NSF's Instrumentation Program Returns With $50M and No Submission Cap — Solicitation 27-500 Closes January 29

Academic PIs who have spent two years deferring a cryo-EM, a campus GPU cluster, or a replacement mass spectrometer now have a hard date: NSF's Major Research Instrumentation Program reopened as solicitation 27-500 on November 6 with $50 million anticipated and proposals due January 29, 2027 (Grants.gov opportunity 364243).

The window is 25 days long and it opens January 4

The Major Research Instrumentation Program went dark for FY2026 — NSF cancelled that competition outright rather than running it on a reduced budget. The replacement solicitation, NSF 27-500, posted November 6, 2026 and listed on Grants.gov as opportunity 364243, formally replaces NSF 23-519 and sets the first full-proposal window at January 4 through January 29, 2027, due by 5 p.m. submitting organization's local time.

Two further windows are already published: October 18 – November 19, 2027, and October 16 – November 17, 2028. Worth noting for anyone planning off the program landing page — that page currently lists only the first two windows, while the solicitation itself lists all three. For the January 2027 window, the earliest expected start date is October 1, 2027. A PI submitting in eight weeks is budgeting for an instrument that gets installed in FY2028, which matters for vendor quotes that carry 90-day price validity.

We wrote earlier this fall that MRI would likely reopen in an October 15 – November 16, 2026 window under the old NSF 23-519 rules, with the four-proposal institutional cap and the five-year cost-share waiver as the two load-bearing strategic facts. NSF did something different. It skipped the fall window and rewrote the program's architecture instead.

Tracks are gone. So is the four-proposal cap.

The revision notes in 27-500 run to two bullets, and both are structural:

• "Organizational submission limits have been eliminated."
• "Tracks have been eliminated (though Helium recovery equipment proposals will still be considered separately from those for research instruments)."

Under NSF 23-519, an institution could submit two proposals in Track 1 (requests above $100,000 and below $1.4 million), one in Track 2 ($1.4 million to $4 million), and one in Track 3 (helium conservation equipment) — four total, and no more. That cap is why MRI generated one of the most contentious internal competitions on any research campus. Sponsored programs offices ran limited-submission pre-proposal rounds in September and October, faculty senates argued about whose instrument got the Track 2 slot, and a deserving proposal could lose without NSF ever reading it.

27-500 removes all of it. The solicitation states flatly that there are "no restrictions or limits" on proposals per organization, proposals per PI or co-PI, or the number of PIs. The internal-selection bottleneck is gone.

That is unambiguously good news for individual PIs and a genuine operational problem for research offices. Every institution that built a limited-submission process around MRI now has to decide what replaces it — because the absence of an NSF cap does not mean an institution should submit twelve proposals. It means the gatekeeping moved from a rule to a judgment call, and the judgment call now has to be made in six weeks over the winter holidays.

A third less money chasing an unbounded number of proposals

Here is the part that should reset expectations. NSF 23-519 anticipated $75,000,000 and approximately 100 awards. NSF 27-500 anticipates $50,000,000 and approximately 60 awards — both "subject to the availability of funds."

Run the arithmetic and the picture is specific: the program pool is down 33%, award count is down 40%, and average award size is actually up about 11%, from roughly $750,000 to roughly $833,000. NSF is funding fewer, somewhat larger instruments out of a smaller pot.

Now layer the eliminated submission cap on top. Under 23-519, total national proposal volume was mechanically bounded — four per eligible organization. Under 27-500, it is bounded only by what institutions choose to send. If submission volume rises even modestly while the award count drops from 100 to 60, the effective success rate falls faster than either number alone suggests. The removal of the cap and the reduction of the budget are not independent facts; together they describe a competition that is meaningfully harder to win than the FY2025 cycle, even though it looks more open.

The second-order effect matters too. NSF 23-519 reserved roughly a quarter of MRI funding for Track 2 — the $1.4 million-and-up instruments. With tracks eliminated, there is no protected slice. A $3.5 million NMR spectrometer now competes head-to-head in a single pool against $250,000 requests. The residue of Track 2 survives as a review requirement rather than a budget set-aside: proposals requesting over $1.4 million must address the instrument's potential impact on the research community "at the regional or national level." If you are asking for seven figures, that regional-or-national argument is no longer a nice-to-have paragraph. It is the thing distinguishing your proposal inside an undifferentiated pool.

Where the budget actually fails: the 70% equipment rule

MRI budgets get returned without review for mechanical reasons more often than they get declined on merit, and 27-500 keeps the rule that causes it. For acquisition proposals, at least 70% of the Total Project Cost must consist of items on the Equipment line of the NSF budget form (Line D).

