How good is the Rigol DG1022Z at calculating voltage from load impedance?

[Ben321]

When I set the load impedance calculation to 50 Ohms in the DG, and the voltage to 1Vpp (and 0V offset), that means that when connected to a 50ohm load, the output waveform should be the specified voltage (1Vpp). That is, the waveform should go between 0.5V and -0.5V as seen on an oscilloscope measuring the signal.  And that's exactly how I have my setup. The BNC cable from the DG's output goes to one input of a T-adapter on my Picoscope. On the other input of the T-adapter is a 50ohm resistor (the kind built-in to a special BNC connector, who's sole purpose is to provide a 50ohm terminator or test load).

With this setup, the waveform's peak and trough points should be 0.5V and -0.5V. However, I'm finding they are instead approximately 0.6V and -0.6V. Was my Rigol DG somehow poorly calibrated at its factory? Is this another example of a low quality "made in China" product? Or is it that the BNC 50ohm resistor I bought was low quality? I think it was the cheapest BNC 50ohm resistor I could find on either Mouser or Digikey. Or is this possibly an issue with the Picoscope's calibration?

[xrunner]

So to answer the question you asked "How good is the Rigol DG1022Z at calculating voltage from load impedance?"

It doesn't really "calculate" anything. It simply doubles the voltage on the screen if you specify that the load is "high impedance". If the load is set to "50 ohms" it divides the displayed voltage in half. But it isn't adjusting anything on the output. It has no way of knowing or sensing anything and you could hook up any pure resistance for the load (other than 50 ohms) and it would not change the display.

So if you are not seeing what you expect it could be more than one cause - but not the setting of "high impedance" or "50 ohms".

[TimFox]

At low frequencies, the value of the 50 ohm resistor you measure on a 3.5 or more digit DMM should tell you the actual load resistance of your fixture.
At what frequency did you measure +/- 0.6 V?  Have you checked the input of the Picoscope against a DC source (e.g., 1.5 V battery that you measured with your DMM)?
Both are worth investigating:  I have found many cheap BNC terminators to be way out of spec on DC.

[Ben321]

It doesn't really "calculate" anything.

It does a calculation. It knows that the device's output resistance is 50ohms, so when you set the load impedance mode to 50ohms, it displays a voltage that's half the actual open-circuit voltage. When I then set the displayed voltage to 1Vpp, I expect that under the intended operating conditions (50ohm load) the actual measured voltage will also be 1Vpp. If the measured voltage ends up being approximately 1.2Vpp (as it was in my case), I know something is wrong.

[xrunner]

It does a calculation. It knows that the device's output resistance is 50ohms, so when you set the load impedance mode to 50ohms, it displays a voltage that's half the actual open-circuit voltage. When I then set the displayed voltage to 1Vpp, I expect that under the intended operating conditions (50ohm load) the actual measured voltage will also be 1Vpp. If the measured voltage ends up being approximately 1.2Vpp (as it was in my case), I know something is wrong.

This was your stated question -

"How good is the Rigol DG1022Z at calculating voltage from load impedance?"

It's bad at that because it doesn't do it.

By calculation I mean it doesn't take into account any actual physical resistance connected to it and making an actual calculation of what would be the voltage drop across the load. If you connect a 4,700 ohm resistor as the load does it calculate anything other than the divide by two or not? nope.  It doesn't know beans about what you have connected to it in the sense you are thinking.

If you want to get pedantic, yea yea OK it divides the setting by two if YOU tell it to - technically that's a calculation but it's a "dumb" calculation because that's all it is capable of doing - displaying the set voltage or dividing it by two. And it's dumb because it depends on the user to set it correctly or it will tell you the wrong thing.

So to answer your question again - it does not calculate voltage from load impedance

Note: By the way I have a DG1032Z

[Ben321]

It does a calculation. It knows that the device's output resistance is 50ohms, so when you set the load impedance mode to 50ohms, it displays a voltage that's half the actual open-circuit voltage. When I then set the displayed voltage to 1Vpp, I expect that under the intended operating conditions (50ohm load) the actual measured voltage will also be 1Vpp. If the measured voltage ends up being approximately 1.2Vpp (as it was in my case), I know something is wrong.

This was your stated question -

"How good is the Rigol DG1022Z at calculating voltage from load impedance?"

It's bad at that because it doesn't do it.

By calculation I mean it doesn't take into account any actual physical resistance connected to it and making an actual calculation of what would be the voltage drop across the load. If you connect a 4,700 ohm resistor as the load does it calculate anything other than the divide by two or not? nope.  It doesn't know beans about what you have connected to it in the sense you are thinking.

If you want to get pedantic, yea yea OK it divides the setting by two if YOU tell it to - technically that's a calculation but it's a "dumb" calculation because that's all it is capable of doing - displaying the set voltage or dividing it by two. And it's dumb because it depends on the user to set it correctly or it will tell you the wrong thing.

So to answer your question again - it does not calculate voltage from load impedance

Note: By the way I have a DG1032Z

Of course it can't know the load impedance actually connected, but it knows what you tell it that it is. So when I tell it the load is 50ohms, and to output a 1Vpp signal, what it does is calculate that the open-circuit voltage must be 2Vpp, and then it outputs a 2Vpp signal (which will be pulled down to 1Vpp if it's attached to a 50ohm load). So what I was saying is how good is this calculation? Is its firmware actually doing the proper calculation? That's what I was calling into question, was if the firmware's math functions were working correctly. Is it taking the 1Vpp I typed in on the controls, and calculating a 2.4Vpp voltage from that (such that it will be 1.2Vpp when attached to a 50ohm load), instead of the 2Vpp it should be calculating, due to a buggy math function in the firmware? Or maybe the DG's supposed output of resistance of 50ohms is actually smaller than 50ohms, due to poor manufacturing? Or is it a problem with the 50ohm BNC terminator I have?

