Square wave question.

[Jesse1973]

Recently purchased a Kenwood CS4125 scope.Scope powers up fine.Controls work.Following the user manual for pre power up settings.So far so good.Attach probes.Follow probe compensation. Then square wave I get using the onboard test point or separate freq generator isnt exactly square.I am a noob when it comes this so bare with me.The peak is flat and parallel to the x axis .Then rise and fall are not perpendicular to the x axis.Actually they are slightly angled .The both lean in the same direction from left to right.I am guessing something is wrong with the scope .What could it be  and can it be fixed.

[DaJMasta]

This is the reality of how signals work.  If you zoom out to see several periods of the waveform it will look pretty square, if you zoom in on the edge it will always be slanted.

This is usually called "rise time" in specifications, and it's effected by the switching speed of the source and the input bandwidth of the measurement device.  If both of those are infinite, it looks like a perfectly vertical line, but in reality, that cannot exist, so provided your equipment is fast enough, you will always be able to see that rise time.

That said, designers of scopes understand this, so even if your scope measures a single point on the low side and a single point after the full rise - what would appear as an instantaneous rise - the interpolation algorithm that draws the line between those points factors in the limited bandwidth of the scope and will always be at least slightly slanted.

[newbrain]

The both lean in the same direction from left to right.

Crude diagrams follow:

This, as DaJMasta says, is normal.

This is definitely not, and means something is wrong/misaligned in your scope.

[Brumby]

Thank you for those diagrams, newbrain.  That clearly shows the concern I had.

[Jesse1973]

The wave looks like the second picture.I guess there is something wrong then.

[Grandchuck]

Wonder if a magnetic disturbance is possible?

[Jesse1973]

Is that something that can be fixed ?If so who would I send an old school scope to?

[TerminalJack505]

Switch the input to GND and ensure the trace is aligned with the X-axis.  If it isn't then use trace rotation to adjust it.

[ebastler]

I struggle to see how this slanted signal shape, with the X axis not orthogonal to the Y, could be caused by a flaw or misadjustment in the electronics alone. Looks more like the CRT's deflection electrodes for the Y beam deflection have been moved out of whack?

[newbrain]

I struggle to see how this slanted signal shape, with the X axis not orthogonal to the Y, could be caused by a flaw or misadjustment in the electronics alone. Looks more like the CRT's deflection electrodes for the Y beam deflection have been moved out of whack?

Frankly, I am as puzzled as you.
I do not think electrodes on a scope CRT can be moved. Aren't they embedded in the glass structure?
I would rather suspect a crosstalk between horizontal and vertical deflection circuits, quite possibly through the supply.

EtA: on a lighter note you can compete on the "Show us your square wave" thread with a negative fall time!

[ebastler]

Thanks, newbrain. Right, the Y circuit cross-talking into the X circuit would also explain it.

[StillTrying]

I think a magnet near the gun end of the CRT could cause that.
Have you tried the scope in a different physical position.
Is the square wave the same shape when it's in the top 3 Y divisions and bottom 3 Y divisions.
A small image of the screen might help.

[Jesse1973]

Okay here it is .Not sure why it is all fuzzy in the pic the screen is quite sharp.

[0culus]

Exposing CRTs in your photographs is certainly an artform...no worries there. It's hard to tell but it almost looks like the trace rotation is off...have you tried turning that adjustment with a small screwdriver? Nevermind, looking at it again the peaks appear to be aligned with the graticule.

I'd expect a strong magnetic field distortion to affect everything. 

[james_s]

I struggle to see how this slanted signal shape, with the X axis not orthogonal to the Y, could be caused by a flaw or misadjustment in the electronics alone. Looks more like the CRT's deflection electrodes for the Y beam deflection have been moved out of whack?

Frankly, I am as puzzled as you.
I do not think electrodes on a scope CRT can be moved. Aren't they embedded in the glass structure?
I would rather suspect a crosstalk between horizontal and vertical deflection circuits, quite possibly through the supply.

EtA: on a lighter note you can compete on the "Show us your square wave" thread with a negative fall time!

