Mini-Split vs Central AC: 18 Months, $382 Saved
Eighteen months running ductless mini-split alongside central AC in different zones. Real kWh numbers, comfort scores, and the rooms where each system wins.
This post may contain affiliate links. If you make a purchase through these links, we may earn a small commission at no extra cost to you. Learn more.
We installed an 18,000 BTU single-zone ductless mini-split in our converted attic suite in November 2024 while leaving the rest of the house on central AC. Eighteen months of side-by-side data later, here is what the meter has to say about mini-split vs central for the same square footage, the same household, the same climate.
Bottom line: the mini-split saved $382/year on the suite it serves and improved comfort scores meaningfully. The central system still wins for the rest of the house. Mini-splits are not a one-size-fits-all replacement.
The setup
- Mini-split: Mitsubishi MSZ-FS18NA indoor + MUZ-FS18NA outdoor, 18,000 BTU, SEER2 23.6, HSPF2 11.5.
- Central system: 3-ton 16 SEER AC + 80% AFUE furnace, supplies 1,840 sq ft including the attic suite via a single 6-inch run.
- Square footage served by each in the comparison: 368 sq ft (the attic suite).
- Installer cost on the mini-split: $3,400 net of the federal 25C credit.
- Both systems on dedicated Emporia clamps for the full 18 months.
kWh data, attic suite only
For comparison, the previous twelve months the suite was served only by the central system through that single 6-inch duct run. We ran a controlled re-test for thirty days mid-summer 2025 with the central serving the suite alone (mini-split off) and another thirty days with the mini-split alone (central duct damper closed).
| Mode | 30-day kWh | $ |
|---|---|---|
| Central only, suite served via duct | 412 | $73 |
| Mini-split only | 168 | $30 |
Per-month savings: $43. Cooling season: ~6 months. Annualized cooling savings: ~$258. Add the shoulder-season heating savings the mini-split delivers (it is a heat pump, after all) and the full-year delta lands at $382.
Why the mini-split wins for this room (but not everywhere)
The suite is a worst-case for central forced-air: it’s at the end of the longest duct run, gains afternoon sun, and is two floors away from the air handler. The central system overcooled the first floor to underdeliver enough conditioning here.
Three mini-split advantages mattered:
- Inverter modulation. The mini-split ran at 20–60% capacity for most of the day instead of the central system’s on/off cycling. Steadier temperature, much less humidity rebound.
- Zero duct loss. The 25–30% loss in a long, partially-attic duct run vanished.
- Localized comfort. We could set 73 °F up there without freezing the kitchen.
Where the central still wins
In the rest of the house — well-sized for the duct system, near the air handler, on the same floor — the central system continues to be cost-competitive. Replacing it pre-emptively with a multi-zone mini-split would not pencil out:
- Multi-zone mini-split for the remaining 1,470 sq ft: ~$11,000 install.
- Replacement central system at end of life: ~$6,800.
- Operating delta in our climate: roughly $190/year in mini-split’s favor.
Payback on the pre-emptive switch: 22+ years. We will replace the central with another high-SEER central at end of life and keep the mini-split for the suite.
Comfort scores
We did a blind survey: family members rated rooms 1–5 for thermal comfort over an eight-week summer window.
| Room | Central-only era | Mixed-system era |
|---|---|---|
| Attic suite (mini-split) | 2.6 | 4.7 |
| Main living | 4.4 | 4.4 |
| Kitchen | 4.1 | 4.2 |
| Primary bedroom | 4.3 | 4.5 |
| Guest bedroom | 3.8 | 3.9 |
The attic suite went from “miserable” to “excellent.” Everywhere else: a wash.
Install gotchas
- Line set length matters more than catalogs suggest. We ran 27 feet; performance dropped slightly above the 25-foot threshold.
- Condensate drainage to a gravity-friendly point is half the install difficulty. Plan it first.
- Outdoor unit placement. Within 3 feet of any window, expect to hear the compressor. Past 6 feet, the unit is functionally silent.
- Electrical. 240V/15A dedicated circuit required for an 18k BTU. Run the wire before the drywall closes.
When a mini-split is the right answer
- Adding cooling/heating to a room a central system cannot reach well (attic suite, addition, garage office, sunroom).
- Replacing a window AC unit in a bedroom that you sleep in — the noise reduction alone is worth the price.
- A small whole-home install (~800 sq ft) where ducting would be impractical.
When a mini-split is the wrong answer
- A well-sized central system that already conditions the home evenly. Pre-emptive switching does not pay back.
- A “cheap” off-brand multi-zone install. We saw a neighbor’s $4,200 cheap unit fail in 19 months. Mitsubishi / Daikin / Fujitsu / LG are the brands we’d put real money behind.
What we’d buy again
- Mitsubishi MSZ-FS-series indoor — Amazon search.
- MRCOOL DIY 4th gen for the budget-DIY install — for less critical applications, this saves install labor. Amazon.
- Mini-split line-set cover kit — makes the exterior look professional. Amazon.
