HP 1743A Oscilloscope Pulls Trace Off-Screen Only in A vs. B / X-Y Mode

[techie1234]

I recently purchased a used HP 1743A oscilloscope - voltages are nominal and everything seems to work except the delayed sweep and X-Y mode (or “A vs B” mode in HP nomenclature, although Channel A controls the Y axis and B the X axis).  The latter is what I’m tackling first, and what I’m seeing has me confused.

In summary, with signal applied to both channels and nominal V/div on the horizontal (“B”) and vertical (“A”) channel the resulting trace is completely off the screen to the right.  Adjusting the horizontal position “full left” doesn’t pull it back on screen.  However, if I set the horizontal V/div to 1/10 of the nominally required setting to make the trace “wide” and increase the input horizontal voltage sufficiently, I can get a partial trace back onto the screen.

For example, I set up a simple equal voltage 1 Hz Lissajous pattern (a rotating circle).  At less than 2.6V P-P applied to both channels, the trace is completely off the screen.  Increasing the horizontal voltage up through 5.2V P-P grows the trace onto the screen as seen in the first picture below.  It rotates as you would expect and is the correct height, albeit quite wide due to the much lower horizontal V/div.  Then, as I increase the voltage past 5.2V P-P the left side of the trace starts to “cave in” back to the right as seen in the second picture below.  This last behavior really has me scratching my head.

I’ve been studying the Service Manual and in Service Sheet 12 on the horizontal pre-amp and amp, for troubleshooting HP suggests putting the scope (conveniently) into A vs. B mode.  If there is horizontal deflection (which there is, albeit too much to the right) HP indicates the problem is in the time base.  However, if the problem is in the time base, why would operating the scope in normal (i.e., non- A vs. B) mode not manifest any problems?

I’m still early in my electronics learning curve – so any pointers or suggestions would be much appreciated!

[scopeman]

It's been quite a while since I worked on one of this series but I would look for resistors that have drifted high in value in the horizontal circuits as this is a typical problem with these scopes. I would also look for leaky capacitors as a second choice in the horizontal amp.

If you use the calibrator signal in Y-T mode is the timing correct and linear throughout several time base ranges?

Sam

W3OHM

[techie1234]

It's been quite a while since I worked on one of this series but I would look for resistors that have drifted high in value in the horizontal circuits as this is a typical problem with these scopes. I would also look for leaky capacitors as a second choice in the horizontal amp.

If you use the calibrator signal in Y-T mode is the timing correct and linear throughout several time base ranges?

Using the on-board 1.4KHz, 1V P-P calibrator in Y vs. time (i.e., normal sweep) mode the horizontal and vertical measurements are all as expected across multiple time bases.  That's what has me stumped - one would think that any horizontal positioning issues would manifest themselves in both A vs B and normal mode, but that's not the case.  I'll start checking the various resistors in situ first to see if anything on the horizontal boards seems to have drifted.

This got me thinking a bit more deeply about X-Y mode.  Since the timebase selector is disabled in X-Y mode, what is the sweep frequency that is used to graph the X vs. Y relationship?

As a side note: the display is responsive to the A vs B cal adjustment trimpot.

[jdragoset]

This got me thinking a bit more deeply about X-Y mode.  Since the timebase selector is disabled in X-Y mode, what is the sweep frequency that is used to graph the X vs. Y relationship?



Whatever the signal fed to x is the "sweep" frequency.
The cal square wave fed to x and y set at same gain will display a faint 45 degree line (square wave hi/lo transition) with a bright spot at each end (hi/lo duration)