGA generator does not receive open access to inject power like a normal network resource. Its injection is tied to the supported HILL — the generator exists, in the tariff's eyes, to serve that load. This coupling is the source of both HILLGA's speed (the grid impact is bounded) and its most demanding engineering requirements (the generator must provably follow the load, including when the load suddenly disappears).

Three structural rules frame every HILLGA project:

The sizing formula

Under Attachment BB §3.1.1, the Maximum Injection Capability of a HILLGA request may not exceed the associated HILL's requested MW load multiplied by the higher of (a) 110% plus the seasonal Planning Reserve Margin in effect at submission, or (b) 125%. In practice: a 245 MW HILL can support a generating facility of roughly 306 MW or more of injection capability. That headroom exists for good reasons — reserve margin, station service, degradation, maintenance overlap — and using it well is a design optimization, not a formality.

The siting geometry

The generation must be local: the Point of Interconnection must be on the transmission system no more than two substations away from the HILL's Local Delivery Facilities ("HILLGA-Local"), and in the same SPP Deliverability Area as the HILL. One generating facility may support multiple HILLs, provided no more than five substations are involved and no more than two existing transmission line segments are utilized — a genuinely useful lever for multi-site campus strategies.

The five-year clock

The HILLGIA (the interconnection agreement HILLGA produces) has a service term of five years from the Commercial Operation Date. To keep operating beyond that — and to shed the load-limited leash — the customer must submit a full Interconnection Request in a subsequent Consolidated Planning Process (CPP) cycle, seeking Network Resource Interconnection Service (NRIS, with an expedited designation pathway available to SPP's Deliverability Areas) or Energy Resource Interconnection Service (ERIS). Treat HILLGA as a bridge with a hard expiration date, and plan the CPP entry early — not in year four.

One threshold question before anything else: is your generation even in the HILLGA lane? If the generating facility interconnects to facilities not under SPP's functional control — behind the fence, on distribution, or to non-OATT transmission owner facilities — it follows the non-jurisdictional generation procedures of Business Practice 7250 §6.3 instead: transmission owner notification for units of 5 MW or more, potential Affected System Study Agreements with SPP, specification requirements (including whether a BESS will register as a Market Storage Unit), and SPP stability analysis under BP7250 §8. Choosing the wrong lane wastes months. Determining the right one is the first piece of advisory work on any co-located generation project.

How Keentel helps:

Keentel Engineering advises developers at the concept stage on HILLGA eligibility, the jurisdictional lane determination (Attachment BB vs. BP7250), injection sizing under the 110%+PRM/125% formula, siting geometry against the two-substation and Deliverability Area rules, and the CPP exit strategy — before deposits are paid and study clocks start.

Part 2: The Complete HILLGA Application Process, Step by Step

Step 0 — The prerequisite.

Before a HILLGA request can even be submitted, the customer must have planned load and supporting generation together with the Transmission Customer, resulting in an executed HILL-Load Connection Study agreement. In other words: the load-side HILL process (Attachment AQ, AX, or BA) comes first or in parallel. HILLGA is a companion process, not a standalone shortcut. Practically, this means the load's ten-year forecast, one-line, load models (CMLD/PERC1), and the Additional HILL Characteristics Form are already in flight — and the generation application must be consistent with all of it.

Step 1 — Assemble and submit the HILLGA request.

The submittal package includes the executed HILLGA study agreement, an identified Point of Interconnection for each request (compliant with the HILLGA-Local geometry), the required study deposit (scaled to requested MW under the filed fee structure — confirm current amounts in Attachment BB before wiring funds), site control evidence, and the technical data package: generator specifications, dynamic models, and — per SPP's explicit recommendation — a PSCAD EMT model at time of submission (more on why in Part 3).

Step 2 — Acknowledgment and validation.

SPP acknowledges the request within 5 business days, then conducts a 15-calendar-day review of the submission. If deficiencies are found, the customer has 10 business days to cure them. This is where incomplete model packages die quietly: a deficiency notice on your dynamic models doesn't just cost the cure window — it costs credibility and, if uncured, the request.

Step 3 — The 90-day System Impact Study.

