Dumb Ass with a scope

[easilyconfused]

I took Dave's advice and bought an old analog scope. It' an Iwatsu SS-5710. I have no business with a scope being the t-total beginner that I am. But I convinced myself it would help me to understand things better if I connected it to my Radio Shack Electronics Learning Lab experiments. The guy wanted a $100. I offered half that. He declined so I walked. Then I went back and gave him the hundred. Anyhow it seems to be out of cal. The PDF I found says the calibration lug should output 0.3v and this one only shows 0.15v. I've tried many settings and get the same result. The time axis is out as well. I looked under the hood but didn't dare touch anything. I'm mindful that there's lots of voltage in there even when unplugged. I didn't see anything that made me feel confident about twisting them. There were a dozen or so trimmers (I guess that's what you call them), but none of them said Y adjust or X adjust so I put it back together. Can anyone suggest how to calibrate it. I doubt if I can afford to have it done professionally.

[ThunderSqueak]

Take a good picture of it on so we can read everything while you are testing the calibration... just curious on how you have it setup while you are testing calibration.

[easilyconfused]

Take a good picture of it on so we can read everything while you are testing the calibration... just curious on how you have it setup while you are testing calibration.

In this pic my settings are 50mV/0.2mSEC. I'm attached to the calibration lug and the probe is switched to the X10 scale. The PDF file calls for 0.3v and it looks to me like I have half that. I hope I'm looking at it wrong.

[EEVblog]

That looks like 300mV to me!
6 divisions of 50mV

Freq looks pretty close to 1KHz, just a bit off. But this could be the calibrator and not the scope.

Dave

[NiHaoMike]

That was probably 0.3Vpp.

[DJPhil]

I picked up an old Tek 465B a while back, and the cal on it is a bit slow, a bit low, and quite noisy. The scope itself performs well, and casual use at <1MHz hasn't unearthed any significant problems. I'm pretty sure it's just the calibrator is just showing it's age a bit in both our scopes.

If you're just getting started I recommend reading through Tek's updated XYZ's of Oscilloscopes. It's a quick read and one of the best all purpose introductions I've seen. With that handy, you can begin to put it through it's paces if you have or can build a signal source or two.

Hope that helps.

[EEVblog]

That was probably 0.3Vpp.

Yeah, scope cal outputs are always rated in peak-peak voltage.

Dave.

[ThunderSqueak]

That looks like 300mV to me!
6 divisions of 50mV

Freq looks pretty close to 1KHz, just a bit off. But this could be the calibrator and not the scope.

Dave

What Dave said

Which is why I asked for a picture, was curious how you were reading it.  

[easilyconfused]

That looks like 300mV to me!
6 divisions of 50mV

Freq looks pretty close to 1KHz, just a bit off. But this could be the calibrator and not the scope.

Dave

Obviously my logic is screwed up. My thinking was as follows. My dial is set on 50mV. I have six large squares with 5 divisions each or a total of 30 divisions. I multiplied the 30 divisions by 50mV (each division) for a total of 1500mV. I divided that by 1000 to reduce the millivolts to volts. The number is 1.5 volts. Then I divided that by the X10 (the probe setting) to reduce the number to a 1:1 result which is 0.15volts. I went to bed after posting the pic. I couldn't sleep. Then I had one of those revelations where you sit up and slam your palm into your forehead and say to yourself "Why don't you check the voltage on the lug to see what it is instead of what it's supposed to be (based on the PDF)". So I got up, came in here and checked the lug with my new Fluke 17B (which I also don't deserve). The voltage was 1.48. So doesn't that support my logic? Or not? As to the timing-- I don't know how to check that with my Fluke. It does have a Hz% button. But when I press it (while in the DCV mode) it simply shows 0.000.

[easilyconfused]

I guess I've pretty well exhausted this thread. But I think the conclusion is that the voltage on the scope is consistent with the multimeter voltage on the lug. And-- the timing is correct IF the lug is actually pulsating at 1KHz (which I have not verified 'cause I don't know how).
And-- with regard to the voltage-- I really don't care how many volts are coming out of the lug as long as the scope reflects that voltage. I hope my reasoning is correct. Thanks for all responses. I really do appreciate it. I know I really don't belong on a forum mainly populated with advanced practitioners. I will take a look at the XYZ PDF. I've been there before but it might mean more now since I've developed a little more understanding of scopes and how to interpret them.

