Tesla Powerwall 3 vs Enphase IQ Battery 5P 2026: Complete Comparison
Picking between a Tesla Powerwall 3 and Enphase’s IQ Battery 5P isn’t really a spec sheet exercise — it’s a decision about architecture. Tesla gives you one big box that also happens to be your solar inverter. Enphase gives you small LFP units stacked behind the microinverters you probably already own. After designing a lot of residential storage systems, I can tell you that most of the “which is better” arguments online miss the point: the right answer depends on your existing equipment, your utility’s rate structure, and whether your state still has real net metering.
Both products got meaningful updates in the last cycle. Tesla’s Powerwall 3 folded the solar inverter into the battery and moved to LFP cells. Enphase pushed the 5P to higher per-unit power and kept the modular story intact. This comparison pulls from manufacturer datasheets, field experience on installed systems, and the 2025 UL/NEC compliance picture.
Quick Verdict
Best overall for most new installs: Tesla Powerwall 3 — 13.5 kWh usable, integrated hybrid inverter with 6 MPPTs, roughly 11.5 kW continuous AC output. Typical installed cost lands in the $14,000–$18,000 range before the 30% federal credit, heavily dependent on panel upgrades and line length to the main service.
Best for existing Enphase homes: Enphase IQ Battery 5P — 5 kWh usable per unit, 3.84 kW continuous (7.68 kW surge for a few seconds), LFP chemistry, 15-year warranty. Street pricing runs roughly $3,500–$5,000 per unit installed, with most backup-oriented homes needing two or three.
Best cost per usable kWh: Tesla Powerwall 3 — once you account for the inverter Enphase customers still need to buy or keep, Tesla’s $/kWh lands meaningfully lower on whole-home jobs. Enphase only catches up when you already have working IQ7/IQ8 microinverters on the roof.
How I Evaluated These Systems
I want to be upfront: I didn’t run a controlled lab test with cycle-counting equipment. What’s below is a synthesis of manufacturer datasheets, UL 9540 and UL 9540A test reports, installer field experience across jobs in the Northeast, Southwest, and Pacific Northwest, and public data from Tesla’s own Powerwall warranty documentation and Enphase’s Enlighten monitoring ecosystem. Any capacity numbers, efficiency figures, and backup durations below are either pulled from official specs or described qualitatively from what installers actually see in the field. Where I give a range, assume real-world variance is wider than the datasheet implies.
Battery Comparison Table
| System | Best For | Usable kWh | Continuous AC Power | Typical Installed | Chemistry | Honest Rating |
|---|---|---|---|---|---|---|
| Tesla Powerwall 3 | New whole-home installs | 13.5 | ~11.5 kW | $14k–$18k | LFP | Strong |
| Enphase IQ Battery 5P | Existing Enphase homes | 5.0 | 3.84 kW (per unit) | $3.5k–$5k each | LFP | Solid but pricey per kWh |
| Tesla Powerwall 2 | Retrofit with existing string inverter | 13.5 | 5.0 kW | $11k–$14k | NMC | Aging, still fine |
| Enphase IQ Battery 10T | Older Enphase expansion | 10.08 | 3.84 kW | $8k–$11k | LFP | Being phased out |
| FranklinWH aPower 2 | Powerwall alternative | 15 | 10 kW | $13k–$17k | LFP | Worth cross-shopping |
Note the rating column isn’t a score — the “9.2 vs 8.7” number theater you see on most affiliate sites is meaningless without knowing the buyer’s utility and roof. Take it as a directional ranking.
Tesla Powerwall 3 — Best All-in-One for New Installs
Best for homes installing solar and storage together
The Powerwall 3 is not just a bigger battery. Tesla rebuilt the box around an integrated hybrid inverter with multiple MPPT inputs, which means on a new install you’re not paying for a separate string inverter or a fleet of microinverters. On a typical 10 kW PV job, that eliminates roughly $2,000–$3,500 of equipment and labor — the kind of saving that actually changes the payback math.
Published specs: 13.5 kWh usable capacity, around 11.5 kW continuous AC output, LFP chemistry, and a 10-year warranty with 70% end-of-warranty capacity retention. Tesla’s DC-coupled architecture on the PV side is genuinely efficient because the solar → battery path skips a DC→AC→DC conversion step. Whether you actually see the advertised round-trip efficiency depends on how you’re measuring — AC→battery→AC on a steady 5 kW discharge is a best case, not your real-world average with variable household loads.
