On Thu, 17 May 2018 16:29:13 -0700, you wrote:
I see that I need to clarify further. I thought the topic would be enough, but I should have been more specific, from the replies this topic has received. I wasn't aware what scope I was referring to would be important. I thought it was a generic enough topic.
I have a Tek 2246A and a Hitachi VC-6025 hybrid analog/digital older scope. They both have the channel 1 and channel 2 invert and add functions, and I just have basic 1x-10x probes for them. I have in the past floated the DUT, and that topic has been beat here in the past I am aware. I really don't fully understand what the add function and invert function can do for me. I have played with them, but probably don't know how to apply the functions, and I thought add/invert could possibly be an alternative to floating the DUT, but with some unknown limitations without a fair amount of experimentation.
OK, going back to basics, the scope has two channels, which may be
added together or run separately. Imagine a circuit with an LED, a
pulse generator sufficient to run the LED, and a series resistor. You
are running the LED in pulse mode, which means that you are putting
MORE current through it than it can tolerate steady state, but not
putting that current through all of the time. It averages out.
Please look up multiplexed LED displays for an explanation.
You'd like to know what the pulse current is for the LED. Even though
you could measure it with a standard DMM, you'd be getting the average
current over time, not the pulse current. Your meter takes that off
time and the on time, averages the current flowing through them, and
gives you a reading. The problem is that even though you are
averaging out to an acceptable value, the peak value may be too much.
(and the average and peak currents are not the same).
So, if you took an oscilloscope with one probe, and measured the
voltage from one end of the resistor to ground, you'd get a reading
and perhaps a waveform (depends on how the LED is hooked up). Now,
you measure that and record the value. Now put the probe on the other
end of the resistor, measure that value and record it. Subtract the
two values to get the voltage drop across the resistor, which is the
peak current, which is what you want to know. (scope probe grounds
are connected to power supply ground).
Your two channel scope can make that nicer and easier.
Put one probe on one end of the resistor, the other probe on the other
end of the resistor. Running in alternate or chopped mode (two
traces), you'd see the two voltage readings at the same time. Well,
it somewhat saves time.
Now, if you added the two voltages (ADD mode), it wouldn't tell you
too much, not in this case. But suppose you subtracted them?
To subtract you add a negative value. In an oscilloscope, you select
ADD (to get a single trace, which is A + B), but you throw the
"invert" switch, (so you add A + (-B)) which works out to (A-B).
Now the reading you see is the true voltage drop across the resistor,
which can be used to tell you exactly what the current through the LED
happens to be.
ADD, and then INVERT.
The difficulty with this is that both readings may be very slightly
different, but both could be at a very high voltage, say 500 volts,
and you're looking to measure that tiny 0.1 volt drop across the
resistor. (it's a thought experiment, so it's not meant to be super
actual). Because the probes and the front end have a voltage limit,
you may not be able to make the measurement since the 500 volts is too
high for the scope.
Some people (and you have to be very careful with this) might put the
UUT (Unit Under Test) to float, connect the ground at say, the 500
volt supply, and then the scope inputs won't see 500 volts, but the
moderate difference.
Unfortunately, this puts the REST of the UUT at 500 volts with respect
to the scope frame, which is NOT a good idea.
Barring the use of special high voltage probes, the only good way to
make this kind of measurement is to use a battery powered scope,
running on battery power, where all the ground parts of the scope are
kept from casual contact. Tektronix did make scopes that were like
that.
These problems are one reason why you have to be careful with not only
what you measure, but how you measure it.
Tektronix did make high voltage probes, so those could be used. Some
will withstand 30,000 volts or a bit more, depending. Differential
measurements *could* be made with those probes, and safely, once all
the safety precautions are observed. You use the UUT ground as
ground, and don't float ANYTHING.
However, the probes are x100 (or X1000) probes, not x10 (divide by
100), so there's a limit of how much signal you get to the scope,
which can limit your measurements.
Hope this helps.
Harvey
So I come here looking for knowledge/experience with add/invert functions, and the application of it that can't be met by other methods.
NielsenTelecom
