# DC offset VS Superimposed DC voltage on AC waveform

Discussion in 'General Electronics Discussion' started by Danny Daviss, Mar 21, 2013.

Not open for further replies.
1. ### Danny Daviss

67
0
Mar 16, 2013
Do you measure DC offset from the Negative Peak cycle of a sine waveform or do you measure the DC offset from the zero crossing points of the sine wave cycle?

If you have a 6VDC offset, is the measurement from the Oscilloscope's zero Origin zero volts to the negative peak of the AC sinewave or to the zero crossing point of the AC sinewave?

When measuring AC ripple on a power supply is the measurement from the Oscilloscope's zero Origin zero volts to the negative peak of the AC sinewave or to the zero crossing point of the AC sinewave?

DC superimposed on a AC waveform is different than DC offset voltage?

DC superimposed voltage means you measure the DC voltage from the zero origin to the negative peak of the cycle?

2. ### john monks

693
3
Mar 9, 2012
Are these homework problems?

3. ### Danny Daviss

67
0
Mar 16, 2013
no at work , testing boards

4. ### Danny Daviss

67
0
Mar 16, 2013
I know you measure DC offset from zero voltage zero origin on the oscilloscope to the zero crossing point of a sinewave

But when measuring AC ripple , do you measure the DC offset from the AC negative peak cycle or from the zero crossing point?

5. ### john monks

693
3
Mar 9, 2012
The second part of your question is neither.

Normally you measure AC ripple from the peak to peak component of your signal.
In other words you look at the signal with an oscilloscope in the AC mode and look at the difference between the most positive part of your waveform and the most negative part of the waveform. Then you will have the AC ripple in peak to peak form.

Sometimes you want this in RMS and this can be much more complicated when the waveform is not a sine wave.

You get to DC offset is whatever the signal is shifted from the usual starting point.
In television for example the signal height starts from black to white. This is adjusted with a combination of an AC gain control and a DC offset or automatic DC corrector. So DC offset is normally whatever the signal is shifted positive or negative to from whatever is considered normal.

6. ### Danny Daviss

67
0
Mar 16, 2013
Yes I know that, But what i don't know is, how do u measure the DC offset of the ripple?
what is the DC voltage level that the AC ripple is riding on top of

Do you measure from the Negative Peak of the AC ripple or the zero crossing point of the AC ripple?

7. ### Danny Daviss

67
0
Mar 16, 2013
What confuses me about DC offset is when measuring square waveform's.

How do you know to measure the square waveform from zero volts origin, which the square waveform has only a Positive cycle , not a negative cycle. It's just low ( zero Volts ) to + 5 volts TTL signals

Or If the square waveform has a positive and negative cycles

How do you know or can tell what kind of square waveform you have?

8. ### john monks

693
3
Mar 9, 2012
Normally you measure the DC offset with a DC voltmeter.
The second part of your question is much harder because you are asking about a definition and I don't know how to define it. The DC component of a square wave is simply halfway between the top and the bottom of the waveform. Now when we are thinking of an offset you have to consider what the normal DC level is. Coming out of a TTL I would think that the top is about 4.5 volts and the bottom is a little less than 0.1 volts. So if I saw the bottom at 1 volt I would think that there is about 1 volt positive DC offset.

9. ### davennModerator

13,947
1,988
Sep 5, 2009
you view and measure it on an oscilloscope
you can then measure its on Vs off ratio and its amplitude

Dave

10. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,501
2,841
Jan 21, 2010
If there's ripple, you measure and DC offset (typically) from the point mid-way between the max and min of the ripple (assuming it is symmetrical).

You do that because you typically define ripple as a pure AC waveform imposed on your desired signal. If that signal is DC, then the ripple will both add and subtract from a constant DC level (your offset) that would exist if the ripple did not.

For a square wave, the DC offset depends on your definition of what the signal is.

How do you define the signal? The variation (if constant) is a DC offset.

11. ### Danny Daviss

67
0
Mar 16, 2013
Yes I measure the DC offsets with my DC volt meter because the oscilloscope confuses me about DC offsets on where they start

Once I get the DC offset voltage, i use the oscilloscope to measure the square waveform

The Problem I'm confused about is , is the square waveform starting at the baseline of the square waveform or is it starting at the halfway between the top and bottom?

So I'm not sure is the square waveform is riding on top of the DC offset voltage from the baseline or from the halfway point

TTL square waveforms start from the baseline right? CMOS square waveforms start at the half way point of a square waveform right?

12. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,501
2,841
Jan 21, 2010
That may not work depending on the signal and how you define it (especially true if it is a pulse train and you define zero volts as being one of the two logic levels)

You need to learn to use the oscilloscope.

Why? I can't see how you need one to measure the other.

How is your square wave defined?

That depends. Where do you define it to be starting?

And neither do we unless we know how the signal is defined.

This is the most common definition.

Why would you expect them to be defined differently from a TTL signal?

Is this a normal logic signal, used for driving logic, in an environment with a single supply rail?

If so, set your scope to DC, and the logic signal should appear with the logic 0 levels at or very close to the baseline (may be slightly above, but should not be below). You are aware that there is a valid range for logic signals, and if your logic 0 is at 0.25V this DOES NOT mean there's a DC offset unless (for some reason) your Logic 1 rises above Vcc/Vdd.

13. ### davennModerator

13,947
1,988
Sep 5, 2009
You do realise that you dont get smooth DC from a AC fed power supply ?

The absolute DC output value of the supply is an avg of the ripple value ... that is, it will be roughly halfway between the positive and negative cycles of the ripple

To measure the amount of AC ripple you are measuring its P-P value

have a look here in wiki

also have a look at this good tutorial

Dave

#### Attached Files:

• ###### Smoothed_ripple.GIF
File size:
22.1 KB
Views:
742
Last edited: Mar 22, 2013
14. ### Danny Daviss

67
0
Mar 16, 2013
How does a tech know how a square wave form is starting? off the baseline Logic zero or Halfway?

How would a tech confirm this?

How do you defined the signal? what do you mean by that sorry? because I don't know either how to defined a square waveform

I'm guessing is the square waveform is either Logic zero base line or starts at the half way point

How does a tech know where the starting point is?

15. ### Danny Daviss

67
0
Mar 16, 2013
When I put my DVM meter on the output of a 12VDC power supply, it measures 12VDC, when an oscilloscope you measure the DC level from zero volts to the half way of the AC ripple

16. ### davennModerator

13,947
1,988
Sep 5, 2009
put your multimeter across the DC output of the PSU in AC mode and you will measure the AC ripple voltage

Dave

17. ### Danny Daviss

67
0
Mar 16, 2013
Yes I got that part , the peak to peak AC ripple voltage

The AC ripple voltage is not riding on top of 12 VDC , the AC ripple voltage is below 12 VDC , the positive peak of the AC ripple voltage is 12 VDC and the negative peak of the AC ripple voltage is 11 VDC

18. ### john monks

693
3
Mar 9, 2012
I typically calibrate my oscilloscope with a voltmeter because oscilloscopes are notoriously inaccurate. I suspect that what you see on your scope is a little off unless the ripple is exceptionally non-symmetrical. And this could be. I wonder if your ripple looks like davenn's waveform.

19. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,501
2,841
Jan 21, 2010
If it's the rising edge of a square wave, how different do you think they'll be?

Technically the rise time of a square wave is often taken to be the time taken to go from 10% to 90% of the final voltage.

Well, perhaps by looking for the falling edge and confirming it's also a sharp transition.

Your questions are so ill-formed that it's really hard trying to figure out what you're after.

Do you understand how to use Google?

They know because they listen, and understand what was told to them.

so it's about 1V peak to peak.

I would tend to say the power supply is producing 11.5V with 1V of ripple, but it depends STRONGLY on the shape of that ripple.

Here's a hint. Draw a picture of the waveform you get on the scope and post it here or I will close this thread because you're not trying to help yourself.

20. ### Danny Daviss

67
0
Mar 16, 2013
The power supply is +12 and -12

The square waveform's range from 5 volts peak , 10 volts peak , 12 volts peak

Some of the square waveform's have DC offset

I measure the DC offset with my DVM meter, it will have 1.5 volts or 2 volts DC offset voltage

Is the square waveform starting at the base line at 1.5 volts DC offset or is the square waveform starting at half way 50% amplitude at 1.5 volts DC offset?

Zero can be either
1.) Starting at the baseline, logic zero volts , which is called "Monopolar" because its just low and high states
2.) Starting at 50% amplitude of a square waveform , which is called "bipolar" because it has a positive and negative cycles

When do you or a tech knows when a square waveform "starts" at 50% amplitude and not at zero volts 0% baseline?

How do you know when the square waveform has positive and negative cycles that its alternating?