Installed pricing varies enormously. A greenfield install with a 200A main panel and the battery mounted near the service drop lands around $14,000–$15,500. Add a panel upgrade (common on 100A services), long conduit runs, a critical load subpanel, or a difficult inspection jurisdiction and you’re closer to $17,000–$18,000. After the 30% residential clean energy credit, most homes net out around $10,000–$13,000. Be skeptical of any quote that comes in dramatically below this — it’s usually missing the interconnection or permit fees, or your installer is cutting corners on the critical load subpanel.
What it does well:
- Integrated hybrid inverter reduces total equipment count on new installs
- 11.5 kW continuous is enough to ride through a central AC startup without load shedding
- LFP chemistry is safer than the NMC cells in the old Powerwall 2 and tolerates deeper daily cycling
- Rapid shutdown compliance handled at the unit level (NEC 2017/2020 690.12)
- Backup transition is fast enough that desktops and most networking gear don’t reboot
Real limitations:
- Tesla’s installer and service network is the single biggest risk. Parts pipeline and truck rolls for Powerwall warranty work have been the #1 complaint from customers for years. If you live somewhere without a nearby Tesla Energy service presence, a failure can mean weeks of waiting, and third-party techs often can’t touch the unit under warranty.
- The integrated inverter is a single point of failure. On a Powerwall 2 + string inverter setup, an inverter death leaves the battery functional as a backup. On a Powerwall 3, an inverter-side failure takes the entire system offline, including solar production.
- Scaling is chunky. You add capacity in 13.5 kWh increments. If your real need is 18 kWh, you’re buying 27.
- It’s a Tesla ecosystem commitment. Integration with non-Tesla monitoring, third-party energy management systems, and some VPP programs is limited compared to open-protocol alternatives.
Enphase IQ Battery 5P — Best for Existing Enphase Homes
Best for homes that already run IQ7/IQ8 microinverters
The 5P makes sense when you already have Enphase on the roof. The unified Enlighten monitoring, the same installer network, the same rapid shutdown architecture — it all drops in without re-engineering your PV system. The modular story (start with one, add more later) is also real, though it’s oversold when people assume linear cost scaling.
Datasheet basics: 5 kWh usable, 3.84 kW continuous per unit, 7.68 kW peak for roughly 3 seconds, LFP chemistry, and a 15-year / 6,000-cycle warranty. That warranty length is legitimately best-in-class and matters if you plan to daily-cycle for TOU arbitrage. The LFP chemistry, per-unit embedded power electronics, and lack of high-voltage DC runs make this one of the safer residential battery architectures on the market.
Pricing reality: installers quote roughly $3,500–$5,000 per unit installed, depending on region and how much electrical work is bundled in. Most homes that want meaningful backup land on two or three units — call it $8,000–$14,000 total, or $5,600–$9,800 after the federal credit. That’s more per usable kWh than a Powerwall 3. People forget this when they fixate on the low per-unit price.
What it does well:
- LFP chemistry and low-voltage DC architecture are genuinely safer than NMC alternatives
- Enlighten monitoring gives you per-microinverter and per-battery data that Tesla’s app does not
- Modular expansion lets you phase investment, which matters when SGIP or state rebates reset
- Distributed power electronics mean a single module failure doesn’t kill the whole system
- 15-year warranty with explicit cycle count is the most transparent in the market
Real limitations:
- You will spend more per usable kWh than with Tesla. This is the single biggest strike against the 5P for new installs. Modular premium is real.
- 3.84 kW continuous per unit is genuinely limiting. A single battery cannot start a 3-ton central AC compressor without load shedding. For whole-home backup on anything larger than a 1,500 sq ft house, you need at least two, often three, just to cover startup inrush.
- Scaling from 3 to 4+ units gets complicated. Additional subpanel work, conductor sizing, and permitting friction accumulate. The “just add another” pitch is true for the first and second unit, less true for the fourth.
- Ecosystem lock-in is real. You’re betting on Enphase’s long-term software and firmware support. That’s been fine historically, but it’s a real dependency.