That constraint quietly kills a lot of otherwise strong projects. Installation labor, site preparation, the first year of a technician's salary, training, extended service contracts — none of it sits on Line D. A $1,000,000 acquisition request can carry at most $300,000 of everything that is not the instrument itself. PIs who have scoped a project around "the instrument plus the staff to run it" need to restructure before January, not after.

The request range is $100,000 to $4 million. Cost sharing is settled and simple: "Inclusion of voluntary committed cost sharing is prohibited." Because voluntary committed cost sharing is off the table, the NSF request is effectively the total project cost — there is no mechanism to make a $5 million instrument work by promising institutional match. Acquisition projects are funded for up to three years (36 months on the cover sheet); development projects for up to five (60 months).

One more budget-adjacent detail that rarely appears in solicitations: 27-500 indicates NSF may give preference to institutions with lower indirect cost rates when proposals are otherwise comparable after merit review. Indirect cost limitations themselves are listed as "Not Applicable," so this is a tiebreaker rather than a cap — but it is in the text, and it is new information for anyone modeling their competitive position.

The operations-and-maintenance letter is the real institutional gate

With the submission cap gone, the sharpest institutional commitment required by 27-500 is a supplementary document: a 1–2 page letter documenting the performing organization's commitment to ensuring successful operations and maintenance of the instrument. That letter must also list the organization's MRI awards with a start date in the previous five calendar years.

Read that second requirement carefully, because it does the work the four-proposal cap used to do. Every proposal from an institution now discloses that institution's recent MRI award history to reviewers. An organization submitting its eighth proposal of the cycle, with four MRI awards since 2022 and a thin O&M commitment, is making a visible argument about its own capacity. The cap was removed; the accountability was relocated.

Practically, this means the signature you need is not your chair's. It is whoever can credibly commit facilities, technical staff, and a service budget for the life of the instrument — and that person's calendar in late December is the actual constraint on your January submission.

Who is eligible, and the under-$100,000 exception

Eligibility is narrower than the general NSF default. 27-500 limits proposals to:

• Institutions of higher education — two- and four-year IHEs, including community colleges, accredited in and with a campus located in the United States
• Non-profit, non-academic organizations located in the U.S., such as museums, observatories, and research laboratories
• Federally recognized Tribal Nations

The floor has an exception worth knowing: MRI proposals must generally request between $100,000 and $4 million, except proposals (a) in mathematics or (b) from non-Ph.D.-granting institutions of higher education, in any NSF-supported discipline. Those may request budgets under $100,000. Non-Ph.D.-granting institutions invoking the exception must submit a separate letter certifying that status. For a community college or a primarily undergraduate institution, this is the most accessible instrumentation money in the federal portfolio — a $60,000 request is in scope, and the organizational cap that previously forced you to compete internally against your own R1 partner institution is gone.

What MRI will not fund has not changed: construction or renovation, general-purpose equipment, sustaining existing infrastructure, instruments used primarily for instructional courses, research primarily outside NSF-supported fields, and "multiple, independent instruments" bundled into one request. That last exclusion is the most commonly misread. A shared-facility upgrade consisting of three unrelated instruments is not one MRI proposal; under the old rules it was not even two, because of the cap. Now it can be three separate proposals — which is the single largest practical change 27-500 delivers to PIs.

The solicitation spans nine assistance listings (47.041 Engineering, 47.049 Mathematical and Physical Sciences, 47.050 Geosciences, 47.070 CISE, 47.074 Biological Sciences, 47.076 STEM Education, 47.079 OISE, 47.083 Office of Integrative Activities, and 47.084 TIP), so the disciplinary reach is effectively all of NSF. Program contact is mri@nsf.gov.

What to do in the next eight weeks

The sequence that works, in order: get a dated vendor quote with price validity extending past October 2027; restructure the budget so Line D equipment clears 70% of total project cost; identify the administrator who will sign the operations-and-maintenance letter and get on their calendar before December 15; and if your request exceeds $1.4 million, draft the regional-or-national impact argument first, because it determines whether the rest of the proposal is worth writing.

If you are hunting for the instrumentation and facilities programs that pair with an MRI request — agency-specific equipment supplements, state matching funds, and the shared-facility mechanisms that can cover the operations costs MRI's 70% rule excludes — search active research instrumentation and equipment solicitations on Granted.

The window is 25 days. The cap is gone. The money is down a third. Plan accordingly.