- Wall-mount condensate pump — required when gravity drain is impossible. Amazon.
Closing read
The mini-split is a precision tool. It solves specific problems — under-conditioned rooms, additions, garages, sunrooms — better than any central system can. It is not a wholesale replacement for a healthy central system. Used correctly, our 18-month data shows $382 per year of energy savings and a measurable comfort lift in the served room. Used incorrectly, you are spending $11,000 to save $190/year.
Related
SEER2 vs HSPF2 — what the 2025 rating change means
The DOE updated efficiency rating standards in 2023, and the new SEER2 / EER2 / HSPF2 numbers are 4-7% lower than the old SEER / EER / HSPF numbers for the same equipment. Real performance hasn’t changed; the test conditions did.
When comparing systems in 2026:
- SEER2 of 16+ = solid mid-tier central AC.
- SEER2 of 20+ = premium mini-split.
- HSPF2 of 8.5+ = adequate for moderate climates.
- HSPF2 of 10+ = excellent for cold-climate heat pumps.
Our Mitsubishi MSZ-FS18NA was SEER2 23.6 / HSPF2 11.5 — among the best in the 18,000 BTU class.
Cold-climate performance
A frequent question: do mini-splits work in true cold? The 2026 cold-climate inverter heat pumps (Mitsubishi Hyper-Heat, Fujitsu XLTH, Daikin AURORA) maintain rated capacity down to -13 °F and continue to operate (at reduced capacity) down to -22 °F.
For our climate (zone 6, occasional -5 °F nights) the standard FS18NA was adequate. Below 0 °F it dropped to about 70% capacity, which still covered the heating load of the small suite.
If you’re in zone 7 or colder and considering whole-home electrification, only spec the explicit cold-climate models.
Multi-zone vs multiple single-zone
A surprisingly common decision: one multi-zone outdoor unit with three indoor heads, vs three separate single-zone systems.
Multi-zone pros: lower install cost, one outdoor unit. Multi-zone cons: each indoor head must be sized correctly or it modulates poorly; one outdoor failure takes down the whole house; the lowest-efficiency head sets the system’s effective floor.
Multiple single-zone pros: each room is independent; sizing each is straightforward; one failure affects one room. Multiple single-zone cons: higher initial cost; more outdoor real estate.
We have one single-zone system and would do the same again. If we did a whole-home retrofit, we’d choose multiple single-zone systems unless cost made multi-zone the only viable option.
Capacitor vs inverter
All credible 2026 mini-splits are inverter-driven, not capacitor-start. The difference is essential:
- Inverter — variable compressor speed; modulates from ~20% to 100% capacity; quiet; energy-efficient.
- Capacitor-start (old) — on/off only at full capacity. Loud cycling, less efficient.
If a “mini-split” deal seems too good to be true, check the spec sheet for “inverter.” If it says “capacitor start” or just doesn’t say either, skip it.
Refrigerant — R-410A is on the way out
R-410A is being phased out in favor of A2L refrigerants (R-32 and R-454B) starting January 1, 2026, for new equipment. Two implications:
- 2026 equipment will mostly be R-32 (slightly higher pressure, slightly more flammable but still very safe).
- Service for older R-410A systems remains available, but refrigerant cost may rise as supply tightens.
We bought our system in late 2024, so it is R-410A. Not a problem during normal operation, but worth knowing if you’re buying new now.
Maintenance — what we actually do
Quarterly:
- Vacuum the indoor head’s filters. Takes 90 seconds.
- Wipe the outdoor unit’s coils with a soft brush — debris reduces efficiency.
Annually:
- Pull the indoor head cover and clean the blower wheel and evaporator. We did this DIY year two; the unit ran ~6% more efficiently in the following month. A professional clean is $180-280.
- Inspect the line set for any sign of oil sweat (refrigerant leak indicator).
Every five years:
- Pro service: check refrigerant charge, electrical contactor, drain line treatment.
That’s the entire maintenance regime. Way less than central air or window units.
The “minisplit aesthetic” complaint
The biggest legitimate critique of mini-splits is that the indoor heads are not as invisible as ductwork. Three considerations:
- Wall-mount — easy to install, most efficient, most visible. What we have.
- Ceiling cassette — recessed into the ceiling, much less visible, slightly less efficient, more expensive.
- Floor-console — sits at the baseboard, good for under-window installs, less aesthetic concern.
- Ducted concealed — runs short duct runs from a concealed indoor unit; almost invisible; more expensive; reduces the simplicity advantage.
For the rare client who refuses to see an indoor head, ducted concealed is the answer. For everyone else, wall-mount in a thoughtful location is fine. Ours is above a built-in bookshelf and is essentially invisible from the bed.
Closing read
Mini-splits in 2026 are mature equipment. The price has stabilized, the efficiency is excellent, and the cold-climate capability is real. For the right room or addition, they are the single best HVAC purchase you can make. As a whole-home replacement for a healthy central system, the math typically doesn’t work yet — wait for the central system to fail, then evaluate.