Tesla Powerwall 2 — Still a Reasonable Retrofit
Best for retrofit jobs with a working string inverter
The Powerwall 2 is still out there through some channels, and for retrofits where you already have a functional SolarEdge, SMA, or Fronius string inverter, it’s a cheaper entry point — typically $11,000–$14,000 installed. Same 13.5 kWh, but older NMC chemistry, 5 kW continuous AC (not 11.5), and lower round-trip efficiency.
The honest take on Powerwall 2 in 2026: it’s AC-coupled, so you lose some efficiency going PV → inverter → AC → battery → AC → loads versus the Powerwall 3’s DC-coupled path. The 5 kW continuous limit means load shedding during backup is routine on any home with central AC. It’s fine if your roof already has a working inverter with useful warranty years left, but I wouldn’t specify it on a new job when the Powerwall 3 exists.
Real limitations:
- 5 kW continuous forces load management during outages
- NMC chemistry has higher fire load than LFP alternatives
- Tesla is prioritizing Powerwall 3 for parts and support
- Round-trip efficiency is notably worse than Powerwall 3 on DC-coupled installs
Use Case Recommendations
Longest Backup Duration
Tesla Powerwall 3. On a typical load profile with fridge, WiFi, a few lights, and intermittent furnace blower, 13.5 kWh gets you most of a day. A single Enphase 5P gets you maybe a quarter of that. Three 5Ps match Tesla’s duration but at higher total cost and more installation complexity.
Existing Enphase Solar Systems
Enphase IQ Battery 5P, obviously. Same monitoring platform, same installer relationship, same rapid shutdown compliance scheme. Swapping in Tesla here means adding a second monitoring app, a second support path, and usually a CT-based energy meter retrofit — not impossible, but annoying and pricier than most installers will tell you up front.
Phased Investment
Enphase if you genuinely plan to add capacity later. One caveat: if your state’s storage incentive program is sunsetting (SGIP in California has been stepping down for years), waiting to add batteries means you claim smaller incentives on later units. The “buy as needed” plan looks best on paper and often underperforms in practice because incentives don’t wait.
Whole-Home Backup on a Larger House
Tesla Powerwall 3 for anything above roughly 2,500 sq ft with central AC. The 11.5 kW continuous handles compressor inrush without tripping. You can get there with three or four Enphase 5Ps, but the cost delta stops being small.
Tight Installation Space
Enphase 5P wins purely on form factor — it’s small, light enough for two people to wall-mount without equipment, and tolerates a wider outdoor temperature range. If your only viable install location is a narrow utility closet or an exterior wall with limited clearance, this actually matters.
Off-Grid or Intermittent-Grid
Neither of these is a real off-grid solution. Both are grid-interactive backup systems. If you’re truly off-grid, look at Sol-Ark or Schneider XW Pro paired with an LFP battery bank — different category. If you just mean a cabin with bad utility service, Powerwall 3’s islanding behavior is more forgiving.
For panel recommendations to pair with either battery, see our Best Solar Panels 2026 guide — the DC/AC ratio you design matters more than the battery choice for production.
Pricing and ROI Reality Check
Here’s where I push back on the rosy payback numbers you see everywhere.
The Math Nobody Runs
The real ROI driver for residential storage is your utility’s rate structure, and it is changing fast. California’s NEM 3.0 replaced net metering with net billing in 2023, slashing export credits by roughly 75% and making battery self-consumption the only way to get reasonable ROI on new PV. Several other states are following. If you’re installing solar in 2026 under a net billing tariff, a battery isn’t a backup luxury — it’s how your solar pays for itself at all.
Under traditional 1:1 net metering (still available in some states), a battery’s economic value comes from backup only, and the payback is genuinely poor — 10 to 15 years isn’t unusual, and if your utility is stable you may never recover the cost. Don’t buy one and pretend otherwise.
Tesla Powerwall 3 Numbers (Honest Version)
Installed: $14,000–$18,000 After 30% federal credit: $9,800–$12,600 Monthly savings under TOU with net billing: highly variable. On California’s E-TOU-C with a 4–9pm peak, a fully cycled Powerwall 3 can save $80–$150/month versus peak grid consumption. On a flat rate utility, savings drop to $20–$40/month from self-consumption. Payback period: 7–11 years under TOU with net billing; 12–20+ years under flat rates or legacy NEM. If anyone quotes you 5 years, ask what electricity rate and discount rate they used.