That's what I'm trying to figure out. Where is the breakdown in this relatively simple equipment setup I've got?

[xrunner]

Of course it can't know the load impedance actually connected, but it knows what you tell it that it is. So when I tell it the load is 50ohms, and to output a 1Vpp signal, what it does is calculate that the open-circuit voltage must be 2Vpp, and then it outputs a 2Vpp signal (which will be pulled down to 1Vpp if it's attached to a 50ohm load). So what I was saying is how good is this calculation? Is its firmware actually doing the proper calculation? That's what I was calling into question, was if the firmware's math functions were working correctly.

No.

You have an idea in your mind that is simply wrong, and I am trying to help you get around it.

Look, here's the text from the DG1022Z manual, you should read it and understand what you are saying is NOT what it does. Specifically look at the bolded text (my bolding). Most important is the last line that is bolded.

If you cannot understand this then I will let some other member take a crack at explaining it.

DG1022Z manual p. 2-88

2. Output Impedance
The output impedance setting influences output amplitude and DC offset. The
instrument has a 50Ω fixed serial output impedance for the [CH1] connector at
the  front  panel.  If  the  actual  load  is  different  from  the  specified  value,  the 
voltage level displayed would not match the voltage level of the device under
test. To ensure correct voltage level, the load impedance setting must match the
actual load.
 
Press Utility  Channel Set  Output Set  Imped to select “HighZ” or
“Load”. The default is “HighZ”. If “Load” is selected, use the numeric keyboard to
set specific impedance value. The default is 50Ω and the available range is from
1Ω to 10kΩ. Impedance setting will be displayed on the screen.
 
The generator will adjust the output amplitude and offset voltage automatically
once the impedance setting is changed. For example, the current amplitude is
5Vpp. At this point, change the output impedance from 50Ω to HighZ and the
amplitude displayed on the screen will double to 10Vpp. If the output impedance
is changed from HighZ to 50Ω, the amplitude will reduce to half of the previous
value (2.5Vpp). Notice that only the displayed values change with the parameter
and the real output from the generator does not change.

[Ben321]

Of course it can't know the load impedance actually connected, but it knows what you tell it that it is. So when I tell it the load is 50ohms, and to output a 1Vpp signal, what it does is calculate that the open-circuit voltage must be 2Vpp, and then it outputs a 2Vpp signal (which will be pulled down to 1Vpp if it's attached to a 50ohm load). So what I was saying is how good is this calculation? Is its firmware actually doing the proper calculation? That's what I was calling into question, was if the firmware's math functions were working correctly.

No.

You have an idea in your mind that is simply wrong, and I am trying to help you get around it.

Look, here's the text from the DG1022Z manual, you should read it and understand what you are saying is NOT what it does. Specifically look at the bolded text (my bolding). Most important is the last line that is bolded.

If you cannot understand this then I will let some other member take a crack at explaining it.

DG1022Z manual p. 2-88

2. Output Impedance
The output impedance setting influences output amplitude and DC offset. The
instrument has a 50Ω fixed serial output impedance for the [CH1] connector at
the  front  panel.  If  the  actual  load  is  different  from  the  specified  value,  the 
voltage level displayed would not match the voltage level of the device under
test. To ensure correct voltage level, the load impedance setting must match the
actual load.
 
Press Utility  Channel Set  Output Set  Imped to select “HighZ” or
“Load”. The default is “HighZ”. If “Load” is selected, use the numeric keyboard to
set specific impedance value. The default is 50Ω and the available range is from
1Ω to 10kΩ. Impedance setting will be displayed on the screen.
 
The generator will adjust the output amplitude and offset voltage automatically
once the impedance setting is changed. For example, the current amplitude is
5Vpp. At this point, change the output impedance from 50Ω to HighZ and the
amplitude displayed on the screen will double to 10Vpp. If the output impedance
is changed from HighZ to 50Ω, the amplitude will reduce to half of the previous
value (2.5Vpp). Notice that only the displayed values change with the parameter
and the real output from the generator does not change.

What you are saying is what I was saying too but maybe in different words. I think we are talking past each other. Well that's beside the point. I'm going to test the terminator resistor now to see if that's the problem.

[xrunner]

What you are saying is what I was saying too but maybe in different words. I think we are talking past each other. Well that's beside the point. I'm going to test the terminator resistor now to see if that's the problem.

Could be. Set the DG1022 to 1 Vpp sine wave. Connect it to an oscilloscope (a high impedance load). You will measure 1 Vpp. If you go into the DG1022 menu and go between telling it you have a 50 ohm load connected, or a high impedance load connected, guess what will happen on the scope screen?

Nothing.

Nothing on the output will change. That menu option is for the operator's convenience in remembering what the load should be experiencing. But it only works for 50 ohm or high impedance and it only changes what the screen is displaying - THE OUTPUT AMPLITUDE WILL NOT CHANGE BECAUSE THAT MENU OPTION IS CHANGED BY YOUR HAND.

Good Luck.