Strange Behavior on Analog Oscilloscope (Tektronix 2225)

[wmundstock]

Hi All.

This is my first post in this forum. I hope I got the best section for my question.

I already have a digital oscilloscope but decided to buy an old analog Tektronix 2225 for learning. I'm having lots of fun playing around with it, but given I never used an analog oscilloscope before, I started wondering about a behavior I have noticed on this particular scope.
The signal used for the testing was a square wave of 1Khz with 500mV peek to peek.
If I measure with the probe in 10x with the vertical scale at 2v/div 10x zoom off, I get a reasonable, crisp trace like the picture below:



So far so good, but if I activate 10x zoom I can see it much better given its a small signal for 10x.



Now if I go back to 10x, and reduce the vertical scale to 0.5v/div, it seems like the trace start getting blurry and hardr to read. See below:



With 0.1V/div its simply unreadeable, and that should be the proper scale for probe compensation adjustment. See below, its absolutely useless.



If I go back to 10x mag with 0.1v/div then its back to acceptable.



If I try a sine wave things get a bit better, but not very much.



I feel like there is something wrong with the scope calibration or something. Given I dont have experience with analogs I dont know what to expect here. Am I doing something wrong? Should this scope be able to display the 500mV signal with 10x probe without 10x magnification?
Below is a print from the same wave on my digital oscilloscope for validation.



Here is a small video of this behavior:


Thanks a lot by the help!
Walter

[srb1954]

You should definitely get a sharper trace than what you are seeing. Your trace brightness is a little high in some of the pictures. For best sharpness it is important to set the brightness no higher than necessary to view the trace. You may also need to adjust the focus control for the sharpest trace as you adjust the brightness.

However, there seems to be a lot of noise riding on top of your signal, which may be causing the fuzziness. This noise could be generated internally to the 2225 although the digital scope trace also seems quite noisy, which suggests that the signal source is at fault or the noise is being picked up from an external source.

Diagnostic steps:
(1) Check to see if the same result is obtained on channel 2. (be sure to adjust the probe compensation if you swap the probe to the other channel)
(2) Check with another probe as there may be a fault in the probe allowing excessive noise onto the signal. The earth leads on scope probes frequently break resulting in excessive noise pickup on traces.
(3) Try another signal source.
(4) Check that the scope is connected to a properly grounded power outlet.
(5) Switch off all other electronic devices in the vicinity as these may be causing radio-frequency interference.

[wmundstock]

Thanks for the suggestions. Unfortunately I already tried all of that.

(1) Check to see if the same result is obtained on channel 2. (be sure to adjust the probe compensation if you swap the probe to the other channel)
- Same results are obtained on both channels.

(2) Check with another probe as there may be a fault in the probe allowing excessive noise onto the signal. The earth leads on scope probes frequently break resulting in excessive noise pickup on traces.
I have 4 probes, 2 are almost like new and 2 bought yesterday. Tried them all with same results.

(3) Try another signal source.
I tried, with both the internal source and external signal generator. External generator seems a bit cleaner than the internal one.

(4) Check that the scope is connected to a properly grounded power outlet.
No ground outlet here. I live on an apartment which does not have it.

(5) Switch off all other electronic devices in the vicinity as these may be causing radio-frequency interference.
I tried going to other locations, other outlets, same results.

Other things that I already tried:
- Focus control (with base signal and with end result signal) no significant change.

All CRT Settings from service manual
- Astigmatism settings - seems spot on.
- CRT Grid Bias, shouldn't really affect anything here as it has more to do with the brightness.
- Trace alignment - no relation.
- Trace Geometry, I did try changing one or two clicks left and right juuuust to make sure, but no impact either.

I reviewed some of the vertical settings, the signal seem to be spot on with regards to the scale, its just that as I get closer to smaller scales it starts blurring. One setting seems. VErtical ajustments have a few more things that maybe could have some relation, but i dont know exactly what they are so thought i shouldn't mess around. They are: Vertical Gain, Deflection Accuracy, and some others that dont seem as related.

[rstofer]

Can you decrease the time/division to look at the higher frequency components?  50 ns/div is as fast as it goes if I am reading the manual properly.  In the most expanded photo, it sure looks like there is a lot of noise riding on that signal.

There is also some capability to magnify the display but I didn't bother reading about that feature

https://www.tek.com/manual/2225-operators-manual

Don't change any control that doesn't have a knob.  You are in no position to re-calibrate the scope and, odds are, it's the signal source, not the scope.

[srb1954]

(4) Check that the scope is connected to a properly grounded power outlet.
No ground outlet here. I live on an apartment which does not have it.


I reviewed some of the vertical settings, the signal seem to be spot on with regards to the scale, its just that as I get closer to smaller scales it starts blurring. One setting seems. VErtical ajustments have a few more things that maybe could have some relation, but i dont know exactly what they are so thought i shouldn't mess around. They are: Vertical Gain, Deflection Accuracy, and some others that dont seem as related.

The lack of a ground connection is a problem. Normally a scope and other electronic instruments would be connected to a ground through the mains power cord both for safety reasons and for minimisation of noise pick-up. In the absence of a grounded power outlet in your apartment the best you can do is power all of your instruments from a single power socket strip (with 3 pin sockets) and using 3-core mains cables to each instrument. That way all of the instruments will at least share a common ground connection between themselves and this will help somewhat in minimising noise pick-up.

