F1080 vs i3200: The Component That Decides Everything Else
Every desktop UV printer on the market runs an Epson piezo printhead, and there are only two of them worth talking about. One is the head out of a home photo printer. The other is an industrial part. Which one a machine uses determines whether it prints decoration or objects — and no software update moves a machine across that line.
Figures from Epson's I3200(8) datasheet and supplier spec sheets; machine-context rows from our own research and bench work. Where a number comes from a supplier rather than Epson, it is called out.
Full side-by-side
The viscosity row is the one that matters. Everything else is degree; that row is kind.
| Spec | F1080 (XP600 / “DX11”) | i3200-U1 |
|---|---|---|
| Class | Consumer-derived — the head from Epson's XP-600 home photo printers | Industrial PrecisionCore MicroTFP, UV variant |
| Nozzles | 1,080 total | 3,200 total (8 rows × 400) |
| Channels | 6 | 8 |
| Nozzles per colour | ~180 | 400 (white on two channels = 800) |
| Ink viscosity window | 3–5 cP (thin, water-like) | 5–7 cP at the nozzle — and built to run heated, so 10–15 cP bottle resin thins into window |
| Firing — grayscale (how real prints run) | 11.52 kHz | 21.6 kHz (1.9×) |
| Firing — binary single-drop | ~46 kHz | 43.2 kHz — roughly equal; the gap only appears in grayscale mode |
| Droplet sizes | 4.5 / 9 / 20 / 40 pL | 3.8 / 6.2 / 9.3 pL four-level · 5 pL binary |
| Physical | Compact consumer module | 69.1 × 59.4 × 35.6 mm · 82 g |
| Replacement price (HeyGears) | $299 | $1,299 |
| Rated / estimated life | 3–6 months (Morpho estimate) | ~2,100 h ≈ 12–24 months (HeyGears; Morpho similar) |
| Head warranty where published | 3 months (xTool, eufyMake) | 180 days (HeyGears) |
| Who uses it | xTool O1 Omni · eufyMake E1 · Morpho Standard · OMTech (×2) · HeyGears G1 | Morpho Pro · HeyGears G1X |
| What it can therefore do | Thin decorative UV ink only — 2D and shallow texture | Heated high-viscosity resin, so structural 3D parts; double-white texture speed |
Why the kHz number is the speed limit
Dots have to land at a fixed spacing set by the print resolution, and the head fires on linear-encoder ticks as the carriage sweeps — Morpho's engineers put it plainly on camera: the encoder is the firing clock. So a firing rate converts directly into a maximum carriage speed.
max carriage speed = firing rate × dot spacing
720 dpi → a drop every 35.3 µm · 1440 dpi → a drop every 17.6 µm
| Head · mode | Firing rate | Max speed @ 720 dpi | Max speed @ 1440 dpi | Reality check |
|---|---|---|---|---|
| F1080 grayscale | 11.52 kHz | ≈ 407 mm/s | ≈ 203 mm/s | xTool O1 measured carriage: 400 mm/s — the head is the wall |
| i3200 grayscale | 21.6 kHz | ≈ 762 mm/s | ≈ 381 mm/s | Morpho Pro runs 600 mm/s — with headroom |
11.52 kHz × 35.3 µm works out to about 407 mm/s. The O1's measured carriage speed is 400. That is not a coincidence and it is not a design choice — it is the head's firing rate hitting the wall. The datasheet's 1.9× grayscale gap is the same 1.5–2× Morpho claims on camera, and it flows through to the 2.1× we measured on large prints in the bench comparison. Physics, datasheet and stopwatch all agree.
Grayscale is the honest number to quote. Real prints fire several drop sizes per nozzle; the ~46 kHz binary figure is a single-drop-size mode nobody prints photographs in, and in that mode the two heads are roughly equal.