Once validated, SPP performs the HILLGA SIS within 90 calendar days, evaluating steady-state performance, transient stability dynamics, short circuit ratio, and SCRCCT screening — the same system-strength gate as the load-side HDPS (SCR/WSCR/CSCR ≥ 6.0; CCT ≥ 0.15 s). In the same 90-day window, the host Transmission Owner performs its Facilities Analysis, and affected-system impacts are screened and studied as needed, with the customer and SPP coordinating with affected-system entities per the tariff.

If SCRCCT screening fails, EMT analysis is required outside the 90-day timeline.

This is the single biggest schedule risk in the process, and it is why SPP recommends the PSCAD model accompany the application: if the screen fails and your EMT model is already on file and validated, the supplemental analysis starts immediately instead of waiting months for model development.

Step 4 — Re-Study, if required.

If system conditions or project changes require it, SPP notifies the customer and performs a Re-Study within 60 calendar days.

Step 5 — Execute the HILLGIA.

After the study report, the customer has 45 business days to negotiate and execute the HILL Generator Interconnection Agreement, with its five-year-from-COD service term.

Step 6 — Plan the exit.

File the full Interconnection Request in a subsequent CPP cycle early enough that NRIS or ERIS is secured before the HILLGIA expires. The expedited designation pathway to SPP's Deliverability Areas exists — but only for those who plan for it.

End to end, a clean run from complete submission to executed HILLGIA is roughly seven to eight months. Every one of the words "complete," "validated," and "clean" is doing heavy lifting in that sentence — and all three depend on the quality of the models in the application.

Part 3: The Models CMLD, PERC1, PSCAD, and Why They Must Agree

A HILL + HILLGA project is studied as a coupled system: a large power-electronic load and its dedicated generation, sharing a small electrical neighborhood, evaluated by parallel 90-day studies. That demands a coherent model set:

CMLD (Composite Load Model)

The WECC composite load model represents the load facility as motor fractions, electronic load, and static load behind feeder/transformer equivalents, with protection-based tripping and reconnection. For a data center, the parameterization must capture the UPS-served electronic fraction, VSD-driven cooling, and the trip/reconnect thresholds that mirror SPP's HILL Fault Ride-Through Requirements.

PERC1 — SPP's preferred load model

PERC1 is the performance-based electronic load model built for power-electronic-dominant facilities. It directly parameterizes the behaviors SPP's requirements regulate: proportional load reduction during voltage sags, constant-current disturbance behavior, deep-sag transfer/trip logic, and the timed recovery ramp back to at least 90% of pre-disturbance consumption within one second of voltage recovery. SPP's HILL submittal guidance lists load modeling data as "CMLD and/or PERC1 (preferred)" — and on a HILLGA project, the load models matter twice, because the generator's studies are run against the load's behavior.

Generator dynamic models

The supporting generation needs its own RMS dynamic models (machine, exciter, governor, and plant controller for synchronous units; the applicable inverter models for BESS and solar) meeting SPP's model quality requirements, plus compliance alignment: IEEE 2800-2022 Clause 7 with SPP's decision points for inverter-based resources, PRC-024-4 for synchronous machines.

PSCAD EMT models

SPP recommends the PSCAD model at time of submission for exactly the reason described above — SCRCCT screening failure triggers EMT analysis outside the 90-day clock. For a coupled project, the EMT model should represent the integrated facility: the data center load (UPS front ends, cooling drives, protection logic), the generation (converters or machines with their full control stacks), and the shared network between them, capable of stable initialization, operation against Thévenin equivalents, impedance scans, and unbalanced fault simulation.

And they must agree with each other

SPP verifies RMS-vs-EMT model consistency. On a coupled project, inconsistency doesn't just delay one study — it can contaminate both the HDPS and the HILLGA SIS simultaneously. The load models, generator models, and EMT model should be developed as a matched set, benchmarked before submission, by a team that sees the whole system.

How Keentel helps:

Keentel Engineering develops the complete model set for HILL + HILLGA projects — CMLD and PERC1 load models parameterized from your actual equipment, generator dynamic models, and integrated PSCAD facility models — mutually benchmarked and packaged to SPP submittal formats, so both 90-day clocks start on time and neither study stalls on a model comment cycle.

