I'm aware I have to accept minor deviations when changing ranges, but I think the arrow should go with the trace.
(on my old CR scopes there are trimpots for each channel to compensate for component aging and such things)
The arrow shows the nominal position and any deviation from that is just the offset error. Of course we would like it to be zero, but all equipment have offset errors, particularly DSOs with such a high sensitivity of 500µV/div like the SDS1202X-E.
For ranges 500µV and 1mV per division, the datasheet specifies the offset error as follows:
±(1% x Offset + 1.5% x 8 x div + 500µV)
The vertical position control actually adds an internal offset to the channel in order to change the trace position. This way, a DC offset of the signal can be cancelled out for instance.
If the trace is centered on the screen (position 0, equivalent to internal offset 0), only the second term applies: 1.5% x 8 x div + 500µV, where div is the channel gain setting. So for 500µV/div we get:
1.5% x 8 x 500µV + 500µV = 1.5% x 4mV + 500µV = 60µV + 500µV = 560µV.
This means, up to 560µV of deviation would still be in spec with the trace at position 0, i.e. the center position.
If the position control is used to apply an internal offset, thus changing the trace position, the first term of the specification comes into effect as well: 1% x Offset. So if we position the trace at the bottom of the screen, the offset is 4 divisions or 4 x 500µV = 2mV and an additional error of 20µV (1% of 2mV) has to be added to the permissible offset error, resulting in a total of ±580µV.
Yes, analog scopes have internal trimmers – and they used to be expensive, partly because someone at the factory had to adjust all these trimmers. These adjustments were not completely stable with time and temperature either. And then, most analog scopes had a maximum sensitivity of 5mV/div and could only go down to 1mV/div with an additional 5x multiplier, which introduced additional errors – just look at the specifications of your analog scope. You just don’t see it so clearly during practical use, because there is usually no indicator (arrow) where the trace should be on an analog scope.
I forgot to mention:
before posting I checked the FW version: 5.1.3.13 and did
Self Cal as well - no joy.
Why was self-cal no joy?
What bothers me most is the effect of CH2 when activated on the CH1-offset (when at center zero div line).
See the screen shots.
Yes, with these symptoms, your scope has to be faulty. Just send the unit back for replacement.
-----------------------------------------------------------------------------------------------------
EDIT: I thought I’d add some more information on the offset error specification.
Some might ask why that first term in the formula
±(1% x Offset + 1.5% x 8 x div + 500µV)
even exists, when it just added negligible 20µV to the 560µV of the 2nd term in the example calculation.
The answer is that the offset range is much wider than just the ±4 divisions of the screen. It is actually up to ±2V (for the gain ranges discussed here, i.e. 500uV and 1mv/div) and if this is used to cancel out a DC offset of the signal, the error margin becomes significant.
Consider an offset of 500mV, then the additional error could be as much as 5mV (1% of 500mV) and the trace might not even be within the visible screen area anymore.
Thankfully, the actual accuracy is much higher than that. See the screenshot below, where an accurate 500mV DC voltage is fed into channel 1 and a -500mV has been set on that channel.
SDS1202X-E_Offset_500mV_500uV
20MHz bandwidth limit and 16x average acquisition mode has been used in order to suppress the noise from the scope and the DC calibrator.
As can be seen, the trace is right at the center of the screen and all the automatic measurements are pretty much spot-on. Since this is pure DC, Max, Top, Mean and RMS measurements should all give identical results, and they certainly do. The most “inaccurate” results are from the mean and RMS measurements, and they are off by 60µV, which is only 0.012% of the total input (and also nominal offset) voltage!