EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: lmarussig on November 07, 2016, 06:01:47 pm
-
Hi everybody,
I have recently bought a second hand Hameg 203-6 oscilloscope on ebay. I've been playing with it for a while and have discovered that it has a couple of faulty bits and bobs (such as the invert function, but I'm going to move these questions over to the repairs section).
One thing I am really not sure about, whether it is a fault of this poor puppy or a poor understanding of mine, is the visualization of a signal in the 5mV/div range.
If I generate a 14mV p/p waveform using my function gen (Instek 8215A) and connect this straight to the oscilloscope (on either channel) via BNC, I can see a clear representation of the signal.
On the other hand, if I measure the signal using a probe (using either channel) at the input of a non-inverting amplifier circuit of mine, I get two functions in anti-phase on the same channel. The same happens if I test the signal, always using the probe, at the BNC output of the function generator.
If I switch to at 10mV/div the anti-phase waveform loses intensity, at 20mV/div it dimly flickers and at 50mV /div it disappears. I am truly puzzled.
I've attached two photos for you to have a look at it. Can anybody help me understand why is this so?
-
I've attached two photos for you to have a look at it. Can anybody help me understand why is this so?
missing the two photos...
-
Sorry, newbie problem with the upload! Here they are
-
probably the trigger system cannot handle such small signals anymore, when you start attenuating it with the vertical scale knob. the signal should be at least 0.5-1 div on the screen, depending on the instrument.
-
What I don't understand is why this behaviour appears to occur only when measuring with a probe. If I connect the signal generator straight into the oscilloscope's input, the problem vanishes.
-
are the probes correctly terminated/adjusted?
What is the signal frequeny (can't see the time per div setting on the photos)?
-
What I don't understand is why this behaviour appears to occur only when measuring with a probe. If I connect the signal generator straight into the oscilloscope's input, the problem vanishes.
probably 1:10 probes? They make the signal ten times smaller for the scope.
-
I have a switchable 1:1/1:10 probe. I've done these tests at 1:1, I've double checked after your suggestion but nada.
However I have tested the function generator directly to the oscilloscope again. It seems that the 5mV/div range works correctly at a 1msec timebase. If I lower the timebase (also just by one point) that flickering anti-phase waveform appears. Any clues?
-
try to adjust the trigger level.
Is the x-magnifier knob pressed?
(should be in unpressed state for normal operation).
With best regards
Andreas
-
when you use your probe, you are likely to pick up high frequency noise "over the air", as opposed to using a fully shielded coax connection. At these low signal levels it can then become problematic to obtain a stable picture, especially when using an analog scope. Does your model have a "HF reject" trigger mode? You could try that and see if it makes a difference. I would not be worried too much that your scope may be broken, you are operating it close to its limits.
-
The magnifier button is out, I'm in auto triggering mode feeding a 1kHz sinewave. The probe seems much more susceptible to this problem (it shows this flickery antiphase up to 50mV/div at all timebases below 1msec), while the other method (function gen direct into oscilloscope) gives problems at all timebases at 5mV/div. Above that vertical division no prob.
-
Alright! So that's already a big clue, I see why the probe might be much more susceptible to this issue. It's just so weird that I get a perfect anti-phase signal on that same channel.. As for the HF rejection, this hasn't got any option
-
Alright! So that's already a big clue, I see why the probe might be much more susceptible to this issue. It's just so weird that I get a perfect anti-phase signal on that same channel.. As for the HF rejection, this hasn't got any option
The anti-phase signal is occuring because of the noise. Picture some random noise wiggling up and down on your main signal. You've got a trigger threshold going through this at some level. Let's say you've got the trigger set to "rising" or "+". Sometimes the random noise on your falling edge of your signal will cross the threshold in a rising fashion, even though your signal is falling. Thus, it will trigger a trace on the falling edge at the same "spot" as it normally does on the rising edge. These two triggered waveforms overlap and you see "both".
Here is a video I did that shows this affect:
https://www.youtube.com/watch?v=O99OTpSCEnM (https://www.youtube.com/watch?v=O99OTpSCEnM)
-
That's good to know!
-
Like the others say, the oscilloscope is triggering on the high frequency content. It is possible that the trigger need to be calibrated.
When you use the unterminated coaxial cable at a low frequency, it looks like a capacitive load to the signal generator with about 25 picofarads per foot which when combined with the 50 ohm output, filters high frequency noise. The coaxial connection also reduces external noise. If you added a 50 ohm feedthrough termination at the oscilloscope end, the signal level would be halved but full bandwidth would also be restored.
Oscilloscope probes have lower capacitance yielding higher bandwidth allowing more high frequency noise and if you do not use a coaxial connection at the probe tip, they will also pick up more ambient noise.
DDS signal generators may have significant high frequency ripple which can cause this problem. Dave made a video showing an example of this but I was not able to find it.
-
if your scope does not have any bandwidth limiting capabilities, you can try and clip some ferrites on the probe cable. but be cautious, the RF is still there ;)
-
Thank you everybody, I would have never had a an understanding of this without asking for any help. Now I feel reassured that it isn't the totally the scopes fault.
Considering the various explanations received, I've also gone through the manual again. I have found out that there is actually a LF and HF rejection filter, that there is a hold-off parameter which can help with low voltage noisy signal analysis and that there are a few paragraphs which relate to the hints found in this thread.