EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: heima on July 03, 2016, 08:40:31 pm
-
Hello All,
I am a mechanical engineer, so please bear with my naivte.
I bought a Tek 2230 for nostalgic reasons (used one back in the 80's) and it seems to be working fine, except for a strange blank/glitch/dropout It is only seen in analog mode, or I cannot see it or seem to capture it in storage mode. I noticed it when viewing video waveforms.
Anyhow, it doesn't seem to be signal dependent, and no signal input is necessary, as you will see in the following pictures. The blank/glitch/dropout (whatever its called) happens every 2.7ms for a duration of about .2ms. Take a look at tek2230dropout1.jpg.
This blank/glitch/dropout is seen regardless the sweep setting or triggering. The scope was configured to be able to get a photo of it. Yes, the scope is triggered. Take a look at tek2230dropout2.jpg.
My question for the experts, is: Any idea what this is? 2.7ms is 370Hz. Is there something going wrong in the Z drive? I think that is the right way to say it. The beam creation. Is this possibly an early sign of a failing power supply that these scopes have a reputation of blowing? Fingers crossed it is not.
Thanks for your help.
-
IMO something is causing the scope to trigger, EMI perhaps.
Adjust the V/div to find the top of the waveform.....the bit that's missing from the flat line trace.
-
That is the multiplexing needed to draw the readout while in analog mode. It does not show up in digital storage mode because then there is no analog display present. In your case, you have the readout enabled but its intensity is so low that you cannot see it but the multiplexing takes place anyway. If you fiddle with the trigger holdoff or time/div variable control, you can get the gaps in the trace to hold still at certain time/div settings.
When you turn the storage/readout intensity control fully counterclockwise, the 2230 toggles whether the readout is displayed or not. In storage mode, this allows the digital storage traces to be displayed without the readout. In analog mode, this allows the readout to be disabled so that there is no interference from the multiplexing needed to display it. This is a ubiquitous feature of Tektronix analog oscilloscopes which include readout capability.
The 2230 is a great albeit peculiar early DSO; I have two of them.
-
Couldn't this effect also be caused by the add/chop/alternate setting? The read-out intensity is almost half way so it should show something (yes, I also used to own a 2230).
-
That is the multiplexing needed to draw the readout while in analog mode. It does not show up in digital storage mode because then there is no analog display present. In your case, you have the readout enabled but its intensity is so low that you cannot see it but the multiplexing takes place anyway. If you fiddle with the trigger holdoff or time/div variable control, you can get the gaps in the trace to hold still at certain time/div settings.
When you turn the storage/readout intensity control fully counterclockwise, the 2230 toggles whether the readout is displayed or not. In storage mode, this allows the digital storage traces to be displayed without the readout. In analog mode, this allows the readout to be disabled so that there is no interference from the multiplexing needed to display it. This is a ubiquitous feature of Tektronix analog oscilloscopes which include readout capability.
The 2230 is a great albeit peculiar early DSO; I have two of them.
Winner, winner, chicken dinner! That was it!
Thank you so much Dave!
Its funny that I never noticed this before, but that was 30 years ago, and the memory fades. Maybe for the better.
-
Couldn't this effect also be caused by the add/chop/alternate setting?
Sometimes you can see artifacts from chop mode but they are more regular and much higher frequency requiring like a 10 us/div sweep or faster to become objectionable. The readout multiplexing is most apparent at around 1 ms/div and slower.
Some Tektronix oscilloscopes can generate the readout between sweeps so there are no artifacts from multiplexing.
The read-out intensity is almost half way so it should show something (yes, I also used to own a 2230).
I thought the same thing when I looked closely at the second photograph showing the position of the readout/storage intensity control. The z-axis calibration may be off.
Its funny that I never noticed this before, but that was 30 years ago, and the memory fades. Maybe for the better.
I did not remember clearly either so broke out one of my 2230s to verify operation before I posted. I was able to replicate your results easily which convinced me it was just the readout and not some other problem.
When you adjust the A sweep intensity control, at what point does the sweep become just barely visible?
The z-axis calibration involving the grid bias may be off or maybe the CRT has a lot of hours on it making it dimmer than it should be. In your first photograph it looks good though.