[DJPhil]

It does have a Hz% button. But when I press it (while in the DCV mode) it simply shows 0.000.

You'll have to be in the ACV mode to read frequency. Switch to ACV, change range to get a good reading, and hit the Hz% once, and you should be good. Hitting the Hz% again will show you duty cycle, which should be very close to 50%.

Hope that helps.

[easilyconfused]

It does have a Hz% button. But when I press it (while in the DCV mode) it simply shows 0.000.

You'll have to be in the ACV mode to read frequency. Switch to ACV, change range to get a good reading, and hit the Hz% once, and you should be good. Hitting the Hz% again will show you duty cycle, which should be very close to 50%.

Hope that helps.

I did as you suggested. In the AC mode I got a voltage very similar to the DC mode (about 1 volt higher). I pressed the Hz% button and got zeros. I pressed it again and got zero percent. 

[ThunderSqueak]

Here, look at this, a cute little tutorial on how to read your scope  I think it will help

Oscilloscope AC Voltage Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE6803

Oscilloscope: DC Voltage Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3403

Oscilloscope Frequency Measurements
http://www.wisc-online.com/objects/ViewObject.aspx?ID=ACE3503

Oscilloscope Coupling Switch
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3603

Oscilloscope Phase Shift Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3803


seriously.... go through each one 


 

[easilyconfused]

Here, look at this, a cute little tutorial on how to read your scope  I think it will help

Oscilloscope AC Voltage Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE6803

Oscilloscope: DC Voltage Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3403

Oscilloscope Frequency Measurements
http://www.wisc-online.com/objects/ViewObject.aspx?ID=ACE3503

Oscilloscope Coupling Switch
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3603

Oscilloscope Phase Shift Measurements
http://www.wisc-online.com/Objects/ViewObject.aspx?ID=ACE3803


seriously.... go through each one 


 

I absolutely will. And thank you.

[ThunderSqueak]

Oh, something that might not be obvious in those tutorials, they want you to click on the controls sometimes during the tutorial and wont show the "next" button until you do so.    Keeps you on your toes  and out of "click next to get it over with" mode.

[EEVblog]

That looks like 300mV to me!
6 divisions of 50mV

Freq looks pretty close to 1KHz, just a bit off. But this could be the calibrator and not the scope.

Dave

Obviously my logic is screwed up. My thinking was as follows. My dial is set on 50mV. I have six large squares with 5 divisions each or a total of 30 divisions. I multiplied the 30 divisions by 50mV (each division) for a total of 1500mV. I divided that by 1000 to reduce the millivolts to volts. The number is 1.5 volts. Then I divided that by the X10 (the probe setting) to reduce the number to a 1:1 result which is 0.15volts.

You have the concept incorrect. Those "large squares" are the divisions. The smaller marks are not counted.
So your signal is 6 "divisions" high or 50mV * 6 = 300mV peak-to-peak. That's exactly what it should be.

I went to bed after posting the pic. I couldn't sleep. Then I had one of those revelations where you sit up and slam your palm into your forehead and say to yourself "Why don't you check the voltage on the lug to see what it is instead of what it's supposed to be (based on the PDF)". So I got up, came in here and checked the lug with my new Fluke 17B (which I also don't deserve). The voltage was 1.48. So doesn't that support my logic? Or not? As to the timing-- I don't know how to check that with my Fluke. It does have a Hz% button. But when I press it (while in the DCV mode) it simply shows 0.000.

You can't measure the peak-to-peak voltage of anything with a multimeter, it doesn't work like that. Meters are designed to measure either the average or true RMS value of a an AC waveform.

Dave.

[djsb]

Also each subdivision equals 10mv (50/5). So if the trace only goes part way across a full division you count the number of subdivisions and multiply by 10mv (at the 50mv per division setting). This is then added to the full division reading as above.
I've probably confused you even more now.

David

[squeezee]

You can't measure the peak-to-peak voltage of anything with a multimeter, it doesn't work like that. Meters are designed to measure either the average or true RMS value of a an AC waveform.

