On Fri, 25 Jan 2013 18:16:19 -0000, "David C. Partridge"
<david.partridge@...> wrote:
Hi Goran,
Yes they are quite the devil to calibrate, however in this case the front panel position control is maxed to the right and the sweep cal is almost all the way clockwise, and the slewing ramp adjustment (R310) is fully clockwise (into the end stop) as well.
You are not kidding about the calibration but from what I have read on
the list in the past, they are easier to deal with than the older
sampling sweep plug-ins. I am only up to having recalibrated the
trigger circuits on my 7T11A which had just enough DC offset to cause
problems. I have gotten better results when triggering off of pulse
outputs that do not cross zero by adding a DC block inline.
Although it is not strictly necessary, I want to build another SMB to
SMB cable so I can diagnose my 7T11A or 7S11 outside of the mainframe
with both the strobe and internal trigger connected. I only just
recently picked up a working S-1 which should make fixing my broken
S-1 and 7S11/7T11 calibration easier.
Clearly something is amiss here. The annoying thing is this morning it behaved almost right, and then about half way through the afternoon, it failed back to the short sweep (8 divisions, getting shorter as the Time/Div control is roatated clockwise).
I am going to assume that when the sweep is short, the time/div is
still calibrated. If the time/div is proportionally low, then I would
look for a problem in the horizontal output circuit path.
Do you get different results when SCAN is set to MANual or EXTternal
INPUT instead of REPetitive? You might want to try using a function
or ramp generator as a signal source for the external scan input
signal.
I would take a close look at TP636 on schematic 7 to see if the slow
ramp is being reset early. The text says that the ramp (in repetitive
mode) should make it to 10 volts but graph 30 shows only 1 volt. I
think the graph was mislabeled 500mV/div instead of 5V/div when
someone forgot to factor in a x10 probe.
Wow, what a clever current source that is with C620 differentiating
the ramp. Doing it that way removes all non-ideal and non-linear
operation of Q636 although I am not sure doing so is really
neccessary. I do not see how the integrating capacitor C631 is
discharged. Somehow Q626 must be latching on.
In Real Time mode the sweep is full length.
This makes me think that the horizontal output circuit path is ok.
David Hess