An additional test to verify whether the problem is in the scope would be to remove the probes, switch the input coupling to GND and select BOTH channels and ALT for the vertical mode. Select PP Auto triggering and adjust the timebase and triggering controls as required to get a steady trace for both channels. The traces should be flat, thin and clean and they shouldn't vary much as the volts/div controls are switched through their ranges or the vertical position is shifted up and down on the screen. Engaging the X10 vertical magnifier may result in small jump in the traces and a slight thickening of the trace but it should be no more than the height between the minor divisions on the graticule scale.

[bob91343]

You are picking up some noise.  Perhaps poor grounding or a strong field in the vicinity.  Check the integrity of the grounding system.  Do you get the wide noisy trace with the calibrator shut off?  With the input shorted?  Something is definitely wrong.

[wmundstock]

Hi All

First of all thank you all for spending the time to help me! After the feedbacks around the noise, I decided to take some measurements, did comparisons and checked manuals.

Back to my original question, my goal is to determine whether or not there was something wrong with the analog oscilloscope and the way it was plotting things with the probe set to 10x attenuator.

I think the answer is: there is absolutely nothing wrong with the oscilloscope and the way its displaying the wave. In fact it is showing a very slightly clearner wave than the digital oscilloscope (probably due to its analog nature). Let me know if you agree to my statement after my analysis below.

Noise: I connected both oscilloscopes to the function generator. It is turned on, but the output for the channel is DISABLED so there is only 1V DC output there (no wave, just a flat DC signal).
Check the screens below. Both scopes are displaying the signal with the same settings. The vertical scale is currently at 100mV. You can see that in both scopes the trace is taking almost 2 divisions. In the digital I added AC RMS measure as well as P-P. The AC RMS is showing 52mv variation which make sense.




If I then turn on the signal, we can see that the wave is displayed consistently on both scopes (they both still have same settings).




Now what was making the two scopes show different lines before was the bandwidth limiting setting which was turned ON in the digital scope (my bad). What that does is to limit the bandwidth to 20mhz on the channel removing high frequency noise from the signal. In the same manner the Zoom feature of the 2225 also limits the bandwith (to 5mhz), thus removing the noise.
With that we have the screens below:




Sweet! So everything is allright with the scope, and I got Nothing to worry. Plus a learned quite a lot in this analysis.

Now, I could just have said it was all about the BW limit feature, but I really wanted to show the full analysis because now the questions are:

• Is the amount of noise I'm seeing too much?
• What would be the expected noise that I should see?

I tested the AC RMS on the probe adjust signal on both scopes and they are quite similar (50 in the digital and 60 in the analog).
If I short the probe on the digital oscilloscope I get 12mv reading on the AC RMS, significant smaller.

Thanks!
Walter

[Keith956]

Excuse me if you're tried it, but what do you see if you ground the input to the scopes:

a) At the probe tip - just connect the probe tip to its ground.

b) At the scope - connect a shorting BNC plug, or just set the channel input to gnd.

If the noise disappears then it's coming from your function generator - what make/model?

Don't forget that in addition to bandwidth limiting, digital scopes can do things like average the input which reduces noise effects considerably. And analog scopes don't tend to show fast noise very well, which may explain the difference you're seeing in your pictures.

[wmundstock]

Thank you for the suggestions:

a) At the probe tip - just connect the probe tip to its ground.

That makes the noise lower, but does not remove it.

b) At the scope - connect a shorting BNC plug, or just set the channel input to gnd.

With the output set to GND, I get a sharp line without any noise. I don't have a BNT terminator to test with.

If the noise disappears then it's coming from your function generator - what make/model?

Its a CHEAP JDS6600 but even though this is may be adding noise, I tested with the internal generator of both scopes and they have similar noise levels.

Don't forget that in addition to bandwidth limiting, digital scopes can do things like average the input which reduces noise effects considerably. And analog scopes don't tend to show fast noise very well, which may explain the difference you're seeing in your pictures.

Sure, the digital scope do have averaging for the acquire settings, but its set to normal. Indeed, if I make it average then I get a very thin line.

Note
- If I touch the cable of the probe with my hand the noise pretty much doubles, but I suppose that's normal due to static or something.

[Keith956]

Touching the cable and finding the noise doubles, and shorting the probe tip and finding that does not remove the noise (but setting the cope i/p to gnd does) suggests your probe - or the probe cable - is bad. I wonder if the connection from the bnc to the cable shield is faulty?

[David Hess]

That behavior is consistent with a dirty switch which is hardly unusual in a 30+ year old oscilloscope.

[wmundstock]

That behavior is consistent with a dirty switch which is hardly unusual in a 30+ year old oscilloscope.

When you say switch, I'm supposing you refer to the coupling switch of the channel, is that right? I can try to clean it a bit with air spray + contact cleaner to check if it does any good.

[David Hess]

That behavior is consistent with a dirty switch which is hardly unusual in a 30+ year old oscilloscope.

When you say switch, I'm supposing you refer to the coupling switch of the channel, is that right? I can try to clean it a bit with air spray + contact cleaner to check if it does any good.

The vertical attenuation and coupling switches are directly in the vertical signal path.  The pull x10 switch is not but could still be dirty.  Also check the variable control.

Wiggling and repeated operating each control will likely reveal which one is causes problems.