Part 4: The Withdrawal Study Know Your Grid Headroom Before SPP Does

Every large-load interconnection ultimately turns on one number that nobody hands you:

how much power can this site actually withdraw from the transmission system, hour by hour, across a real year of grid conditions?

That number — the site's withdrawal capacity — is not the load's nameplate, and it is not a constant. It moves with seasonal facility ratings, system load levels, generation dispatch, and outage conditions.

A POI that supports 245 MW of withdrawal on a mild spring night may support far less on a peak summer afternoon with a line out of service.

The transmission owner and SPP will eventually tell you what the grid can deliver — in their study, on their clock, after your deposits are paid.

A withdrawal study answers the question first, on your side of the table.

A rigorous withdrawal study has three layers:

1. The Shadow Load Study (SLS)

Before submitting anything, replicate the analyses SPP and the host transmission owner will perform on your application: power flow with thermal and voltage screening at the POI, the applicable contingency set (aligned to NERC TPL-001-5.1 and SPP planning criteria), short circuit review, and system-strength indication consistent with SPP's SCRCCT methodology. The output is foresight — the expected study outcome, the binding constraints, the likely network-upgrade exposure, and whether SCRCCT screening (and therefore the supplemental EMT stage) is a live risk — all known before the 90-day clock starts and before the application locks in your requested MW.

2. The 8760 withdrawal-capacity model

Extend the SLS across the year: map representative system conditions — seasonal ratings, load levels, dispatch patterns, and outage states — onto all 8,760 hours to produce an hourly time series of the grid headroom available to your load. The deliverables that matter: the hourly withdrawal-capacity profile, its duration curve, and the deficit-hour analysis — the hours, depths, and durations when grid withdrawal capability falls short of the facility's demand.

3. BESS and generation sizing from the deficit hours

The deficit analysis converts directly into design: the behind-the-meter battery's power rating comes from the deepest deficits, its energy rating from the longest ones, and the dispatch strategy from their frequency and clustering. The same profile sizes on-site generation, shapes the phasing plan (how much load can energize in year one versus after upgrades), and — for a HILLGA project — informs the generator's operating envelope under LLRIS, where injection follows the load's forecast. This is the arithmetic of powered land development in a grid-constrained era: grid withdrawal capacity + BESS + behind-the-meter generation = firm capacity for a tenant, without waiting years for network upgrades.

Why does this analysis make or break projects?

Three reasons. First, it sets the application's requested MW intelligently — oversubscribe the POI and the study returns upgrade costs and delays that kill the pro forma; undersubscribe and you leave capacity, and revenue, on the table. Second, it converts study risk into design certainty: the BESS and generation are sized against measured headroom, not hope. Third, it collapses surprise: when SPP's HDPS or HILLGA SIS results arrive, a developer who ran the shadow study is confirming expectations and negotiating details — not discovering, five months and a deposit later, that the site supports half the load.

The dynamic dimension still matters

Withdrawal capacity is the planning layer; the dynamic behavior of withdrawal is the compliance layer. Under LLRIS, the co-located generator's injection is tied to the load — so when the load suddenly drops (a fault ride-through event, a UPS transfer during a deep sag, a feeder trip, or SPP's six-reclose sequences), the generator must run back fast and stably, without overspeed, overvoltage, or uncontrolled islanding — and then take the load back as it recovers to ≥90% of pre-disturbance consumption within one second of voltage recovery. A complete withdrawal study therefore pairs the 8760 planning analysis with EMT simulation of load-rejection and load-recovery events, verified runback and emergency power controls, and protection coordination across the load, generation, and transmission interface.

How Keentel helps:

Keentel Engineering performs the complete withdrawal study stack — Shadow Load Studies replicating SPP/TO methodology, 8760 hourly withdrawal-capacity modeling with deficit-hour analysis and BESS power/energy sizing, and the dynamic load-rejection and runback verification that LLRIS projects must ultimately demonstrate — so your application is sized right, your storage is sized right, and SPP's study results arrive as confirmation, not surprise.

Part 5: Seven Pitfalls We See in HILLGA Planning

Submitting the HILLGA request before the HILL-Load Connection Study agreement exists.

