Test Equipment: Oscillocopes

[TinselKoala]

Yes, that calibrator trace does look a little funky. It should be more level across the tops and bottoms. I presume you have adjusted the probes for the squarest-flattest possible display as per the instructions.  Did you say at one point that you found some voltages that were off-spec in the calibrator circuit? The spec on the calibrator output is 300 mV plus or minus one percent at normal temperatures. So this should produce a flat, horizontal  top and bottom parts of the square wave output. It looks like the output voltage is sagging a bit in the pulses, or at least the display of them is. Now is when you'd like to have a good external square wave generator, and/or another scope,  to see if the problem is in the calibrator or somewhere else in the scope! If the "known good" square wave generator gives you a good flat-topped trace then you know that it is the calibrator sagging, but the scope itself is displaying correctly.

As to the voltage asymmetry: Since you are in AC-coupled input mode I would expect the displayed trace to be more symmetrical about the zero-baseline even if the calibrator signal itself isn't perfectly symmetrical. This may be a "DC bias" issue, which is an internal adjustment that you will come across when you run through the performance checks or calibration procedures. Again, it would be nice to see what happens when you display other signals that are known not to have any DC offset. Have you tried looking at the GenRad oscillator's sine wave output yet? If that instrument is working properly it should be giving a symmetrical sine wave.


I don't know if you have the short "operator's manual" in addition to your SM. I'm attaching it below just in case you don't have it. It has lots of instructions for making the various measurements that are possible with the scope. (It also includes instructions for the DMM option which you don't have, obviously.)

[TinselKoala]

Ah.... do you have the probes in 1x or 10x attenuation? I just noticed that my Tek 2213a does the same thing if I have the probes set to 1x instead of 10x and only when AC-coupled. When DC-coupled the traces flatten out nicely.  The cal trace slopes in exactly the same way yours is doing when I use 1x probe atten and AC coupling.

[TinselKoala]

OK, I've made a short video that I think accounts for both issues: the sloping trace and the voltage asymmetry. At least it accounts for those symptoms on my Tek 2213a, and I think that the same thing is going on with the 465 as well.

My Tek 2213a shows the same thing when the probe is set to 1x attenuation and the channel is AC-coupled. Also,when the probe is set to 1x, the compensation capacitor adjustment has almost no effect on the shape of the trace. When set to 10x attenuation the probe compensation can be adjusted as per the instructions in the probe manual.

When DC-coupling is selected with 1x attenuation, the sloping of the calibrator trace goes away. When the probe is set to 10x attenuation, there is no sloping in either AC or DC coupled input.

The voltage asymmetry around the zero volt baseline when AC-Coupled is due, _I think_, to the slight time asymmetry, or duty cycle, of the calibrator signal. Since  the AC coupling brings the average of the trace down to the zero baseline, it is the _areas_ of the positive and negative going parts of the "square" waveform that are averaged. The narrower (shorter duration) part of the pulse waveform will have a slightly higher peak value than the wider part.

http://www.youtube.com/watch?v=HmHRdh8OnUw

[Brian516]

Ah.... do you have the probes in 1x or 10x attenuation? I just noticed that my Tek 2213a does the same thing if I have the probes set to 1x instead of 10x and only when AC-coupled. When DC-coupled the traces flatten out nicely.  The cal trace slopes in exactly the same way yours is doing when I use 1x probe atten and AC coupling.

Yes, I did have the probes on 1X attenuation.

OK, I've made a short video that I think accounts for both issues: the sloping trace and the voltage asymmetry. At least it accounts for those symptoms on my Tek 2213a, and I think that the same thing is going on with the 465 as well.

My Tek 2213a shows the same thing when the probe is set to 1x attenuation and the channel is AC-coupled. Also,when the probe is set to 1x, the compensation capacitor adjustment has almost no effect on the shape of the trace. When set to 10x attenuation the probe compensation can be adjusted as per the instructions in the probe manual.

When DC-coupling is selected with 1x attenuation, the sloping of the calibrator trace goes away. When the probe is set to 10x attenuation, there is no sloping in either AC or DC coupled input.

The voltage asymmetry around the zero volt baseline when AC-Coupled is due, _I think_, to the slight time asymmetry, or duty cycle, of the calibrator signal. Since  the AC coupling brings the average of the trace down to the zero baseline, it is the _areas_ of the positive and negative going parts of the "square" waveform that are averaged. The narrower (shorter duration) part of the pulse waveform will have a slightly higher peak value than the wider part.

http://www.youtube.com/watch?v=HmHRdh8OnUw

The 465 acts exactly the same as your 2213A in these aspects.
Definitely very helpful information.  Not even the scope manual nor the probe manuals from Tek specified anything at all about them having to be on 10x and DC coupling in order to properly adjust the compensation cap.

