Tektronix 2246 Oscilloscope CRT Sparkle/Noise - Normal?

[jordankersten]

Hello everyone. I am new and have a question. I just got a Tektronix 2246 Mod A Oscilloscope. Everything works great on it. The CRT seems to have some "sparkle" or "noise" in it. It is very slight and it is visible both on the traces and the readout. It looks like the traces and readout displays and cursors have sparkling spots in it. Is this normal? If not, what could be causing it?

 

[jordankersten]

One more thing to add...the problem gets worse the more things that are on the screen: it is worst when all 4 channels are displayed with cursors, etc. It smoothes out when the timing is faster but is still present.

 

[Phrak]

Sparkling reminds me of dram memory access aliasing. It could be the guy who designed the video memory at Tektronix didn't bother to do the worst case timing analysis. Thing is, you can suffer some of this without crashing the cpu, because its just the user interface, so sloppy design work gets by. Aging might aggravate marginal timing.

But if this scope has a vector drawn display, I don't know how much of this applies.

 

[NoTime]

Hard to tell from the description.
Could be a leaky HV power supply or noisy transistor in the Z channel.

 

[jordankersten]

It gets worse the more traces you put up on the screen (its a 4 channel scope). The "dashing" is worse on the vertical deflection. A problem in th CRT circuitry? Power supply? AC Line noise?

 

[NoTime]

It gets worse the more traces you put up on the screen (its a 4 channel scope). The "dashing" is worse on the vertical deflection. A problem in th CRT circuitry? Power supply? AC Line noise?

First guess would be to look at the blanking circuitry in the Z channel.
This is supposed to shut the beam off while the vertical beam position is being stepped.
So vertical (or somewhat slanted) dashes on the screen would indicate a problem in this area.

 

[f95toli]

I guess you already know this but...
The 2246 is an old analogue scope meaning there is probably only a single beam drawing all four channels; this means that there is presumably a CHOP/ALT button somewhere which controls how the channels are drawn.
If this button it set to CHOP the lines will indeed look like they are made up of spots since the beam continuosly jumps between the lines spending 1/4 of its time on each channel meaning only 1/4 of each line is actually drawn (in ALT mode the scope draws one complete channel at a time).

 

[Averagesupernova]

I guess you already know this but...
The 2246 is an old analogue scope meaning there is probably only a single beam drawing all four channels; this means that there is presumably a CHOP/ALT button somewhere which controls how the channels are drawn.
If this button it set to CHOP the lines will indeed look like they are made up of spots since the beam continuosly jumps between the lines spending 1/4 of its time on each channel meaning only 1/4 of each line is actually drawn (in ALT mode the scope draws one complete channel at a time).

Not likely to see this because the scope will not be triggering at rate to see these spots. I used to work on analog scopes for a living. Another scope is needed to troubleshoot the defective one.

 

[jordankersten]

Not likely to see this because the scope will not be triggering at rate to see these spots. I used to work on analog scopes for a living. Another scope is needed to troubleshoot the defective one.

Agreed. It doesn't matter what mode its in (chop or alt) it does the same thing. It gets worse when Line triggering is used. Could it be a power supply filter? Line noise leaking in?

 

[Averagesupernova]

Agreed. It doesn't matter what mode its in (chop or alt) it does the same thing. It gets worse when Line triggering is used. Could it be a power supply filter? Line noise leaking in?

Sounds likely that it is some type of power supply noise. NoTime has already suggested things of this nature.

 

[jordankersten]

Here are some pics of the display problem...one is 10X mag chop and one is 10X mag alt. It smoothes out when you turn the readout off...intensity of the readout has no affect, only if it is on or off.

 

[jordankersten]

The problem goes away when the readout is turned off. The readout intensity doesn't affect it at all, but when you turn the readout intensity all the way down (off) the problem disappears.

 

[jordankersten]

Ok, the service manual says that When ROR (Readout Request or something like that) input is low, chop blanking is disabled and the ROB (Read Out Blanking) is inverted and controls the blanking. When ROR goes from low to high, the blank output remains connected to the readout blank signal for an additional four to six TC clock periods to mask vertical source-switching transients. My question is this: on my ROR, high input is 5.06V and low is 4.65. Is 4.65 low enough to be considered "low" and enable the readout blanking? When the readout pot is turned all the way off, I get the 5.06 supply voltage on the ROR pin. When the readout is on (doesn't matter to what intensity, just on) then the ROR drops to 4.65V to enable the readout blanking (ROB). Does that voltage difference sound right?

 

[NoTime]

The 4.65 is unlikely to be a DC voltage, so a DVM is not going to show a meaningful low.

It has been a while since I looked at one of these.
But, The display images you posted don't really look out of line.
So barring another sample that doesn't show this artifact I'm inclined to think there is nothing wrong.
Perhaps someone else here has seen this model more recently.

 

[jordankersten]

If the 7.5 Volt adjust is at exactly 7.5V, the "dashes" produced by the readout scans across the trace. If you bring the 7.5V adjust lower, it scans faster and faster. If you bring the 7.5V up, the "dashes" scan faster and faster disappearing right around 8 volts. All other voltages check out ok. The 15v supply is almost 16v when 7.5v is exactly 7.5. All other voltages are right on.

 

[jordankersten]

One more thing: if there are two readout displays on (CH1 and time) there are two sets of dashes that scan across the trace. If you add other things to the readout (CH1 CH2, time, etc.) another set of dashes scans across the trace for everything you add to the readout.

 

[Phrak]

From the display, it looks like the same goofy tectronixs i'd been using for the past 10 years.
I don't see the sparkling you're having trouble with, just some segmented traces.

 

[jordankersten]

It's not really sparkle. It's just "dashed" or broken up when the readout is on.