Skip to content

Latest commit

 

History

History
106 lines (87 loc) · 5.89 KB

README.md

File metadata and controls

106 lines (87 loc) · 5.89 KB
Raw

owowon - A small oscilloscope UI for the Owon HDS series portable oscilloscopes

Screenshot of the program, reading a 10 MHz sine wave generated by the HDS2102S itself:

Screenshot of the program, reading a 10 MHz sine wave generated by the device itself

Videos of a couple of actions on the calibration square wave (everything is synchronized with the oscilloscope):

vid1.mp4
vid2.mp4
vid3.mp4

Windows only

I mostly created this for myself, putting on the reverse engineering hat to figure out how to talk to the device, giving egui a spin UI-wise, and checking out the WinUSB support of the Windows Runtime. The USB communication is thus tied to Windows right now, but should be able to be ported without too much hassle. The rest of the program is platform agnostic, to the extent that egui is.

Supported devices

So far, this program has only been tested with the HDS2102S oscilloscope, with firmware version 1.5.1. I assume it's going to work mostly fine with other oscilloscopes from the same model range, though more or fewer bugs than what's listed here might be present. Feel free to create issues or pull requests if you like to expand the support.

Installation

The official Owon software uses libusb as driver, but since I wanted to give WinRT/WinUSB a spin, I "converted" it into a WinUSB device using Zadig. The corresponding driver is provided in the winusb driver folder in this repository, but you can generate your own through that tool if you like. You'll need it for the program to detect the oscilloscope.

Features

  • Much, much faster data acquisition than the Owon tool (up to ~100 readings per second, depending on the number of active channels and whether measurements are enabled or not), and allows changing settings while acquiring.
  • More measurements (see limitations below)
  • Almost everything needed for everyday usage is controllable through the UI (image of the sidebar UI):
    • Auto function
    • Channel enablement
    • Timebase, Horizontal offset
    • Acquisition mode, Sample depth
    • Per channel: Vertical offset, scale, probe attenuation, coupling
    • Trigger: Source, level (see limitations below), mode, triggering edge, coupling
    • Measurements
    • Function generator (see limitations below)

Note: DMM readout/control support is technically possible, but currently not implemented.

Keyboard Shortcuts

  • Toggle measurements: M
  • Horizontal zoom in/out (time base): PageUp/PageDown or mouse wheel up/down
    • with Ctrl: vertical zoom in/out (voltage scale). This affects Ch1 (or Ch2 if Ch1 is disabled). Press Alt to force Ch2 (if enabled).
    • with Alt: move trigger level up/down
      • with Shift (only with keyboard): bigger steps
  • Horizontal offset: ArrowLeft/ArrowRight
    • with Shift: bigger steps
    • with Ctrl: smaller steps
  • Channel vertical offset. ArrowUp/ArrowDown. This affects Ch1 (or Ch2 if Ch1 is disabled). Press Alt to force Ch2 (if enabled)
    • with Shift: bigger steps
    • with Ctrl: smaller steps

Limitations

  • Switching to DMM/Function generator mode on the device itself will permanently slow down the acquisition of data by a factor of ~2-4. Only switching back to the oscilloscope view and then restarting the device restores the full performance.
    • You can also hit the play/pause button to hold the current DMM value, which temporarily returns it to full speed. However, it seems that the hold is automatically released when switching modes back to function generator/oscilloscope view. Since there doesn't seem to be a way to disable the DMM entirely through the USB interface, the only way to get back to full speed is to restart the device entirely.
  • Opening any menu or enabling drawing measurements on the device itself will temporarily slow down acquisition. Closing the menu or disabling the measurements, respectively, will restore the full acquisition speed.
  • Only the trigger modes Auto and Single work properly. Normal fails to transmit (and render on the oscilloscope screen) the signal data after entering trigger status and leaving it again. It seems that the transition between Trig and Ready causes the device to misbehave.
  • There is no software way to re-arm the trigger in Single mode, you have to press the play/pause button on the device.
  • Setting negative trigger levels will cause it to apply incorrectly, but reading it out will return an overflowed value (4295V, probably some sign issue with u32 - 4,294,967,295), so you'll have to manually type the value in, and the trigger level visualization will not work. When setting the trigger level through the device itself, it reads out correctly.
  • Most measurements only reliably work inside the area that matches the screen, not the full measured area.
    • I wouldn't necessarily trust the measurements that aren't displayed on the device itself.
  • Function generator
    • The :FUNCtion? call to get the currently active function never returns AttALT, StairDn, StairUD, StairUp, Besselj, Bessely, Sinc and instead returns AmpALT for these. Setting these works, however.
    • :FUNCtion:FREQuency? returns values mis-scaled by a factor of 1 million (= effectively in micro-Hertz), worked around in software
    • :FUNCtion:AMPLitude?/:FUNCtion:OFFSet? return values mis-scaled by a factor of 1000 (= effectively in Millivolts), worked around in software
    • I haven't bothered adding support for the parameters of all function generator modes so far.
  • The DMM (multimeter) is also controllable through USB (undocumented), but not implemented yet.