The Obsolescence Brief Subscribe
Home TeardownsUSB-to-serial bridges

PL2303 replacement: the FT232RNL, CP2102N, CH340G and PL2303GC crosses compared (2026)

USB-UART Bridges · Issue 07
PL2303 EOL versions: blocked on Win 11 PL2303GC: keeps Prolific driver & VID FT232RNL / CP2102N / CH340G: new driver Every cross: a board redesign

The Prolific PL2303 USB-to-serial bridge's end-of-life chip versions (HXA, TA, TB, HXD, RA, SA, EA) no longer load on Windows 11, and the replacement search returns four crosses, the PL2303GC, FT232RNL, CP2102N and CH340G. None is a drop-in: the one fact that decides each swap is the host driver and USB VID/PID, which only the PL2303GC keeps inside Prolific's family. Candidates to validate in your own design, not drop-in equivalents.

01

Status // The part in question

Lifecycle status

Prolific's driver notice marks Windows 11 and later unsupported on the EOL chip versions (and Windows 8, 8.1 and 10 on the oldest HXA/X), enforced by the driver refusing to load: a Code 10 fault and the "DO NOT SUPPORT WINDOWS 11 OR LATER" message. The earliest cutoff traces to counterfeiting. The HXA and X were phased out around 2012 after clones appeared.

The root cause is not a foundry retirement but a vendor decision to drop the older silicon in the current driver, with the G-series as the supported line. So a last-time-buy buys no support time, because the driver block travels with the OS, not your stock. Design in one named cross and budget the requalification.

Current stock, broker availability and any last-time-buy window are dated data for the email, not this page.

02

Candidates // Why these

The candidate set

Four active parts, each with the fit-catch to validate in your own design, plus the older Prolific parts named below to redirect a bad cross:

PL2303GCProlific · same-vendor successorACTIVE
Package SSOP/QFN (G-series)
Driver Prolific family, VID 0x067B
Supply 3.3V VBUS, integrated LDO

Catch: a new chip generation, not a relabel of the HXD. The only cross that keeps Prolific's driver family and VID, but confirm package and pinout against your board; the G-series is not stated to share the H-series footprint.

FT232RNLFTDI · read the generationREDESIGN
Package SSOP-28
Driver FTDI VCP/D2XX, VID 0x0403
Supply 3.3 to 5.25V

Catch: read the generation. The FT232RL is itself end-of-life (LTB 1 June 2024), so the live FTDI part is the FT232RNL, a cross-vendor swap: FTDI's driver and VID 0x0403 replace Prolific's, and SSOP-28 is a different footprint.

CP2102NSilicon Labs · mainstream crossREDESIGN
Package QFN-28
Driver Silicon Labs VCP, VID 0x10C4
Supply 3.0 to 3.6V core

Catch: a board redesign. A mainstream Silicon Labs cross, but a QFN-28 with its own pinout, VCP driver, VID/PID and GPIO scheme; footprint, driver and identifiers all differ, and any GPIO or config must be re-created.

CH340GWCH · the low-cost crossREDESIGN
Package SOP-16
Driver WCH driver, own VID:PID
Supply 5V / 3.3V

Catch: a maker-grade profile. The low-cost option, but the CH340G needs an external 12MHz crystal (the CH340C/N integrate it), the package differs, and the WCH driver and VID/PID replace Prolific's. Weigh traceability against unit cost.

References, named to redirect: the older Prolific parts in the search cluster, the PL-2303HX and PL-2303TA, are not candidates. They are the EOL chip versions this page is about, and crossing to another discontinued Prolific part relocates the problem rather than solving it.

03

Same function // The driver is the tell

Same function, not the same driver

Pinout is not the axis that separates these crosses. No candidate is a pin-for-pin match to the PL2303: the FT232RNL is SSOP-28, the CP2102N is QFN-28, the CH340G is SOP-16, and the PL2303GC is a different chip generation. Every cross is a board redesign, so a per-pin graphic would show no meaningful match on any part; the single load-bearing axis is the host driver and USB VID/PID.

PL2303 EOL reference · no cross keeps the footprint · only the PL2303GC keeps the driver family and VID

PL2303 (HXA/TA/HXD)Prolific · subject
EOL
Package
varies by version
Driver
Prolific
USB VID
0x067B
Windows 11
Blocked
PL2303GCProlific
ACTIVE
Package
G-series (differs)
Driver
Prolific (same)
USB VID
0x067B (same)
Windows 11
Supported
FT232RNLFTDI
ACTIVE
Package
SSOP-28
Driver
FTDI
USB VID
0x0403
Windows 11
FTDI driver
CP2102NSilicon Labs
ACTIVE
Package
QFN-28
Driver
Silicon Labs
USB VID
0x10C4
Windows 11
SiLabs driver
CH340GWCH
ACTIVE
Package
SOP-16
Driver
WCH
USB VID
own VID/PID
Windows 11
WCH driver
Green same as the PL2303 Amber driver and USB identifiers change Red footprint changes (board redesign)
ONE KEEPS THE DRIVEROnly the PL2303GC keeps Prolific's driver and VID (0x067B); the three cross-vendor parts each bring their own.
NO CROSS KEEPS THE FOOTPRINTFT232RNL (SSOP-28), CP2102N (QFN-28), CH340G (SOP-16) and the new-generation PL2303GC all differ, so every cross is a board redesign.
A CROSS-REFERENCE HIDES THE WORKA parametric filter matches on function and data rate, but the deciding axis, the host driver and USB VID/PID, is not a filterable column.

