Nuclear: Existing Designs vs Upcoming SMRs
TL:DR; Starting now with the existing designs makes more sense
Make the most of today, for you never know what tomorrow may bring.
— Gemini (A.I.)
The Colorado Energy Plan and several state legislators are strong on nuclear power. But only for Small Modular Reactors (SMRs) and not considering the existing designs that can be built today. I think a large part of this is the SMRs are no more than wonderous marketing promises today and so they sound great. Perfect even.
But they’re not. Reasonable estimates can be made as to when they will start coming off the assembly line, what they will cost, and the permitting and reviews required. So let’s dive in to what we know we can get today vs. what we estimate we can get tomorrow.
Two nuclear options dominate the conversation: building proven large reactors like South Korea’s APR14001 now or waiting for emerging SMRs. I’ll breaks down the costs, timelines, and trade-offs of each approach, assuming replacement of a coal plant (no new transmission lines needed).
Option 1: Build the APR1400 Now
Overview: The APR1400 is a 1,400-megawatt (MW) pressurized water reactor with a track record in South Korea and the UAE. Four APR1400 units (totaling 5.6 GW) would match the output of a large coal plant.
Timeline:
Regulatory approval: The U.S. Nuclear Regulatory Commission (NRC) certified the APR1400 design in 20192, but site-specific licensing under the National Environmental Policy Act (NEPA) typically takes 2–5 years for environmental reviews, public hearings, and safety evaluations.3, 4
Construction: South Korea’s KHNP estimates 7–10 years from groundbreaking to operation for new APR1400 units.5 Recent projects like Shin Hanul 3&4 (delayed by policy shifts) highlight risks, but standardized designs aim to streamline timelines.6
Cost:
Capital cost: ~$5–6 billion per reactor ($3,500–4,300/kW).7 For 5.6 GW (4 reactors), total costs could reach $20–24 billion.
Advantages: Proven technology, economies of scale, and predictable output (1,400 MW per reactor).8
Pros: Proven technology, high output, lower cost per GW.
Cons: Long lead time, large upfront investment, risk of delays.
Option 2: Wait for Small Modular Reactors (SMRs)
Overview: SMRs are, as the name suggests, smaller nuclear reactors, typically producing 300 MW or less per module. The idea is that they can be factory-produced and shipped to the site, reducing construction time and costs.
Several designs are under development, with varying levels of maturity. There are no existing units. Not even a prototype. I believe they’ll get there but there is great uncertainty as to when, the actual pricing once they are available, etc.
Timeline:
Regulatory approval: No SMR design has full NRC certification yet. The NuScale VOYGR (77 MW) took 42 months for design approval,9 but site-specific licensing under NEPA may still take 4–6 years.10
Deployment: Analysts project SMRs won’t be mass-produced until the mid-2030s.11 Supply chains and factory infrastructure are still nascent.12
Cost:
Capital cost: Current estimates range from $6,000–9,000/kW for SMRs.13 For 5.6 GW (~19×300 MW units), total costs could hit $33–50 billion.
Economies of scale: Costs may drop with mass production, but early projects like NuScale’s canceled Utah plant saw costs rise 75%.14
Head-to-Head Comparison
Key Trade-Offs
Time to Deployment
If immediate action is necessary to replace retiring coal plants, the APR1400 offers a clear advantage. With a shorter overall timeline (12–16 years versus 15–20+ years for SMRs), it ensures earlier decarbonization benefits. Delaying investment in favor of SMRs risks prolonging reliance on fossil fuels during the interim period.
Scalability and Flexibility
One major benefit of SMRs is their modularity. Unlike the massive APR1400, SMRs can be added incrementally, allowing utilities to match supply with demand growth more precisely. Additionally, their smaller size makes them suitable for locations where large reactors aren’t feasible.
Financial Risk
The APR1400 represents a tried-and-true technology with predictable costs and performance. While SMRs could eventually undercut large reactors, today’s costs are far higher.15
Grid Stability
APR1400s provide steady baseload power. Grid inertia is critical and large generators, i.e. large nuclear and coal, provide that. SMRs will be similar to large gas turbines where they increase inertia but are not large enough to provide the base inertia.
Conclusion
Building APR1400 reactors today offers a known path to decarbonize quickly, albeit with higher upfront costs. Waiting for SMRs gambles on unproven cost reductions and regulatory efficiencies—a risky bet for regions needing reliable power now. For utilities, the choice hinges on whether “perfect”16 (SMRs) should be the enemy of “good enough” (APR1400) in the race to net zero.
I think rapid deployment and proven reliability are paramount and so the APR1400 is the better choice. Four APR1400s delivering 5.6 GW would come online within 7-9 years, providing carbon reductions within a decade.
If we wait for SMRs we’ll wait longer and very likely pay more. I know the SMR companies are promising a better solution real soon now. But that puts me in mind of a common statement in the software industry - “What’s the difference between a car salesperson and a software salesperson? The car salesperson knows when they’re lying.”17
I choose the Korean APR1400 over the Westinghouse AP1000 because the Korean company builds the plants at about half the cost.
Prediction - they won’t be perfect.
The people involved in SMRs should be highly optimistic. That’s required in a successful startup. But we need to be realistic.
APRs all the way, proven design and years of operational experience is key to safely and reliably operating these machines,
Trust me, I know, 42 years in nuclear plant operation, maintenance and engineering. The US industry has a 90%capacity factor for mainly one reason, operational experience, OE.
We have learned how to operate and maintain these large plants. We have zero operational experience with SMR’s. that’s a problem.
It's still a generous assumption that a 4-unit APR1400 could be constructed in the United States for $24B. I'd offer that a decent estimate would be to take the original Vogtle per unit estimate of $7B and adjust to 2025 dollars which produces a $10B/unit cost. That would see a $41B cost for a 4-unit APR1400 facility. A 50% premium for infrastructure construction in the United States feels right.
Trying to get that IRP approved by the PUC would be challenging. You'd also need a significant amount of transmission upgrades for such a facility as the output is at the level that the Colorado Power Pathways supports . Transmission export capacity out of Colorado would be necessary as well to market the output above load as such a plant would see Colorado as one of the largest power exporters in the WECC. That right there is likely another $2B or in network upgrades. Colorado could probably see some nice energy sales in this scenario, though probably not enough to offset the higher resource portfolio cost.