My oscilloscope just did something strange

Thread Starter

adam555

Joined Aug 17, 2013
858
I prefer to mess as little as possible with the oscilloscope. In any case, it's been almost a week since I cleaned the pot that MrChips pointed out, and so far I had no more problems.

There is something though that I'm unable to tune: the screen doesn't seem to trace perfectly straight lines (e.g. when you set a channel to GND); they bend downwards a little bit at the right hand side. It's not much of a problem, but it would be nice to set it straight.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
There should be a little screw adjustment marked trace rotation on the front panel.....

HTH Steve.
I tried that, it rotates the trace; but it rotates the trace including the bent. Not sure if I'm explaining it clearly. You never get a perfectly straight trace, whether you rotate it to the left or right it's always slightly bent.
 

ian field

Joined Oct 27, 2012
6,536
I tried that, it rotates the trace; but it rotates the trace including the bent. Not sure if I'm explaining it clearly. You never get a perfectly straight trace, whether you rotate it to the left or right it's always slightly bent.
Wrong adjustment of probe compensation can make the tops of pulses droop - if the whole trace is wonky while not displaying a signal could mean the mu-metal shield round the tube needs degaussing.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
Wrong adjustment of probe compensation can make the tops of pulses droop - if the whole trace is wonky while not displaying a signal could mean the mu-metal shield round the tube needs degaussing.
Yes, it's bent without any signal. I mean, if I set a channel to GND, it should trace a straight line. Well, this line is never completely straight; it's always slightly bend towards the end (the right-hand side). You can also see it with a signal, for example: if the signal is a square wave, and you amplify it until it covers the full grid, you can also see this slight curvature of the trace on the top right and bottom right corners.

If it needs degaussing, how do I do that?
 
Last edited:

ian field

Joined Oct 27, 2012
6,536
The X & Y deflections systems should be pretty much independent from each other - the only interconnecting system I can think of is the sync takeoff from Y to X, maybe a glitch is getting back through there - but you should be able to isolate that by selecting ext trig.

Supply ripple could affect the trace, but IWHT you'd be able to resolve the 100Hz half cycles if you slow the timebase right down. That would seem quite likely if the bend in the trace moves with the trace rotation adj.

If its become randomly magnetised, you used to be able to get demagnetising wands for the TV trade, generally you switch on the wand and sweep it every which way over the scope and move it well away from the scope before killing the current.

There used to be a plentiful supply of scrap colour TVs, they contain a degauss coil that could be put around the scope - but you can't just bung mains through the coil, it'll burn up pretty quickly! Best to wire it up with the original degauss posistor from the TV board - this takes about 6 minutes to cool down so you can move the coil along and degauss the next few inches of scope.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
Thanks Ian, I just tried changing the timebase but it does it in all time divisions.

Adding a degauss coil seems to complicated to me. If it doesn't have an easy solution -like setting a pot or something like that- then I'm going to leave it as it is. You barely notice it in everyday use; it's just a pain when you do a full calibration -that's how I noticed it.
 

ian field

Joined Oct 27, 2012
6,536
The timebase makes the trace go sideways, so it probably isn't the culprit for a vertical aberration.

Can you pull the vertical amplifier? - that might help you narrow things down a bit, but there's a few gotchas! Often the high voltage Y-plate drivers are on the chassis and not on the Y-amp module, so removing it may not remove the fault. Also if on-chassis drivers have an undefined input, you may not even see any trace. If the fault is taken out with the Y-amp, you still have to determine whether its the amp or PSU.

I wasn't suggesting any modification to degauss the scope - you simply use the degauss coil from an old TV, salvaging the original degauss posistor is the safest way to avoid setting fire to the coil. The degauss posistor takes 6 minutes to cool before you can hit it again so you have to edge the coil along the outside (preferably surrounding) of the scope a few inches at a time and wait 6 minutes before you can hit it again.

If you have a variac it all becomes very easy - but not everyone has a variac.

With a variac I'd suggest twisting a degauss coil into a figure of 8 and folding it on itself (if its thin and big enough to fold twice - do so) make the coil as small as you can and remove the scope covers so you can get in and around the mu-metal shell round the tube. When done, move the coil at least a couple of feet from the scope, then turn the variac right down before killing the power.

If you have a reel of magnet wire, you can wind your own degauss coil - a few inches dia, a few hundred turns of anything from 0.2 - 0.4mm wire.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
I really appreciate your help, but I don't think I'll be able to fix this one; all that is way beyond my league. :(

In any case, I don't have the variac or the degauss coil; and to be honest, I just had to look on the internet what they are.

If the fault was getting worse or if it wasn't working at all, I wouldn't mind risking it; but since it's not that bad, I'm a bit afraid of messing with it without knowing exactly what I'm doing.
 

alfacliff

Joined Dec 13, 2013
2,458
check the power supplies for hum, filter caps tend to fade in older gear, and could cause a problem common to both horiz and vert.
cliff
 

ian field

Joined Oct 27, 2012
6,536
check the power supplies for hum, filter caps tend to fade in older gear, and could cause a problem common to both horiz and vert.
cliff
If the aberration was due to supply ripple, it would be different at different timebase settings.

Linear PSU ripple would be resolved at slow timebase speeds and SMPSU at the faster speeds.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
Doesn't seem to be humming.

I just realized I made a small mistake when explaining the fault: it bends downwards on the top side of the screen, but just a tiny bit upwards on the bottom; both on the right-hand side. So, if you plot a square wave covering the whole grid, it would be slightly taller on the left-hand side than on the right-hand side.
 
Top