One issue I have noticed with these Chinese probes is that the scope does not automatically detect whether I have it set on 1X or 10X.  I seriously doubt it, but are all these fairly cheap Chinese probes this way?  If not, I would have to say that I do not recommend them to anyone. These ones are the model "PP-200", although the seller's listing did not specify this at all. If I would have bothered to ask them, I'm sure they would have had no choice but to tell me.  They said that they have never had any returns, though, and they have 100% feedback and all that jazz.  Seems that really doesn't mean all that much these days.  Do you think that possibly the reason why it doesn't recognize the attenuation setting on these cheap probes is because it is so old?
Is there any way for me to manually change what attenuation the scope shows, so I can match the display to the probes?  I don't really know if this is particularly necessary, since the screen shows exactly the same trace with the cheap ones as it does with the 6062A and 6065A. The only real difference is that the indicator light at the V/DIV knob doesn't change.

There does seem to be a few other quirks with this scope as far as what the screen shows when the settings are not set for the trace to appear entirely on-screen.  At this point, I'm not too concerned about it since, well, who really would want any part of their trace to appear off screen??? And maybe in time after some use, the rest of whatever gunk/oxidation remains inside of the pots and switches will work itself loose with the deoxit that is in there still. If not, here in a few weeks maybe I'll put a drop of WD40 or PB in each one and let 'er sit for a few days and see where it stands then. (of course after I try putting some deoxit and WD/PB in an old pot I have laying around and testing it several times to make sure it doesn't cause any issues/shorts/etc)

Anyway, what are your thoughts on the attenuation sensing w cheap probes issue?  Just keep it in mind and keep on rollin along the knowledge superhighway?

Brian

[Brian516]

Ok, so it appears that I still have a fair amount of "noise" and bad contacts on channel 2. 

-I did a quick test with an old pot similar to the ones used in the scope:
-I hooked it up to my DMM on the Resistance setting, and observed how smoothly/roughly it functioned without any treatment. It was OK, but not all that great. It had quite a fair amount of 'noise', meaning that the R value would jump around a bit while adjusting, instead of smoothly increasing or decreasing.
-I left it connected, and put a couple drops of Deoxit on the shaft and let it seep in for a minute, then slowly adjusted and observed. This created a good bit of "extra noise" in the pot, and what I mean by this is that, when adjusting the pot, the resistance would 'jump' up to 2x or even 3x of the overall R value of the pot. It would immediately settle to where it should be upon stopping rotation.
-I let it sit for a couple minutes and repeated, with close to the same results, but a little less 'jumping'.
-I then put a single drop of PB blaster on the shaft and let it seep in. Waited a couple min, then repeated the adjusting/observing process.  It continued to do the same thing ('jumping' effect) but a little less frequently.
-After sitting for approx 10 minutes, still connected and DMM powered on, while adjusting the pot I get absolutely no 'jumping' and the pot works almost flawlessly, and much, much better than it did before any treatment.  Another 10 minutes, and I am noticing that it is jumping again, but only by about 300 ohms, and only occasionally. It seems to be occurring in about the same positions, though, which I would think might be grime that has settled on the contact surface after the PB has ran off.

I am going to set this pot aside and see how it performs later, check for any shorting, "noise", etc.  I am also going to take a similar pot and do the same but without the PB Blaster.  Then after several hours or so I am going to rinse them out with IPA and see if their performance changes.

I am thinking that, since I still have a fair amount of 'noise' in some of the pots on my scope, I may take the cover and the knobs off and put a drop or two of PB Blaster in them, work it in by turning, let it sit overnight, and then rinse them all out with IPA twice and let them dry out before I power it up again, and see if it's any better.  I just don't think the Deoxit is up to the task in it's entirety (of course I will find out for sure if I can find another old pot that is close to the same as the one I just tested). Not to mention it's only D5, not the PowerBooster stuff that is actually a cleaner instead of mostly conditioner.  Plus I think it would be smart to rinse out the cleaner and loose gunk, anyway. Hopefully that FINALLY solves the rest of my "noise" issues once and for all!!!

Oh, and for everyone who might not know, PB Blaster is the pro-grade equivalent of WD40. PB = "Penetrating Blaster".  It's somehow also made to attract to rust and ferrous metals so it will free up heavily rusted and/or stuck parts. Just a little heads-up in case some of you end up being in need of something like that when your WD40 can't handle the task.

If, after the overnight PB/IPA rinse treatment, she's still got issues with CH2, I will make a vid showing what's happening. I just wish I had a tripod so I don't give you all nausea having to watch my shaky vids....

Brian