Sure you can, using the PEAK button if the meter has it.

(although the 17B doesn't)

[easilyconfused]

"You can't measure the peak-to-peak voltage of anything with a multimeter, it doesn't work like that. Meters are designed to measure either the average or true RMS value of a an AC waveform. Dave."

Okay. This is going to be one of those moments where this learner must take something on faith. I understand what you said about the hash marks, and I'm VERY pleased that my scope is apparently in cal in the X and the Y. But I hope it's not hard for the forum to understand how I would expect the lug to show a voltage on my mm to match the stated lug voltage (from the manual). And for the sake of expanding my knowledge I still want to learn why it doesn't. Thanks to all.

[DJPhil]

It does have a Hz% button. But when I press it (while in the DCV mode) it simply shows 0.000.

You'll have to be in the ACV mode to read frequency. Switch to ACV, change range to get a good reading, and hit the Hz% once, and you should be good. Hitting the Hz% again will show you duty cycle, which should be very close to 50%.

Hope that helps.

I did as you suggested. In the AC mode I got a voltage very similar to the DC mode (about 1 volt higher). I pressed the Hz% button and got zeros. I pressed it again and got zero percent. 

Hmm, I'm a bit baffled too. I assume the probes are where they're supposed to be if you're getting an AC voltage reading, but just to be sure, you have one lead on scope ground and one lead on the cal bar? I'm also assuming the range switch doesn't need adjustment, I'm not sure.

It's possible that the 17 isn't sensitive enough to measure that weak a signal. All I can find online is the calibration guide, so I'm not sure what the actual specs are. The performance test on page 11 says to use a 1V signal to check frequency to 5KHz. I don't think it's very likely that this is the problem, but I can't be sure.

Sorry if I added to the confusion. 

[MTron]

I know I really don't belong on a forum mainly populated with advanced practitioners. I will take a look at the XYZ PDF. I've been there before but it might mean more now since I've developed a little more understanding of scopes and how to interpret them.

This is exactly the forum you should be on, and there are plenty of beginers here, like me for example. I just purchased my first oscilloscope as well, and your not alone in your confusion...there is quite a learning curve

as for your division confusion. 50mV per div means every large square is 50mV. Between each square are hash marks, if there are 5 between each large square then each hash is 1/5 x 50mV or 10mV.  Also keep in mind your probe x10 probes will output 1/10th the actual voltage. Your scope might have markings on the volts/div knob that display what the reading will be with a X1 or X10 probe.

Remember...ask questions...make mistakes....its the best way to learn.

[djsb]

The measurement should be done with the meter set to AC voltage otherwise any DC offset will also be measured. Some meters take this into account, others maybe not. This is why an oscilloscope is so useful as it allows you to see any DC offset present.
As an example on my Tek 2225 scopes 0.936Hz 0.52v Pk-Pk cal output my cheapo maplins precision gold multimeter reads 0.257V on ACV and my Gossen Metrohit x-tra reads 0.2479V true RMS. Both pretty close, but I trust the Gossen more. You should get around 150mv on your multimeter. RMS values also depends on the waveform. For a squarewave the RMS value is the same as the amplitude (peak value). For a sine wave the RMS is amplitude divided by the square root of 2 (or multiplied by 0.707).
All explained here.

http://en.wikipedia.org/wiki/Root_mean_square

David.

[alm]

Tektronix has an appnote XYZs of Oscilloscopes that is helpful for beginners, not just for Tek scopes (most scopes work very similar).

[easilyconfused]

Also each subdivision equals 10mv (50/5). So if the trace only goes part way across a full division you count the number of subdivisions and multiply by 10mv (at the 50mv per division setting). This is then added to the full division reading as above.
I've probably confused you even more now.

David

No you haven't. I understand that. Thanks

[easilyconfused]

I know I really don't belong on a forum mainly populated with advanced practitioners. I will take a look at the XYZ PDF. I've been there before but it might mean more now since I've developed a little more understanding of scopes and how to interpret them.

This is exactly the forum you should be on, and there are plenty of beginers here, like me for example. I just purchased my first oscilloscope as well, and your not alone in your confusion...there is quite a learning curve

as for your division confusion. 50mV per div means every large square is 50mV. Between each square are hash marks, if there are 5 between each large square then each hash is 1/5 x 50mV or 10mV.  Also keep in mind your probe x10 probes will output 1/10th the actual voltage. Your scope might have markings on the volts/div knob that display what the reading will be with a X1 or X10 probe.

Remember...ask questions...make mistakes....its the best way to learn.

Thanks bro

[easilyconfused]

"Hmm, I'm a bit baffled too. I assume the probes are where they're supposed to be if you're getting an AC voltage reading, but just to be sure, you have one lead on scope ground and one lead on the cal bar?"

I put the red mm lead on the scope's output (cal) lug. I put the black mm lead on the scope's ground lug. The voltage displayed in the auto-ranging DC voltage mode was 1.48. I switched to AC voltage mode and got a value about 1.52v or so. Nobody ever told me that the calibration voltage would coincide with the multimeter. I just assumed it would. And I still don't understand why it doesn't. Maybe it's got something to do with the pulsations so that I'm only seeing half of it.

[alm]

A good DMM will read the RMS value, which is the root of the mean of the squares, this is derived from the equation for  power (V²/R for voltage over a resistor), Wikipedia (link above) has the proper derivation. But for a square wave, the RMS and average value are actually identical, so for this purpose, you might as well use the average voltage. For a square wave from 0 to Vpp with a 50% duty cycle, the average voltage is Vpp/2. So ~1.5V RMS/average corresponds with 3Vpp. I assume that you use a 10x scope probe, which attenuates the signal ten times (voltage divider, 9Mohm / 1Mohm). So the 3Vpp becomes 300mVpp.

An average responding meter will multiply the average with 1.1 to correct for the difference between RMS and average for sinusoidal signals, but many average responding meters won't have the bandwidth for 1kHz square wave anyway (something like 10kHz might be OK, many cheap ones are just 50-500Hz), so I wouldn't trust their answers, although the results from your Fluke 17B (which is not true RMS if I remember correctly) seem plausible in this case, maybe the frequency roll-off and +10% correction cancel each other out.

[easilyconfused]

A good DMM will read the RMS value, which is the root of the mean of the squares, this is derived from the equation for  power (V²/R for voltage over a resistor), Wikipedia (link above) has the proper derivation. But for a square wave, the RMS and average value are actually identical, so for this purpose, you might as well use the average voltage. For a square wave from 0 to Vpp with a 50% duty cycle, the average voltage is Vpp/2. So ~1.5V RMS/average corresponds with 3Vpp. I assume that you use a 10x scope probe, which attenuates the signal ten times (voltage divider, 9Mohm / 1Mohm). So the 3Vpp becomes 300mVpp.

An average responding meter will multiply the average with 1.1 to correct for the difference between RMS and average for sinusoidal signals, but many average responding meters won't have the bandwidth for 1kHz square wave anyway (something like 10kHz might be OK, many cheap ones are just 50-500Hz), so I wouldn't trust their answers, although the results from your Fluke 17B (which is not true RMS if I remember correctly) seem plausible in this case, maybe the frequency roll-off and +10% correction cancel each other out.

Even thought 99% of what you've said is over my head I think I get the gist of it. And I think the gist of it is that the 0.148 volts I'm reading on my Fluke non-RMS is one half of the pulsating signal. If that's true- there is some diagnostic value in doing it as one of several steps in verifying the cal on your newly acquired garage sale scope. PS: I'm not sure but I suspect I've given that value of 0.148 with the decimal in the wrong place in parts of this string. 0.148 is correct and the lug value is said to be 0.3v.

[EEVblog]

"Hmm, I'm a bit baffled too. I assume the probes are where they're supposed to be if you're getting an AC voltage reading, but just to be sure, you have one lead on scope ground and one lead on the cal bar?"

I put the red mm lead on the scope's output (cal) lug. I put the black mm lead on the scope's ground lug. The voltage displayed in the auto-ranging DC voltage mode was 1.48. I switched to AC voltage mode and got a value about 1.52v or so. Nobody ever told me that the calibration voltage would coincide with the multimeter. I just assumed it would. And I still don't understand why it doesn't. Maybe it's got something to do with the pulsations so that I'm only seeing half of it.

As I said, multimeters can ONLY read RMS or Average values for AC waveforms which his what you are looking at. That is how they are designed.
If you want to measure the peak-to-peak value of a signal, like you just did with the calibration signal, you have to use an oscilloscope so you can actually see the waveform.
i suggest you go and study the difference between RMS, average, and peak-to-peak waveforms.

So trying to use the multimeter to measure the calibration signal was wrong in this case, it can't do it.

Dave.

[alm]

Even thought 99% of what you've said is over my head I think I get the gist of it. And I think the gist of it is that the 0.148 volts I'm reading on my Fluke non-RMS is one half of the pulsating signal. If that's true- there is some diagnostic value in doing it as one of several steps in verifying the cal on your newly acquired garage sale scope. PS: I'm not sure but I suspect I've given that value of 0.148 with the decimal in the wrong place in parts of this string. 0.148 is correct and the lug value is said to be 0.3v.

My guess about the 10x probe was wrong, you're either using a 1x probe or a plain piece of wire. The manual says 0.3V, your picture shows 0.3V, and your meter shows half that. Everything looks fine to me.

Just forget all the RMS stuff for a second. Go draw a square wave from 0V to 0.3V (i.e. it's 0V for the same amount of time as it's 0.3V) on a piece of graph paper. Use one box for 50mV (so the square wave is six boxes high), and make each horizontal line two boxes wide. Now imagine what the average value would be (hint: a straight line exactly in the center between 0V and 0.3V). That's the 0.15V you would get if you would measure the average value, or would use a true-RMS meter. A non-true-RMS meter measures this value, and multiplies it by 1.1 so it will show the average value for sines. The actual math for RMS is more complex, but would give the same answer for a square wave.

Your meter has a specified frequency response of 50-200Hz, so measuring this kind of signal is way out of spec. So the fact that it's only 10% low is actually pretty good. You should only measure AC signals that are 50/60Hz and not heavily distorted with this type of meter, other signals may give inaccurate results.

[scrat]

Even thought 99% of what you've said is over my head I think I get the gist of it. And I think the gist of it is that the 0.148 volts I'm reading on my Fluke non-RMS is one half of the pulsating signal. If that's true- there is some diagnostic value in doing it as one of several steps in verifying the cal on your newly acquired garage sale scope. PS: I'm not sure but I suspect I've given that value of 0.148 with the decimal in the wrong place in parts of this string. 0.148 is correct and the lug value is said to be 0.3v.

My guess about the 10x probe was wrong, you're either using a 1x probe or a plain piece of wire. The manual says 0.3V, your picture shows 0.3V, and your meter shows half that. Everything looks fine to me.

Just forget all the RMS stuff for a second. Go draw a square wave from 0V to 0.3V (i.e. it's 0V for the same amount of time as it's 0.3V) on a piece of graph paper. Use one box for 50mV (so the square wave is six boxes high), and make each horizontal line two boxes wide. Now imagine what the average value would be (hint: a straight line exactly in the center between 0V and 0.3V). That's the 0.15V you would get if you would measure the average value, or would use a true-RMS meter. A non-true-RMS meter measures this value, and multiplies it by 1.1 so it will show the average value for sines. The actual math for RMS is more complex, but would give the same answer for a square wave.

Your meter has a specified frequency response of 50-200Hz, so measuring this kind of signal is way out of spec. So the fact that it's only 10% low is actually pretty good. You should only measure AC signals that are 50/60Hz and not heavily distorted with this type of meter, other signals may give inaccurate results.

I agree. I can't see the problem: on the scope, with your 10x probe, you measured 0.3 Vpp, which is as specified (and it means 3 Vpp at the cal output). When you measured it with the DMM, you got 1.48. If your DMM is a true-rms, you should measure 1.5 V (while you got 1.48 V), since this is the rms value for a 3 Vpp square wave. Isn't it OK for an old scope? How could it be anything different?

The first definition they gave me for rms value was "the value of a DC that will have on a resistance the same thermal effect as the considered signal". Even if I could complicate it by saying that "the rms value is the square root of the mean value of the square of the signal", and that you should calculate it with an integral, the old informal definition still works