-
When you adjust the A sweep intensity control, at what point does the sweep become just barely visible?
If the marking on the knob was akin to the hour hand on a clock, it would be just barely visible at about 9:30. For the picture, the marking on the knob was pointing at the "E" in "INTENSITY A-@-B". The sweep can get much, much brighter. Bright enough that I suspect it might burn the phospor.
I cropped the picture of the sweep, because the flash from the camera was so bright. In doing so, I also cropped out the readouts. Sorry if that threw anybody for a loop.
I do not think this scope has seen much use. The timebase knob is very stiff, as are the vertical amplifier (???) knobs. In fact, all knobs and switches seem fine, with the exception of the CURSORS knob in the upper center. It turns smooth, but feels floppy. I think I need to open this up and see if a retaining collar or sleeve has slid down a shaft. It also does not move the cursors smoothly. They kind of jump from point to point, not move around the waveform. The service manual does not have a good diagram of this assembly, so I am only guessing why its floppy.
Thanks to all who provided support, I really appreciate it. :-+
-
I cropped the picture of the sweep, because the flash from the camera was so bright. In doing so, I also cropped out the readouts. Sorry if that threw anybody for a loop.
I do not think this scope has seen much use. The timebase knob is very stiff, as are the vertical amplifier (???) knobs. In fact, all knobs and switches seem fine, with the exception of the CURSORS knob in the upper center. It turns smooth, but feels floppy. I think I need to open this up and see if a retaining collar or sleeve has slid down a shaft. It also does not move the cursors smoothly. They kind of jump from point to point, not move around the waveform. The service manual does not have a good diagram of this assembly, so I am only guessing why its floppy.
It could be the pot just needs some "exercise" with the wiper or the track crusty due to age.
-
When you adjust the A sweep intensity control, at what point does the sweep become just barely visible?
If the marking on the knob was akin to the hour hand on a clock, it would be just barely visible at about 9:30. For the picture, the marking on the knob was pointing at the "E" in "INTENSITY A-@-B". The sweep can get much, much brighter. Bright enough that I suspect it might burn the phospor.
9 o'clock for barely visible is about right and matches the performance of mine so there is no reason to mess with the grid bias adjustment. The CRTs on these oscilloscopes have a long operating life and the phosphor is very resistant to being burned.
I cropped the picture of the sweep, because the flash from the camera was so bright. In doing so, I also cropped out the readouts. Sorry if that threw anybody for a loop.
That explains it then. I was worried that there was a problem with the intensity of the readout which would be weird.
I do not think this scope has seen much use. The timebase knob is very stiff, as are the vertical amplifier (???) knobs.
The volt/div and time/div controls should feel snappy without friction. The time/div control is naturally stiffer until the A and B time/div controls are unlocked.
In fact, all knobs and switches seem fine, with the exception of the CURSORS knob in the upper center. It turns smooth, but feels floppy. I think I need to open this up and see if a retaining collar or sleeve has slid down a shaft. It also does not move the cursors smoothly.
I found that sometimes the potentiometers get stiff and removing the knob to lubricate the shaft from the front is enough to restore smooth operation. A couple drops of a light silicon oil works well.
They kind of jump from point to point, not move around the waveform. The service manual does not have a good diagram of this assembly, so I am only guessing why its floppy.
The cursor controls all seem to do this to one extent or another. I have not messed with any of mine yet to try and improve them. Maybe try spinning it through a few complete rotations in both directions to see if it improves.
The 2230 is more difficult to work on then the other 22xx oscilloscopes including the 2232 which replaced it. The outside enclosure is easy enough to remove but the digital board on top prevents access and does not easily tilt out of the way unlike the 2232.
-
It could be the pot just needs some "exercise" with the wiper or the track crusty due to age.
Just some mechanical exercise may be enough.
The control itself uses two potentiometers in parallel but rotated with respect to each other so by reading both, there is no dead zone and the processor can use them as a continuous rotary encoder. Tektronix continued to use these in later oscilloscopes however the 2230 and 2232 always seem to be "scratchier" and I do not think it is because of age.
I have not had the moxie to tear one of these small double potentiometers apart yet. My guess is that like the other high quality Mod-Pots, it uses a carbon brush for the wiper. Carbon residue wears off of the carbon brush and cakes with the lubrication eventually preventing good contact between the wiper and element.
