Author Topic: Real-Time Curve Tracing on Keysight Scope  (Read 765 times)

0 Members and 1 Guest are viewing this topic.

Offline bb1Topic starter

  • Contributor
  • Posts: 44
  • Country: us
Real-Time Curve Tracing on Keysight Scope
« on: April 12, 2022, 12:57:41 pm »
While usefulness of curve tracing is limited, the curve tracing is a lot of fun.
Attached XY_infPersist_hiRes.png shows one of the screenshots of
live traces on Keysight scope DSOX1102G in XY mode.

The curves were obtained using floating scope and grounded two-channel Rigol DG1102Z
waveform generator.

The scope ground was connected to drain of the J113 jfet,
scope Ch1 to the Rigol ground,
and the scope Ch2 to the Rigol Ch2 output providing symmetric triangular signal.
Gate of the J133 was connected to the Rigol Ch1 output providing 8-step signal.

The Waveform Gen settings can be seen on the attached screenshots
rigolGenCh1.png and rigolGenCh2.png.

Approximate schematics of the setup is shown on LTspice jfetCurveTrace_realTime.asc,
where capacitor C1 describes stray capacitance between Keysight and Rigol grounds.
LT simulation results are shown on
LTsimCurves.png

Drain_source voltage and the drain current measurement resistor voltage are
shown as functions of time on
the scope screenshot drain_sourceV_resistorV_vsTime.

Gate voltage (Rigol Ch1) and Rigol Triangular pulses (Ch2 output) are shown
as a function of time on gateV_RigolTriangV_vsTime.png.

In normal XY Keysight mode the traces are bright, but too noisy, as shown on
normalXYcurves.png.

Unfortunately, Keysight XY mode is not compatible with Average mode.
Fortunately, it is compatible with HiRes mode. Unfortunately, in HiRes mode
the traces are dim, as can be seen on HiResXYcurves.png.
Improved image can be obtained by combining HiRes mode with Infinite persistence, as
shown on XY_infPersist_hiRes.png.

More careful examination of the curves shows that each gate voltage curve
actually consists of two slightly separated curves.
This is due to the stray capacitance between the scope and the Wave Gen grounds,
as can be verified by LT simulation changing the value of C1.
The true curve should be average of these two slightly separated curves.

This small separation can be further decreased by reducing Wave Gen frequencies.
 
The following users thanked this post: egonotto


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf