# LimeSDR mini v2.2 in Windows 11: No Samples or timeout

**URL:** <https://discourse.myriadrf.org/t/limesdr-mini-v2-2-in-windows-11-no-samples-or-timeout/13356>\
**Category:** LimeSDR\
**Created:** [23 October 2026 22:32 UTC](https://discourse.myriadrf.org/t/limesdr-mini-v2-2-in-windows-11-no-samples-or-timeout/13356 "2026-10-23T22:32:39Z")\
**Posts on this page:** 1\
**Showing post:** 8

<div class="post-metadata">

**Author:** ![ricardas](https://yyz2.discourse-cdn.com/flex034/user_avatar/discourse.myriadrf.org/ricardas/32/86_2.png) [@ricardas](https://discourse.myriadrf.org/u/ricardas)\
**Post date:** [3 November 2026 15:58 UTC](https://discourse.myriadrf.org/t/limesdr-mini-v2-2-in-windows-11-no-samples-or-timeout/13356/8 "2026-11-03T15:58:55Z")

</div>

Hi, just FYI limesuiteng has been updated to allow select which FTDI driver to use:

> <https://github.com/myriadrf/LimeSuiteNG/commit/3978812860dfa444366bdac633a78521fb63ad49>

And there has been some fixes regarding the WinUSB driver:

> <https://github.com/myriadrf/LimeSuiteNG/pull/417>
>
> Since 39788128 Windows 11 builds link \`WU\_FTD3XX\` by default. With FTDI's WinUSB…-based D3XX driver/library (1.4.0.1) a LimeSDR Mini is enumerated, but every control read fails and RX streaming produces no samples:
> 
> \`\`\`
> limeDevice --full
> LimeSDR-Mini: reference clock detection failed, assuming 40 MHz
> Firmware version	: 0
> Gateware version	: 0
> Gateware target board	: UNKNOWN
> \`\`\`
> 
> \## Root causes
> 
> Verified with a standalone test program against \`FTD3XXWU.dll\` (same handle, same 64-byte GET\_INFO packet):
> 
> 1. \*\*Stream mode on the control pipes.\*\* After \`FT\_SetStreamPipe(h, FALSE, FALSE, 0x82, 64)\` every \`FT\_ReadPipe\`/\`FT\_ReadPipeEx\` on 0x82 returns \`FT\_OTHER\_ERROR\` (32). Without stream mode the 64-byte reply is read correctly. The stream pipes (4096) are not affected and keep stream mode (needed for packet alignment).
> 2. \*\*Synchronous completion.\*\* \`BulkTransfer\` and \`BeginDataXfer\` treated every status other than \`FT\_IO\_PENDING\` as an error; the WinUSB library can complete immediately with \`FT\_OK\`.
> 3. \*\*Auto-reset events.\*\* \`FT\_InitializeOverlapped\` in the WU library creates auto-reset events. \`USBDMAEmulation\` waits on the same transfer twice (\`Wait()\` and then \`UpdateProducerStates()\` with timeout 0), so the first wait consumes the signal and the transfer is never finished. \`WaitForXfer\` now checks \`HasOverlappedIoCompleted()\` before and after waiting.
> 4. \*\*Stale control replies on connect.\*\* \`FT\_FlushPipe\` is not supported by the WU library (\`FT\_NOT\_SUPPORTED\`). A reply left in the control IN pipe by a previous session was parsed as the answer to the first command and could crash device detection (access violation in \`make()\` on every second start). The control IN pipe is now drained with a short timeout on connect (the libusb path already flushes it).
> 
> (1) is guarded by \`LIMESUITENG\_FTDI\_WINUSB\`, defined for \`WU\_FTD3XX\` builds, so the legacy \`FTD3XX\` (WDF) path keeps its current behaviour. (2)–(4) are harmless for both libraries.
> 
> \## Testing
> 
> \- LimeSDR Mini v1 (HW 2, gateware 1.30, firmware 6), FTDI WinUSB D3XX 1.4.0.1, Windows 11
> \- x64 build: \`develop\` → all zeros (above); with this PR → correct device info, RX 10 MS/s at 30 MB/s without overruns, 3/3 reconnects. The x64 binaries were run under Windows-on-ARM x64 emulation (x64 user-mode \`FTD3XXWU.dll\` on top of WinUSB); native x64 hardware not tested.
> \- ARM64 build (with the follow-up ARM64 PR): RX up to 30.72 MS/s at 92.5 MB/s, no overruns; SoapySDR plugin streaming for hours; 4/4 consecutive reconnects (previously every other one crashed).
> \- Legacy \`FTD3XX\` configuration still compiles (x64).
> \- Not tested: TX, LimeSDR Mini v2, the legacy WDF driver at runtime.
> 
> Possibly related: #377 (overlapped handling in FT601).

---

_[View the full topic](https://discourse.myriadrf.org/t/limesdr-mini-v2-2-in-windows-11-no-samples-or-timeout/13356)._