State incentives change the picture significantly:
- California SGIP: Still available for equity resilience applicants, much smaller general market rebates
- Massachusetts SMART: Battery adders remain but step down each program block
- Texas: Utility-specific rebates (Oncor, CenterPoint) come and go
- New York NY-Sun: Storage adder available in some zones
Enphase IQ Battery 5P Numbers (Honest Version)
Installed, three-unit system: $10,500–$15,000 After 30% federal credit: $7,350–$10,500 Monthly savings: similar to Tesla scaled by usable kWh (15 kWh on three units vs 13.5 kWh on one Powerwall) Payback period: typically 1–2 years longer than Powerwall 3 on equivalent capacity, because you paid more per kWh to get there
The modular story helps if you’re only cycling a few kWh a day and your outage risk is low — you can start with one unit for a fridge-and-lights backup and save real money versus a full Powerwall.
State-by-state incentive details live in our Solar Panel Cost by State 2026 guide. The incentive landscape changes frequently, so verify current program status before finalizing your financials.
Performance — What Actually Matters in the Field
Round-Trip Efficiency Is a Slippery Number
Manufacturers quote round-trip efficiency under ideal steady-state conditions. Real-world efficiency drops because standby losses, inverter conversion losses, and household load variability all eat into the number. In the field, I’d expect a Powerwall 3 to deliver somewhere in the low-to-mid 90s depending on how you measure, and the Enphase 5P to land a few points behind that. Both are fine. The difference doesn’t change the economics.
Also: round-trip efficiency matters less than you think if your alternative is curtailing solar production. A battery at 90% efficient beats a solar system curtailing to zero at 100% efficient.
Capacity Degradation
Both manufacturers warrant 70% capacity retention at end of warranty (10 years Tesla, 15 years Enphase). Real-world LFP cells usually outperform this — 80% retention at 10 years is a reasonable expectation under moderate cycling. If you’re planning financials on 100% of nameplate for 25 years, redo them. Use 85% as a weighted average over the system’s life.
Backup Duration Reality
Backup runtime depends entirely on your load profile, and the “24–36 hour” claims on marketing pages assume essential loads only — fridge, LED lights, WiFi, phone chargers. The moment you run central AC, electric heat, an electric water heater, or charge an EV, runtime collapses.
Practical estimates:
- Essential loads only (200–400W average): Powerwall 3 runs 24–40+ hours. Single 5P runs 8–14 hours.
- Typical loads including some HVAC cycling (800–1,500W average): Powerwall 3 runs 8–16 hours. Single 5P runs 3–5 hours.
- Heavy loads (central AC, EV charging): both systems run down in 3–6 hours regardless of capacity.
If you live in a hot climate and expect to keep AC running through an outage, size for it. One Powerwall 3 will not cool a 2,500 sq ft house overnight.
Cold Weather
Lithium cells lose usable capacity at low temperatures — that’s physics, not a defect. Both systems include internal heaters to keep cells in their operating window, which draws from the battery and reduces effective capacity in winter. Expect 10–20% capacity loss at 20°F and more below 0°F. If you live somewhere that hits -20°F regularly, make sure the battery is installed in a conditioned or semi-conditioned space.
Installation and Integration
Tesla Powerwall 3
Installation takes a Tesla-certified crew roughly a day for a straightforward job, longer if your main panel needs upgrading. The integrated inverter simplifies wiring on new PV installs but requires the installer to be trained on the specific Powerwall 3 commissioning flow, which Tesla gates tightly.
Typical requirements:
- Compatible main service (most 200A panels work; 100A services often need upgrades)
- Critical load subpanel if you want partial-home rather than whole-home backup
- Indoor or sheltered outdoor location with adequate clearance
- Internet connectivity for commissioning and firmware updates
- Tesla Gateway 3 for grid interface and backup switching
Permitting: 2–6 weeks depending on jurisdiction. Some AHJs have slowed down on battery interconnections post-2022 because of fire code updates around UL 9540A testing.
Enphase IQ Battery 5P
Installation is simpler per unit because each battery is smaller and lighter, but the total time scales with unit count. Three-battery installs typically take a day.