No, the deflection plates are spot welded to the metal structure of the electron gun but they are otherwise mostly unsupported and being relatively large they are quite fragile. I've seen bent deflection plates on multiple occasions, it can happen if the scope is dropped or knocked over. An xray of the CRT can verify the condition.

[frogg]

I'm sure others would already say this, but given that the Kenwood CS4125 has a CRT, and the CRT is scanned line by line, it's pretty obvious to me that whatever is vertically syncing with the horizontal scan is slightly off.

You'd have to find the schematics of the scope to know what's syncing the horizontal and vertical scans.

[shakalnokturn]

...Kenwood CS4125 has a CRT, and the CRT is scanned line by line, it's pretty obvious to me that whatever is vertically syncing with the horizontal scan is slightly off.

I'm not sure about that... Not sure about what would cause such a display either...

In an CRT scope the syncing business is usually called the trigger if my understanding is right.

[Circlotron]

I'm sure others would already say this, but given that the Kenwood CS4125 has a CRT, and the CRT is scanned line by line, it's pretty obvious to me that whatever is vertically syncing with the horizontal scan is slightly off.

What you say might apply to early DSOs with CRTs but analog scopes like this one draw the trace as one continuous wiggly line from left to right. No raster scanning here.

[Brumby]

..... but analog scopes like this one draw the trace as one continuous wiggly line from left to right. No raster scanning here.

Worth repeating!


Yes little Jimmy.  There is such a thing as 'analogue'!

[Brumby]

I would rather suspect a crosstalk between horizontal and vertical deflection circuits

That does explain what is displayed.  Specifically, the X-axis getting a sniff of the Y-axis deflection.

I see the other suggestions as having a poor fit.

[Fungus]

I would rather suspect a crosstalk between horizontal and vertical deflection circuits

That does explain what is displayed.  Specifically, the X-axis getting a sniff of the Y-axis deflection.

Yep, and it sounds like a very difficult thing to fix.

[james_s]

I'm sure others would already say this, but given that the Kenwood CS4125 has a CRT, and the CRT is scanned line by line, it's pretty obvious to me that whatever is vertically syncing with the horizontal scan is slightly off.

You'd have to find the schematics of the scope to know what's syncing the horizontal and vertical scans.

This is just plain wrong.

You are describing a raster display, while an analog oscilloscope uses a vector display. The image is not scanned line by line, the timebase directly controls the horizontal sweep while the signal you want to look at drives the vertical deflection and the result is the trace is drawn in real time. Scopes that have a readout use a time sharing technique where the beam shoots off to trace out the characters then returns to the trace.

There's more than one way to drive a CRT. Analog scopes use electrostatic deflection tubes but there are magnetic deflection vector monitors too, mostly used in vintage arcade games like Asteroids and Tempest.

[rf-loop]

I'm sure others would already say this, but given that the Kenwood CS4125 has a CRT, and the CRT is scanned line by line, it's pretty obvious to me that whatever is vertically syncing with the horizontal scan is slightly off.

You'd have to find the schematics of the scope to know what's syncing the horizontal and vertical scans.

No. It is absolutely sure normal xy vector tube as in nearly all pure analog scopes what I know over tens of years. It can also see even nearly without eyes when look trace in image. Raster monitor do not draw like this. Never.
--------------------


For OP

It looks bit like magnetic error if not like some others have told about some crosstalk (what I bit suspect when I look trace shape, but ... all is possible.). I do not know Kenwood internals but some scopes have small permanent magnets for fine adjust symmetry (of course expensive higher class scopes have also other methods for adjust some linearity and trace bending like example in many old Tektronix) but also I have seen scopes what have been in strong magnetic field and some parts around tube are magnetized  so it may need perhaps demagnetize.

[trevatxtal]

Simple the crt tube has been rotated in its clamps.
If you have ever replace a crt tube then this is the first thing you set up.

[StillTrying]

If the tube was rotated to fix it, the whole width of the scan would then slope upwards, and the straight tops and bottoms of the square wave would then slope in an unusual way. I don't think interaction between the Y and supplies would that perfect, so I'm sticking with magnetism.