Oscilloscope strange measurement values

[Edgarus]

Hi,
I have a Siglent SDS 1052DL oscilloscope. It looks good, works.. almost good.
The problem is strage values of voltage measurmets depend on a set scale:
10V scale:

5V scale:

2V scale:

1V scale:

500mV scale:

and 200mV scale (which is very ...ech. strange):


A circuit: simple voltage divider, power supply stable 5V. Out voltage of the divider 1.552V measured by meter.
At 500mV scale the measurement is similar to indications of the meter, but other values are rounded.
I've made a test with 12V on second channel, and with a large scale  (I do not remember the value), oscilloscope shown me 18Volts!.

My question is, if it is normal for that kind of equipment, or maybe this Sglent is a crap?

Best regards
Edgarus.

[alsetalokin4017]

Well.... part of the  problem may be that you are showing a 5 v DC signal on CH2, but your Trigger Level appears to be set to 12 volts on CH2.

Another part of the problem (the last scopeshot) may be that usually the scope is calculating the measurements from what is actually displayed on the screen, and when your signal is off the screen it can't calculate properly.

Can you repeat the measurements with the scope trigger set to some value within the range of the actual signal you are trying to trigger on?

ETA: Never mind, if you are using AUTO mode then the scope should be running along untriggered. But it seems to be showing mostly correct values, considering scale resolution errors, except for the last shot.

[Edgarus]

I've made some screens as you asked, one channel active, 5V trigger to 0V. This time I've started from 100V/div.











This is ridiculous, that it shows 12V on 5V rail. And this 244mV on last picture. Factory reset, autoset or other kung fu is not correcting this issue, of course latest firmware is installed.

So I have to be sure that it's ok and it's normal, or I can return it to the shop.
I wonder how it looks on tektronix, rigol or agilent devices.

Best regards.
Edgarus

[bktemp]

This is normal:
At 100V/div range, the ADC has a input range of at least 800Vpp. Divided by 256 steps of the 8bit ADC gives at least 3V per step. High speed ADCs often have errors of a few LSBs, so 7V offset at 100V/div is not unusual.
At <1V/div 5V is simply out of range. The readout shows the maximum voltage the ADC can measure. It would be nice if it would show something like "OVER" but the scope measures the best it can with the range the user has set.

[Edgarus]

Thank you for your replies.
Anyway, I will try to return it to the shop, or change it to Rigol 1054Z.

Best regards
Edgarus.

[tautech]

We have no way of knowing if your connection method is not influencing the measurements. 
Be aware also the scope will "grab" any glitches/ripple from the environment and add it to calculations.
Input filters and BW limit can be used to manage/reduce these.

Please also study the specifications to check if measurements are WITHIN stated accuracy.
A scope is NOT a 5 1/2 digit DMM.

[dom0]

8 Vpp noise at 100 V/div is pretty good, I don't see your problem there.

[AG6QR]

Thank you for your replies.
Anyway, I will try to return it to the shop, or change it to Rigol 1054Z.

Best regards
Edgarus.

It looks like it's perfectly normal, as far as I can tell.  My Rigol (an older 1052E) would behave pretty much the same way.  If you try and return it or change it for another scope, I think you'll be disappointed.

Those measurement functions are based on what's displayed on the screen.  If you don't have the trace properly displayed on the screen, where it's fully visible, with good resolution, then you won't get reasonable results.  Even if you DO have the trace fully visible, the Vpp measurements will often show at least one or two pixels worth of noise, so they won't be hugely valuable for a DC signal (if you want accurate Vpp in that case, use AC coupling and a scale sufficient to show the waveform with more than a few pixels worth of height).

First, get the trace displayed so that you can make the required measurements reasonably well by using your eye and a ruler against the screen.  After you do that, the "measure" function will do a more accurate and easier job than your eye and the ruler would do.

[w2aew]

As the others have said, what your seeing is normal.  This is not an autoranging voltmeter.  You have to be sure to set the scope up appropriately to get an accurate measurement.  If your vertical scale is too large with respect to your signal, then the signal deflection will be very small and will suffer from the coarse 8 bit resolution.  If the scale is such that your signal is off of the screen, then your 8 bit digitizer is out of range and you can't get a reading. This is the same for basically any scope.