Typical requirements:
- Existing Enphase envoy/gateway (or IQ System Controller for new installs)
- A dedicated 20A circuit per battery pair (varies by configuration)
- Adequate wall space and clearance per Enphase’s installation manual
- Compatible with existing IQ7 or IQ8 microinverters
Permitting: generally friendlier to AHJs because of the lower DC voltages and distributed architecture, but inspection is still required. Rapid shutdown is handled at the microinverter level — same as it’s always been on Enphase.
Monitoring
Enphase Enlighten gives you per-microinverter and per-battery data down to the individual unit level. Tesla’s app gives you slick energy flow visualization and storm watch, but less granular diagnostic data. If you are the kind of owner who wants to know exactly which module is underperforming, Enphase wins. If you just want to see “battery is at 67%” on your phone, Tesla’s app is fine.
Safety and Warranty
Chemistry Matters
Tesla Powerwall 3: LFP. Tesla moved the Powerwall 3 to LFP (lithium iron phosphate) cells — a meaningful improvement over the Powerwall 2’s NMC chemistry from a fire safety standpoint. LFP has a much higher thermal runaway threshold and won’t propagate fire cell-to-cell the way NMC packs can. This is the correct chemistry for a residential backup battery and I’d spec it over NMC every time.
Enphase IQ Battery 5P: LFP. Same chemistry story, with the additional advantage of distributed architecture — no high-voltage DC bus between units. UL 9540A testing has been favorable for both systems.
Warranty Comparison
- Tesla Powerwall 3: 10 years, 70% capacity retention, unlimited cycles for self-consumption and backup
- Enphase IQ Battery 5P: 15 years, 70% capacity retention, 6,000 cycles
Enphase’s 15-year warranty is genuinely better on paper. Whether you get value from it depends on whether Enphase is still a going concern in year 12. Both companies have financial ups and downs; neither is a sure thing for the full warranty period. Factor in installer warranty backup where possible.
Service Is Where Tesla Struggles
I’ll say it again because it matters: Tesla’s field service for residential Powerwall warranty claims is the biggest single weakness of buying one. Enphase’s installer-delivered support model generally gets a truck out faster, especially outside major metros. If you live somewhere rural, this weighs the decision toward Enphase.
For a broader comparison including Generac, FranklinWH, and SolarEdge options, see our Best Home Battery Systems 2026 guide.
EV Charging Integration
Tesla Powerwall 3
11.5 kW continuous is enough to run a Level 2 charger at most realistic home charging rates (typically 7.2–11.5 kW). Pairing a Powerwall 3 with a Tesla Wall Connector gives you coordinated charging logic — set your EV to charge from stored solar during off-peak hours, or pause charging during outages to preserve backup capacity. This coordination is genuinely useful under TOU rates.
Enphase IQ Battery 5P
A single unit delivers 3.84 kW continuous, which limits Level 2 charging to the slower end — roughly 16A at 240V. Practical home EV charging from an Enphase system really wants two or three units working together. Enphase’s coordination with third-party EV chargers is improving but isn’t as tight as Tesla-to-Tesla.
See our Best EV Chargers for Home 2026 guide for pairing options.
Scaling Up
Tesla scales in 13.5 kWh chunks. Most homes top out at 2–4 Powerwalls before panel capacity, interconnection limits, or AHJ rules start causing friction. Adding a second Powerwall after the fact is less cost-efficient than buying two at once because you pay mobilization and permitting twice.
Enphase scales in 5 kWh chunks and is designed around the idea of incremental adds. Most homes can accommodate 4–8 units before the critical load subpanel and main service limits push back. The phased approach only saves money if you actually catch future incentive windows — otherwise you’re just paying the installation crew twice.
Grid Services and VPP Realities
Both manufacturers participate in Virtual Power Plant programs, and the economics are more marketing than substance for most homeowners.
Tesla VPP in California pays roughly $2/kWh exported during emergency events, with typical annual participation yielding $100–$400 per Powerwall for most households. The higher numbers you see quoted are from outlier heatwaves. Don’t buy a battery based on VPP income — treat it as a bonus.
Enphase grid services programs work through utility partnerships and generally pay less per event but activate more frequently. Expected annual revenue is broadly similar — low hundreds per household.
Real talk: VPP income is nice, but it shouldn’t be in your ROI pitch unless you’re comfortable with the program being restructured or killed mid-contract. Utility incentive programs get changed or eliminated regularly.
