开云体育

Are you terminating the attenuator? Most attenuators are designed to
operate with a 50 Ohm source and load.

-John

==============

Thanks for the response. I thought of the same thing so I tried three
different oscilloscopes:

1) Tektronix 2465 - 300MHz
2) Tektronix TDS3034B - 300MHz
3) HP 54502A - 400MHz

All produced roughly the same results. I even tried turning on the 150MHz
bandwidth limit on the TDS to see if it made any difference but it didn't,
implying that there are not harmonics that high.

-Matthew


--- In hp_agilent_equipment@..., "Dave B" <dave@> wrote:

On 19 Dec 2012 at 13:52, hp_agilent_equipment@... wrote:

______________________________________________________________________
7. Triangle Wave Flatness HP 8116A
Posted by: "medasaro" medasaro@ medasaro
Date: Tue Dec 18, 2012 11:10 pm ((PST))

All -
I am trying to calibrate one of these generators and I can't seem
to make it past step 29 in the calibration procedure. The
procedure reads:

Flatness
23. Set the HP 8116A frequency to 2.99 MHz.
24. Connect the HP 8116A's main output to the oscilloscope via a
6 dB attenuator and set the channel attenuation-factor to 2.
25. Connect the HP 8116A's trigger output to the oscilloscope's
trigger input via a 20 dB attenuator.
26. Set the oscilloscope to 160 mV/div and 100 ns/div.
27. Record the ampltude of the output signal.
28. Set the HP 8116A frequency to 9.99 MHz.
29. Check that the output amplitude has decreased by between 2%
and 4% of the amplitude at 2.99 MHz. If not, you must change
the values of both A1R220 and A1R223 (both must have the
same value). Refer to Table 9-1.


However, my amplitude decreases by more then 5%. I have tried three
different slope ICs (out of various generators I have) and some
produced better flatness then others, but none met spec. I also tried
the same procedure on a different generator (also in need of
calibration) and it too failed, but not as badly. I have tried the
test on two oscilloscopes with the same results. Finally, I tried
increasing the value of R220 and R223, but that, too, didn't seem to
help, at least not enough. Any other ideas?

As a final note, at these high-frequencies the triangle wave gets
rounded-off at the peaks. I have been measuring the amplitude to the
actual peak of the waveform, not where the triangle wave would peak if
it was not rounded. I assume this is correct...

Thanks in advance!
Matthew

Hi.

Cant' help but wonder, what 'scope are you using, and has it enough
realtime bandwidth to not distort (filter some harmonic content of) the
Triangle wave, so that it begins to aproximate a sine shape.

You're (at a guess) going to need a truly flat response on your 'scope
out way past 50 or more MHz I suspect.

Just musing.

Dave B.