The prerequisite is structural — the request isn't valid without it.

Sizing the generator to the load's nameplate instead of the formula.

The 110%+PRM/125% headroom is there to be used; leaving it unexamined strands optionality.

Skipping the PSCAD model at submission.

f SCRCCT screening fails — likely in weak areas — EMT analysis outside the 90-day clock begins only when a model exists.

Treating load models and generator models as separate workstreams.

SPP studies a coupled system; inconsistent models stall both studies at once.

Applying without a withdrawal study.

Requesting MW the POI cannot support — or failing to demonstrate stable generator behavior through load-rejection events — invites upgrade shocks, comment loops, and restudies.

Ignoring the jurisdictional lane question.

Behind-the-fence and distribution-connected generation may belong in BP7250's non-jurisdictional process, not Attachment BB — with entirely different steps.

Deferring the CPP exit plan.

Five years arrives quickly; NRIS/ERIS conversion, potential network upgrades, and the expedited designation pathway should be scoped at financial close.

Technical FAQ

How Keentel Engineering Supports HILL + HILLGA Projects

Keentel Engineering provides end-to-end engineering and application support for large loads and their supporting generation across SPP and North American markets:

• Complete SPP application assistance — HILL (Attachment AQ/AX) and HILLGA (Attachment BB) submittal packages, prerequisite sequencing, deficiency-cure support, and study-cycle coordination through HILLGIA execution.
• Withdrawal studies — our signature analysis for grid-constrained development: Shadow Load Studies replicating SPP/TO methodology, 8760 hourly withdrawal-capacity modeling with deficit-hour analysis and BESS power/energy sizing, plus the dynamic layer — load-rejection EMT simulation, generator runback and emergency power control demonstration, reclosing-sequence stability, and anti-islanding verification.
• Dynamic load modeling — CMLD and PERC1 — site-specific parameterization encoded to SPP's HILL Fault Ride-Through Requirements, delivered in SPP submittal formats with Business Practice 7850 documentation.
• PSCAD EMT modeling — integrated load + generation facility models, supplemental-stage-ready (Thévenin operation, impedance scans, unbalanced faults, energization), validated and benchmarked against the RMS models.
• Generator interconnection engineering — RMS dynamic models, IEEE 2800-2022 Clause 7 and PRC-024-4 compliance alignment, and SCRCCT risk assessment at the POI.
• Strategy and exit planning — LLRIS operating envelope analysis, forecast sensitivity, and CPP conversion (NRIS/ERIS) roadmaps aligned to the five-year HILLGIA term.

With a 32-engineer organization, P.E. services in all 50 states, and recent large-load engagements from 50 MW to 1 GW across ERCOT, CAISO, and PJM, Keentel brings data-center-scale modeling experience to the newest — and fastest — interconnection pathway in SPP.

Planning a co-located generation project — or already staring at a deficiency notice?

Contact Keentel Engineering at contact@keentelengineering.com or 813-389-7871, or visit keentelengineering.com to schedule a technical consultation.

Our Offices:

• Head Office – Tampa, FL: 400 N Ashley Dr, STE #2600, Tampa, FL 33602 · (813) 389-7871
• Austin, TX: 5900 Balcones Drive, STE 100, Austin, TX 78731 · (512) 591-0752
• Sacramento, CA: 1401 21st St, Ste R, Sacramento, CA 95811 · (916) 913-4524
• Baltimore, MD: 306 W Redwood St, STE 200, Baltimore, MD 21201 · (410) 225-2181
  
  This article reflects the SPP HILLGA framework as filed October 24, 2025 and approved by FERC in January 2026, based on SPP's published HILLGA overview (Attachment BB), the HILL process documents (RR696/RR720/RR724), and SPP's HILL Fault Ride-Through Requirements (V2.0). Deposit amounts, forms, and procedural details continue to evolve through the SPP stakeholder process — always confirm current requirements against Attachment BB, Business Practice 7250, and SPP's OASIS studies page before submitting. This article is general information, not legal or financial advice.

Keentel Engineering — Power System Studies | EMT & Dynamic Modeling | Interconnection Support