Function is necessary, not sufficient. Anything on the host keyed to Prolific's VID (0x067B) stops matching the moment the silicon changes vendor. Suggested candidates to validate in your own design, not drop-in equivalents.

04

Parametrics // Side by side

Comparison table

PL2303 HXA/TA/HXD (EOL)PL2303GCFT232RNLCP2102NCH340G
VendorProlificProlificFTDISilicon LabsWCH
LifecycleEOL chip versionActive, maker-rec.ActiveActiveActive
Driver / USB VID:PIDProlific 0x067B; Win 11 blockedProlific family, 0x067BFTDI VCP/D2XX, 0x0403SiLabs VCP, 0x10C4WCH, own VID:PID
Pin-compatiblesubjectNew generationNoNoNo
Packagevaries by versionSSOP-28 / QFN-24 / SSOP-16 / QFN-16SSOP-28QFN-28SOP-16
Supply / I/Opublic value to confirm3.3V VBUS, integrated LDO3.3 to 5.25V3.0 to 3.6V core5V / 3.3V
Internal clockpublic value to confirmIntegratedIntegratedIntegratedExternal 12MHz crystal
Windows 11 supportNot on EOL versionsSupportedFTDI driverSiLabs driverWCH driver

The load-bearing row is the host driver and USB VID/PID: the PL2303GC keeps Prolific's, while the FT232RNL, CP2102N and CH340G each change both: the host-software and driver-provisioning work a footprint or data-rate match does not surface. Lifecycle and electrical figures are vendor-reported, not independently audited.

05

FAQ // Frequently asked

Frequently asked questions

Is the PL2303 compatible with Windows 11?

No, not on the EOL chip versions: Prolific's own driver notice states Windows 11 and later are not supported on the PL-2303HXA, TA, TB, HXD, RA, SA and EA versions, and the driver reports a Code 10 fault with the message "DO NOT SUPPORT WINDOWS 11 OR LATER" for those parts. Prolific directs affected designs to the PL-2303GC chip.

Which PL2303 chip versions are EOL?

The end-of-life chip versions Prolific names are the PL-2303HXA, TA, TB, HXD, RA, SA and EA. Prolific labels these the "EOL chip version" in its Windows driver notice and recommends the PL-2303GC as the current-generation part.

What is the PL2303 chipset, and what replaces it?

The PL2303 is a Prolific USB-to-serial (USB-to-UART) bridge that presents a USB device as a virtual COM port to the host. For the EOL versions, the maker-recommended replacement is the PL-2303GC; the cross-vendor options engineers search are the FTDI FT232RNL, the Silicon Labs CP2102N and the WCH CH340G, each a suggested candidate to validate, not a guaranteed drop-in.

What is the difference between the PL2303 and FTDI?

The difference is the driver and USB identifiers: the PL2303 uses Prolific's driver and Vendor ID (0x067B), while the FTDI FT232RNL uses FTDI's VCP/D2XX driver and Vendor ID (0x0403), so host software keyed to Prolific will not enumerate the FTDI part without changes. Note also that the FT232RL many readers name is itself end-of-life (last-time-buy 1 June 2024 passed), so the live FTDI cross is the FT232RNL.

PL2303 vs CP2102 vs FT232, which is good?

All three are USB-to-UART bridges, so the choice is not about the serial function but about the host driver, the USB VID/PID and the package, which differ on each. The CP2102N is the mainstream Silicon Labs cross in QFN-28, the FT232RNL is the live FTDI cross in SSOP-28, and both change the driver and identifiers away from Prolific, so each is a board redesign plus a host-driver change to validate.

Can I replace an FT232RL over a PL2303 on the same board?

Not as a footprint swap: the PL2303 and the FTDI parts are different packages and pinouts, so putting an FTDI bridge where a PL2303 sat is a board redesign, not a like-for-like part change. The FTDI part to design in is the FT232RNL, since the FT232RL is itself end-of-life.

Are there counterfeit PL2303 chips?

Yes, the PL2303 has a documented counterfeit history: legacy-driver documentation records that the PL-2303HXA and PL-2303X were phased out around 2012 after counterfeit copies appeared, and Prolific's Windows driver blocks non-genuine and EOL chips with the Code 10 fault. Buy any remaining genuine stock only through authorized or franchised distribution, and treat a counterfeit or clone as a separate requalification.

Do the driver and USB VID/PID settings carry over when I cross off the PL2303?

Not on a cross-vendor swap: the PL2303 enumerates under Prolific's Vendor ID (0x067B) and driver, so anything on the host keyed to that VID/PID, to a device path built from it, or to the Prolific driver, must be re-pointed. The PL-2303GC keeps the binding inside Prolific's driver family; the FT232RNL, CP2102N and CH340G each bring their own driver and identifiers.

Editorially independent and from public sources only; lifecycle, package, driver and rating data is vendor-reported, not independently audited. Alternates are candidates to validate in your own design, confirm every detail with the manufacturer before committing.

About the author

Hermann Njike is a systems obsolescence manager who leads the cross-functional response when critical components go end-of-life, and is the single point of contact for thousands of production parts.

Editorially independent and from public sources only; lifecycle, pin and rating data is vendor-reported, not independently audited. Alternates are candidates to validate in your own design. Confirm every detail with the manufacturer before committing.

Hermann Njike

Hermann Njike is a systems obsolescence manager who leads the cross-functional response when critical components go end-of-life, and is the single point of contact for thousands of production parts.

Related briefs
The Brief, weekly

One obsolete part, worked out in full.

Read a sample issue →