Noob uTracer3+ Q: no current displayed when Vg=0

[Bo Diddly]

I'm a uTracer noob. I just completed building a uTracer3+ board, which went very smoothly.  All testing and calibration carried out per manual.  Before wiring the up in an enclosure, I thought I'd test a couple of tubes and get familiar with the GUI.

I tried both a 12AX7 and EL84 with v3.11.  I get decent triode and pentode plate characteristic curve shapes, but for both tubes, when Vg = 0, I just get flat line of Ia = 0, instead of the expected maximum Ia(Va) curve.  If I set the Vg stepping variable to anything smaller than  -0.25V, Ia goes flat line at 0mA. I am running the board off a bench top DC voltage supply, and have the tube socket wired to the board with test leads. 

Any help would be greatly appreciated! a 

[Martin Manning]

How long are the test leads? Oscillations can occur with long leads, which produces odd curves. I have successfully done what you ae trying to do by soldering short  2-3" leads to a tube socket and connecting them directly to the ASGK terminal block. 

[Bo Diddly]

Martin, Yessir,  I will shorten the test leads to minimum, and also increase my DC max current available.  (I’m actually happy to see evidence of oscillation in some pentode grid curves, so I know what to look for in gauging the effectiveness of final lead dress and RF mitigation.) I must work today, but will to get back to the bench tomorrow!

Thank you for your help!

[Martin Manning]

I've seen all manner of weird behaviors resulting from oscillations, and no two have been alike. Adding a snap-on ferrite to the grid or plate lead has cured it for me.

[Bo Diddly]

Issue resolved.  

During initial calibration (build manual section 11.9), fine tuning Vg(set) to Vg(meas) at low grid voltages, I had to adjust the Vg(sat) slider nearly to it's maximum, but still got ~0.2V difference between set and measured.  This seemed to be about the same as the threshold voltage I was seeing to get any Ia, so I reset the calibration slider to a default of 0.96, and now everything is measuring beautifully.  

Thank you Martin for your help!

[Martin Manning]

That's strange, seems like if you were measuring grid voltage near zero with the Vsat slider where it was you should get a Vg1 = ~0 curve. Was that new trace (looking good, BTW) done with the long leads?

[Bo Diddly]

I can’t even pretend to know how the adjustments on the calibration page of the GUI interface with the various controllers on the board, haha.  I’m guessing the Vg(sat) may be equivalent to Vg(set) - but I don’t know if the calibration signals from the GUI are linear or not.  I’m curious to play with this interaction - I don’t know it may have been coincidental that the closest offset I could get at low grid voltages [i.e. Vg of -0.2 to around -0.6V) Vg(set) and V(meas) was around 0.2V - very close to the “threshold” for generating plate conduction at -0.18V.  So having the Vgrid(sat) slider at 1.09 (end of range) moved the zero 0.18V?

[Bo Diddly]

Martin, sorry I forgot to include

Yes, I shortened the test leads to the DUT very similar to your arrangement. (I sort of wanted to see what oscillation behavior looked like so I could recognize it if it occurred in my final enclosure build out, it was very interesting to see if affected not only the smoothness of the curves, but the pentode gain on certain grid voltages but not others)