The built in diagnostics include an "exercise" mode which reads back the status of all of the controls including the double potentiometer cursor control in real time as you mess with them. With care this might show if there is a problem but I tried it with mine and despite the cursors being jumpy, I could not see any indicated problem.
-
The built in diagnostics include an "exercise" mode which reads back the status of all of the controls including the double potentiometer cursor control in real time as you mess with them. With care this might show if there is a problem but I tried it with mine and despite the cursors being jumpy, I could not see any indicated problem.
Well, I got up the gumption to take the scope apart. At least the housing off. There are some EMI fingers at the front of the chassis, just behind the front panel. One got loose and wedged between the chassis and the case. That caused it to chew up the inside of the housing when trying to get the housing off. It was tough too. Once apart, I used a knife and a wet towel to scrape down the chewed up part. I don't think I got any aluminum flakes into the scope electronics. Fingers crossed.
Anyhow, looking at the shaft for the cursor control, the reason it is floppy is because it is mounted to a flexible joint. That joint permits the shaft to flop around, yet still apply torque. To tighten this floppy-ness up would require adding a sleeve to the shaft to increase its diameter, or somehow reducing the inner diameter of the hole which the shaft passes through. Another project for another day.
Oh, I noticed the behavior of the cursor knob. It appears to be velocity sensitive. Twist fast, and the cursor moves more smoothly for a greater distance. Slow down, and the cursor move less smoothly at a smaller distance. Slow even more, and the cursor behaves like it is jumping from one sample position to another.
It appears that there are 3 cursor movement speeds. Fast, Medium, and Slow (sample by sample). So what do I mean by velocity sensitive? Twist the cursor through an angle, say 90 degrees very quickly. Notice how the cursor moves smoothly for a given distance. Now twist the cursor through the same angle, but at a slower rate. The cursor does not move as far as before and not as smoothly. Now twist the cursor through the same angle, but really slow. The cursor might move from one to two sample positions. It is when you move the cursor at a speed between Fast, Medium, or Slow, that it does the jumping.
Its not a bug, its a feature! :)
-
Well, I got up the gumption to take the scope apart. At least the housing off. There are some EMI fingers at the front of the chassis, just behind the front panel. One got loose and wedged between the chassis and the case. That caused it to chew up the inside of the housing when trying to get the housing off. It was tough too. Once apart, I used a knife and a wet towel to scrape down the chewed up part. I don't think I got any aluminum flakes into the scope electronics. Fingers crossed.
The EMI fingers are annoying sometimes. One of my 2230s had the same problem getting the housing off.
Anyhow, looking at the shaft for the cursor control, the reason it is floppy is because it is mounted to a flexible joint. That joint permits the shaft to flop around, yet still apply torque. To tighten this floppy-ness up would require adding a sleeve to the shaft to increase its diameter, or somehow reducing the inner diameter of the hole which the shaft passes through. Another project for another day.
I have not had one of my 2230s apart in a while so did not remember the details about the cursor control. The 2232 cursor control is slightly different because the push-button switch is part of the rotary control instead of having a separate shaft.
Oh, I noticed the behavior of the cursor knob. It appears to be velocity sensitive. Twist fast, and the cursor moves more smoothly for a greater distance. Slow down, and the cursor move less smoothly at a smaller distance. Slow even more, and the cursor behaves like it is jumping from one sample position to another.
It appears that there are 3 cursor movement speeds. Fast, Medium, and Slow (sample by sample). So what do I mean by velocity sensitive? Twist the cursor through an angle, say 90 degrees very quickly. Notice how the cursor moves smoothly for a given distance. Now twist the cursor through the same angle, but at a slower rate. The cursor does not move as far as before and not as smoothly. Now twist the cursor through the same angle, but really slow. The cursor might move from one to two sample positions. It is when you move the cursor at a speed between Fast, Medium, or Slow, that it does the jumping.
Its not a bug, its a feature! :)
It *is* a feature but at least on mine when I reverse direction, it tends to jump in reverse unless I am at the slowest speed. Even with the jumpiness I find these cursor controls to work better than the cursor controls on most current DSOs. The fast mode is really needed to scroll through a 4k record length.