Pressure is a slope, not a voltage
The pressure pulse that ejects a drop comes from how fastthe chamber wall moves, not how far. Waveform slope is pressure: a slow flex just oozes fluid backward, because the ink has time to get out of the way; a fast flex spikes the pressure because it does not. How far the wall moves sets drop volume. And the crystal's force determines whether it can hold the commanded speed against thick fluid pushing back — which is what “industrial head” actually means.
One correction worth making explicitly, because our own earlier notes had it wrong: do not say the i3200 runs higher drive voltages than the F1080. No Epson-published drive-voltage figure for either head could be found, and thin-film piezo generally trends toward lower drive voltage than bulk piezo, so the claim may be backwards. Say the i3200 delivers a stronger pressure pulse — the published viscosity and frequency specs demonstrate that, and they carry the argument on their own.
A second correction, from Epson's own I-series spec table: the i3200 lists “Heater: Not available” and “Ink Recirculation: Not available,” and Epson publishes no viscosity range at all. The heat in these machines is applied to the ink path and reservoir by the machine builder. The head does not warm the ink; the machine does, so that it arrives in-window at the nozzle. The cP figures everyone quotes come from supplier sheets, not Epson. Why that matters physically is the honey rule.
What a head actually costs
Listings retrieved August 2026. Individual prices are vendor listings — claimed. The bands are solid, and the bands are the point.
The open world: Epson
| i3200-A1 (aqueous), Alibaba factory listing | $840 (1–9 pcs) |
| i3200-A1, US distributors | $889–$1,198 |
| i3200-U1 (UV), US distributors | ~$1,150–$1,180 |
| i3200-E1 (solvent) | $1,408–$1,925 |
| DX5 / F186000, the salvage-era head, still listed today | $825–$1,450 |
The closed world: everyone else
| Ricoh Gen5 MH5440 | $2,377–$2,596 |
| Konica Minolta KM1024i (LHE / MHE-D / MAE-C) | $2,650–$2,750 |
| Fujifilm Dimatix StarFire SG1024/SA-2C | $2,738 |
| Fujifilm Dimatix StarFire SG1024/LA and /MC | $4,300–$5,150 |
| Fujifilm Dimatix Samba | No public price found anywhere |
The Samba result is itself the finding. Fujifilm's own StarFire pages carry full specifications and no pricing and no purchase path; Samba is contract-only and nothing citable exists. The i3200 has an eBay listing. The Samba has a sales representative. That is the difference between the two worlds in one data point — and the reason every machine in this category is Epson-shaped is that the distribution model is Epson-shaped, not that the physics demands it.
Two things this corrects. First: the “$400–700 from Chinese suppliers” figure that circulates does not hold up — the firmest China-direct number is $840. Second, and more usefully: HeyGears sells the U1HD replacement head at $1,299 with a 180-day warranty, against a street price of roughly $1,150–$1,180 for a bare i3200-U1. That is essentially cost plus handling, not a markup. We had implied otherwise, and correcting it makes the rest of the cost criticism more credible, not less.
The trick, stated plainly
The cheap head is cheap partly because it is unheated, non-recirculating and spec-thin. Fujifilm publishes for the StarFire everything Epson does not: 8–20 cP with an optimum of 10–14, integrated temperature control to 50 °C, recirculation, a metal nozzle plate. The i3200 is a commodity aqueous and eco-solvent head being pushed into a job the expensive heads were designed for, with the machine builder supplying externally what a $3,000 head does internally. That is not a scandal. It is the whole trick, and it is why these machines exist at all.
Which is also why the printhead dates sit on the patent timeline: Epson closing external sales around 2008, reopening in 2019, and naming the i3200 in its OEM catalogue in May 2021 is the second of the two curves that had to cross.
More from this research
- UV printer research hub
- The interactive patent timeline
- How UV printing actually works
- Six desktop UV printers compared
Sources: Epson I3200(8) series datasheet (inkjet-solution.epson.com, 2023-02-08); Epson I-series product pages; DigiPrint Supplies, AllPrintheads, zzhstar and eBay listings retrieved August 2026; Fujifilm StarFire SG1024 product page; HeyGears and Morpho campaign material and our own bench testing.