Environmental Footprint
Both batteries use LFP chemistry, which eliminates the cobalt supply chain concerns that dominated the NMC conversation. LFP cells also last longer, so the lifetime material footprint per kWh delivered is better than older NMC packs. End-of-life recycling infrastructure for residential LFP is still immature in North America, but it’s moving — expect viable recycling paths by the end of this decade.
The carbon payback on either system paired with solar is typically 2–4 years in most US grid mixes, assuming the battery cycles daily. If your use case is backup-only with minimal cycling, the carbon payback stretches significantly — a battery that sits idle 355 days a year is embodied carbon without offsetting benefit.
Final Recommendation
For a new solar + storage install: Tesla Powerwall 3 is the default recommendation if you have solid Tesla service coverage in your area. The DC-coupled architecture, integrated inverter, and 11.5 kW continuous make it the highest-utility single-box solution for whole-home backup on the market. The service risk is the real tradeoff.
For an existing Enphase solar home: Enphase IQ Battery 5P. Don’t fight your existing architecture. The integration advantages, unified monitoring, and installer continuity are worth the higher $/kWh.
For retrofit homes with a working string inverter: Cross-shop Powerwall 2 (if still available in your market), FranklinWH aPower 2, and a smaller Enphase configuration. None is obviously the right answer — it depends on your inverter’s remaining warranty and your backup priorities.
The situation where neither makes financial sense: You live in a stable utility area with 1:1 net metering, infrequent outages, and flat-rate electricity. The payback under those conditions is genuinely bad — 12+ years — and I tell customers that honestly. A battery in that case is peace-of-mind equipment, not an investment. Buy one anyway if resilience matters to you, but don’t lie to yourself about the economics.
Frequently Asked Questions
Which battery works better with existing solar panels?
If your existing solar uses Enphase microinverters, get Enphase batteries — the integration is tight and the monitoring is unified. If you have a string inverter (SolarEdge, SMA, Fronius), either battery works. Powerwall 3’s integrated inverter is wasted in a retrofit with a healthy string inverter; Powerwall 2 or a competitor may be a better match. Powerwall 3 shines on new installs where it replaces the inverter entirely.
How long will these batteries run my house during an outage?
Honest answer: it depends entirely on what you’re running. Fridge, lights, WiFi, and a phone charger — Powerwall 3 easily runs 24+ hours, a single Enphase 5P runs 8–14 hours. Add central AC or electric heat and both systems run down in hours, not days. If outage protection is your main driver, size your battery to your actual backup loads, not to “whole house.”
Can I add more batteries later?
Enphase is designed for it — the 5P’s modular architecture scales cleanly in 5 kWh units up to a point. Tesla Powerwall scales in 13.5 kWh chunks and is lumpier. Either way, you’ll pay mobilization and permitting costs again on the add-on install, which eats into the “phased investment” narrative. The biggest reason to add batteries later is if you think incentive programs will improve, which is generally a bad bet — they usually step down.
Which battery has better monitoring?
Enphase Enlighten is more granular and gives you per-module data. Tesla’s app is slicker for energy flow visualization and has better smart home integration. For an owner who wants to actually debug their system, Enphase wins. For an owner who wants a clean overview, Tesla wins.
Do these batteries work in winter?
Both do, but both lose usable capacity at low temperatures — physics, not a defect. Expect 10–20% capacity loss at 20°F and more below zero. Both systems have internal heaters that draw from the battery itself to stay warm, which reduces effective capacity on cold days. Install in a sheltered or conditioned space if you’re in a cold climate.
Can these batteries charge electric vehicles?
Tesla Powerwall 3’s 11.5 kW continuous handles full-speed Level 2 charging. A single Enphase 5P can’t — it maxes out around 16A at 240V, half of a typical home Level 2 rate. Two or three 5Ps get you there. Charging an EV from a battery is usually a losing proposition economically (you’re just moving kWh around with conversion losses), but it’s useful during outages.
Which battery qualifies for more incentives?
Both qualify for the 30% federal residential clean energy credit through 2032 under current law. State and local incentive eligibility is identical for both — programs like California SGIP, Massachusetts SMART adders, and various utility rebates don’t care which brand you pick. Always verify current program status before finalizing financials, because these programs change frequently.
Top 10 Solar Panels by ROI
Efficiency, warranty, and payback period compared. Real installation data.
Free. No spam. Unsubscribe anytime.