Re: Scope input capacitance

Fabio Trevisan
 

Colin,
I have myself spent the whole last year doing some "poor man's" calibration
fixtures to restore and calibrate a Kenwood 60MHz oscilloscope (my previous
project that is almost concluded...).
To calibrate the Vertical stage of this scope (for square wave response), I
built a very small square wave generator using an SMD 74AC14 mounted on a
double sided blank circuit board, everything with very, very short
connections... The board has not more than 2 by 2 inches.
This is a classic schmitt trigger oscillator.
OK... Besides having an output connector, I added a BNC connector (without
cable, just the plug, meant to be hooked directly to the scope's input)
connected to the point where the capacitor of the schmitt trigger
oscillator goes.
In this circuit, I can remove the capacitor altogether so to have nothing
else connected, than the BNC to the scope. (so the only capacitor that the
oscilator "sees" is the capacitance of the input of the oscilloscope and,
of course, the stray capacitance of the 74AC14 input).

Then, I select the scope's attenuator for the input that DOESN'T have any
attenuation and measure the frequency the oscillator generates...
Eventually changing the resistor of the oscillator so that the frequency is
high enough (in order to be sensitive to the small capacitance differences
you will measure) but still in the reach of my DMM (4MHz).

Then, all I needed to do was to switch ranges in the oscilloscope and
adjust the input trimmers to match that same frequency.

Of course that this in not "perfect"...but it gets to the most important
point in adjusting the input capacitance, which is
is to make sure that it's the same throughout all attenuator ranges, and
the only reason for that is to make sure that the probe compensation
adjustment doesn't change when you change ranges...

One could be perfectionist and wish the oscilloscope to have "exactly"
20pF... (or 15, depending on the model) but for what?
The input capacitance of any scope - in fact - is just an unavoidable evil.
Would it be possible, it should rather be 0pF.
Of course that you don't want your input capacitance to be so "OFF" that it
falls out of the range of compensation of the usual probes... but the
scopes don't even provide for such big variation.

Ohhhh... in time... if you DO HAVE a square wave generator capable of
generating at all levels to cover all ranges of the oscilloscope, from say
300mVpp to 50Vpp, take any 10x probe, put the generator at 300mVpp, set the
scope to 5mV/div range (30mV will be 6 div)... and adjust the probe
compensation for proper square wave.
Then, increase the range of the scope, as you increase the output of the
generator, and at each range, don't touch anymore the probe's compensation,
and adjust the compensation on the scope's respective range attenuator...
When you switch through all voltage ranges without having to re-adjust your
probe's compensation trimmer, then your scope's input capacitance is good.

Rgrds,

Fabio









2016-09-29 18:35 GMT-03:00 colingherbert@... [TekScopes] <
TekScopes@...>:



Hi,
According to the datasheet for the 067-0538-00 Calibration fixture (input
normalizer 20pF), it is possible to measure the input capacitance of a
scope channel with a Tektronix type 130 L-C meter. Does anyone have any
idea of how this might be achieved and if any other L-C meter could be
used?
I have tried just connecting the test-leads of two or three of my L-C
meters across the input BNCs while the scope is powered up to no avail. I
am puzzled.
Colin.

[Non-text portions of this message have been removed]



[Non-text portions of this message have been removed]

Join TekScopes@groups.io to automatically receive all group messages.